From owner-sam-users@nvg.unit.no Thu Nov 7 19:27:19 1996 Date: Wed, 6 Nov 1996 11:50:04 +0100 Message-Id: <96110611495899@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: sam-users@nvg.unit.no Subject: Re: Re : Killer apps. X-Vms-To: SMTP%"sam-users@nvg.unit.no" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 233 Lines: 7 That's one thing we'll have to contend with. The number of people who I have heard say "ooh - look. Intel Inside. My uncle's [whoever] says that and that cost loads, so it *must* be good" just before I've mangles their head... Al. From owner-sam-users@nvg.unit.no Thu Nov 7 19:27:20 1996 Date: Thu, 7 Nov 1996 19:24:54 +0000 (GMT) From: "A.S. Collier" X-Sender: asc25@navy.csi.cam.ac.uk To: sam-users@nvg.unit.no Subject: Re: SamSon Interrupts In-Reply-To: <691803D435B@eigg.sms.ed.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1498 Lines: 32 On Thu, 7 Nov 1996, Dave Hooper wrote: > Here's a question ... why would we need line-interrupts to control > the speed of a sample player or tracker... if we're going to have a > whizzy sound/graphics card (or was this idea scrapped at some stage?) The idea isn't scrapped, but the graphics and sound will probably be arriving after the processor board has been developed. In the meantime, we still need some way to play samples, so interrupts are necessary if we want to be able to do anything else - like scanning the keyboard, or screen effects - at the same time. It may be acceptable to use the processor and timings to play samples through something like the Quazar (16 bit, surround stereo. Hey, shouldn't Colin be advertising his own stuff?) The other suggesttions would probably need a seperate processor, seperate (but accessible) memory - quite hard to implement. > Can we not also / instead (whatever) have a counter/timer circuit? > That'd be nice, easy, useful and cheap. Bonus! Like I said, a user definable time-based interrupt system would be ideal. But can that be designed cheaply or easily? Maybe... Remember it needs to be accurate, so perhaps the easiest route would be to have a port (or two) to which a 16 or 32 bit number can be sent; the circuit counts that number of clicks from the clock crystal before generating an interrupt. Yes, that sounds do-able. (Forgive me for my electronically uninformed comments, I may be asking the impossible there) Andrew From owner-sam-users@nvg.unit.no Thu Nov 7 19:32:46 1996 Date: Wed, 6 Nov 1996 11:59:08 +0100 Message-Id: <96110611590524@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: sam-users@nvg.unit.no Subject: Re: Tv or not TV X-Vms-To: SMTP%"sam-users@nvg.unit.no" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 300 Lines: 12 I use a TV which has a composite video input, which is linked up to my SAM via the SCART socket. The sound goes straight to my stereo, if anyone's interested. I reckon one of the selling points could (poss?) be: You don't need a monitor if you don't want one! (Though more catchy) Comments? Al. From owner-sam-users@nvg.unit.no Thu Nov 7 19:36:34 1996 Date: Thu, 7 Nov 1996 19:32:02 +0000 (GMT) From: "A.S. Collier" X-Sender: asc25@navy.csi.cam.ac.uk To: sam-users@nvg.unit.no Subject: Re: Fred 74 In-Reply-To: <96110611485335@morse.ntu.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 591 Lines: 20 On Wed, 6 Nov 1996, ALLAN CLARKSON wrote: > Yeah, I can't get RGB Demo to work either. > > Luckily, however, being less than 25 mins walk from Anderton's house > as I am... Nor does it work for me (FRED 74 only arrived today, but then it did have to be redirected from home) That said, I already have a copy of that demo (Thanks Martijn), just like the rest of MNEMOtech (except the ones who've joined since I posted the last disk round.) Deary me, looks like Anderton's really messed up this time. What has he got to say about this? (You're still on the list aren't you?) Andrew From owner-sam-users@nvg.unit.no Thu Nov 7 19:41:21 1996 Date: Thu, 7 Nov 1996 19:34:29 GMT Message-Id: <199611071934.TAA19692@pipe1.uk.pipeline.com> To: sam-users@nvg.unit.no Subject: Message understood. From: samsboss@uk.pipeline.com (Samsboss@uk.pipeline.com) Cc: sam-users@nvg.unit.no X-Pipeuser: samsboss X-Pipehub: uk.pipeline.com X-Pipegcos: (Samsboss@uk.pipeline.com) X-Mailer: Pipeline v3.5.0 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1374 Lines: 50 On Nov 07, 1996 03:35:27, 'BrenchleyR@aol.com' wrote: >Everyone please read this. BUT DO NOT REPLY USING THIS HEADING - invent your >own. >(sorry, but this heading business is doing my headin) I know, its so bad isn't it. > >Right. (lots cut to save space) >I think a couple of people have started to get way ahead of things over the >last week or so. We have to learn to walk first. We even have to prove that >we can co-operate enough to get a single small product out. Can the hardware >person get the design done, can the software man keep to a deadline? We >cannot realistically talk about building a new machine until we can prove we >can build a small interface together. Ok, but what should the first interface be? I would cast my vote for a proper bi-direction port so I can talk to the PC. > >(lots more cut) > >Add to that a buffer board so we don't overload the address/data lines. Now dat is a goody idea. >(more cut) > > >Now a few extra points that have been thought of. Each new board shouldl use >a 96 way connector so that the extra 32 lines can be allocated for new >signals between boards. All boards will have through connection so there is >no need for two-ups ect. Best idea we could have. Why did Bruce not opt for the 96way? > >(bit more cut) > >Bob. -- Samsboss. (Likeing the sound of it very much). From owner-sam-users@nvg.unit.no Thu Nov 7 19:41:21 1996 Date: Thu, 7 Nov 1996 19:34:32 GMT Message-Id: <199611071934.TAA19696@pipe1.uk.pipeline.com> To: sam-users@nvg.unit.no Subject: Re: Clock From: samsboss@uk.pipeline.com (Samsboss@uk.pipeline.com) Cc: sam-users@nvg.unit.no X-Pipeuser: samsboss X-Pipehub: uk.pipeline.com X-Pipegcos: (Samsboss@uk.pipeline.com) X-Mailer: Pipeline v3.5.0 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 724 Lines: 24 On Nov 02, 1996 11:24:33, 'BrenchleyR@aol.com' wrote: >In a message dated 02/11/96 00:31:39, you wrote: > >>> From: "Gavin Smith" >>> What ever happened to the clock that WC were going to bring out? We >>> really do need one now that the Sambus is no longer made. >> >>Well, I took out the clock circuitry from the sambus and designed a >>little board for a stand alone clock module, perfect for the 2-up or >>sumfin' like that, but it's just a case of producing the PCB's thats >>the problem. >> >> > >See this this months FORMAT fro the latest on the SAM_CLOCK. > >Bob. -- Any news in advance Bob? Could do with one to go with my hard disk interface. Samsboss. From owner-sam-users@nvg.unit.no Thu Nov 7 19:51:10 1996 From: Dave Hooper <9531427@EIGG.SMS.ED.AC.UK> Organization: Edinburgh Uni (CS and MATHS) To: sam-users@nvg.unit.no Date: Thu, 7 Nov 1996 19:53:13 +0000 Subject: Re: SOS colour Priority: normal X-Mailer: Pegasus Mail for Windows (v2.42a) Message-Id: <6978D6F481F@eigg.sms.ed.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 681 Lines: 15 er ... then isn't a PPM just the same as a RAW file with a zero-length header?(three bytes per pixel in raster order...) I started out on a PCX-SAM converter but it was pants. I did a SAM-PCX converter, but that's not really relevent, now, is it? (written for a PC and a SAM... I think... can't remember) I've also written an entirely useless SAM basic (tokenised, etc) to ASCII converter, for a PC. Anyone care? davee +---------------------------------------+---------------------+ | doubleback : complete bobbins | torus?- 1997? it'll | | http://www.geocities.com/area51/5636/ | never get finished | +---------------------------------------+---------------------+ From owner-sam-users@nvg.unit.no Thu Nov 7 20:18:22 1996 Message-Id: From: ee31ag@ee.surrey.ac.uk (Andrew M Gale) Subject: Re: Message understood. To: sam-users@nvg.unit.no Date: Thu, 7 Nov 1996 20:15:04 +0000 (GMT) In-Reply-To: <199611071934.TAA19692@pipe1.uk.pipeline.com> from "Samsboss@uk.pipeline.com" at Nov 7, 96 07:34:29 pm X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 247 Lines: 8 > Best idea we could have. Why did Bruce not opt for the 96way? > > Well, there's no point having 96pins if you've only got 64 signals to put down them! It's just as well he didn't anyway, because it leaves the middel row free for us to use... From owner-sam-users@nvg.unit.no Thu Nov 7 20:32:59 1996 From: Gavin Smith Organization: University of Ulster To: sam-users@nvg.unit.no Date: Thu, 7 Nov 1996 20:32:16 GMT Subject: What I would like in the SOS X-Confirm-Reading-To: "Gavin Smith" X-Pmrqc: 1 Priority: normal X-Mailer: Pegasus Mail for Windows (v2.23) Message-Id: <1BD3F947B9@smserver1.ulst.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2202 Lines: 45 Okay, I certainly don't have any of the technical abilities (yet) that a good lot of the people have on this list - the most techy thing I can do is program in C! :) I have, however used and learnt about a lot of computers and I know what I would like to see in a new Sam. Maybe you could take me as an example of a Sam owner who likes a bit of programming, plays the odd game, enjoys the Sam community, does the odd bit of word processing etc etc. I thought I would make a list (in no particular order) of very basic, general things I would like to see in a new Sam... Maybe someone could comment on which are practical? 1 - PC tower case (already decided upon it seems?) 2 - High density floppy drive 3 - Able to easily plug in standard 72-pin SIMMS and come with 4 or 8 meg 4 - Comes with internet software (mainly cos it well sell, and not cos I want to go "surfing" - although I would love to use my Sam for e-mail and IRC, both of which should be practical(?) 5 - It follows on that some kind of modem is needed - I wouldn't mind if it wasn't a standard modem 6 - On start-up I would like it to either go into either a CLI or a GUI where I could call the CLI in a window - the choice would be pre-configured in a previous sitting. 7 - Possibility of plugging in a CD-ROM which I understand shouldn't be too hard 8 - Hard Drive MUST be standard 9 - Standard PC keyboard and mouse 10 - Conversions of 2 or 3 big games - Worms, Civilisaton (2?) and Championship Manager 2 would be a very nice mixture 11 - Able to swop between Sam formatted and PC formatted disks with no hassle 12 - TV's should be the standard SAM display for marketing purposes but I think there are a lot of us who would like to use PC monitors with Sam 13 - A good C/C++ - though why we have to wait for the new SAM for that, I'm not sure... 14 - A new name would be nice :) SAMson is kind of cute for the prototype but I dunno if the finish thing should be called that. 15 - Able to use Zip drives/ Syquest drives etc - would they plug into the parallel port or would we need a SCSI? 17 - stlg400 or under As I said, very basic comments, but things that, as an average user, I would like to see. From owner-sam-users@nvg.unit.no Thu Nov 7 20:38:59 1996 Subject: RGB demo To: sam-users@nvg.unit.no Date: Thu, 7 Nov 1996 20:30:11 +0000 (GMT) Cc: Simon.Cooke@nessie.mcc.ac.uk (Simon Cooke) X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit From: Simon Cooke Message-Id: <96Nov7.213643+0100_met.46991-31612+49@sabre-wulf.nvg.unit.no> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 133 Lines: 5 Okay - someone explain how to get the RGB demo from Fred 75 working... I've hacked it to bits, and I can't get it to work... Simon From owner-sam-users@nvg.unit.no Thu Nov 7 20:56:04 1996 Message-Id: <199611072055.UAA12690@mail.enterprise.net> From: davewhitmore@enterprise.net (Dave Whitmore) To: sam-users@nvg.unit.no Subject: Daddy! Date: 07 Nov 1996 19:55:35 References: <96110513225090@morse.ntu.ac.uk> X-Newsreader: Spot 1.3 X-Mailer: NetGate 1.3 (Amiga; TCP/IP) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 615 Lines: 16 In a message of 05 Nov 96 ALLAN CLARKSON wrote to SAM-USERS@nvg.unit.no: AC> Awww, Dave's going to be a daddy again! AC> It just goes to show: nights *away* from the VDU can be just as AC> productive! Yeah, it took ages to get the spec right. We can use a lot of the hardware from the first model, but if it is a boy this time we're gonna need new software. I think a bundel will need to be prepared before the launch. Or we could issue that when the time comes.. We can't decide on any names yet though.. DaveSon, SAM, SAMantha.... :)) _ |_)ave (/\)hitmore email - davewhitmore@enterprise.net From owner-sam-users@nvg.unit.no Thu Nov 7 20:56:05 1996 Message-Id: <199611072055.UAA12700@mail.enterprise.net> From: davewhitmore@enterprise.net (Dave Whitmore) To: sam-users@nvg.unit.no Subject: Re : Killer apps - Internet Date: 07 Nov 1996 20:49:35 References: <26783.199611071333@lily.csv.warwick.ac.uk> X-Newsreader: Spot 1.3 X-Mailer: NetGate 1.3 (Amiga; TCP/IP) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1767 Lines: 52 In a message of 07 Nov 96 Mr P R Walker wrote to sam-users@nvg.unit.no: Hi Paul, >> Why don't we provide tools to use Fidonet too? MPR> The good thing about it is that you don't need special tools as such, MPR> just a bog standard terminal program. MPR> Like Termite. MPR> Simon? Sorry for jumping in. Yes, Termite or Comm'ix should both be great for fido. Comm'ix can do the downloading/uploading of mail packets right now. Kev Cooper uses Comm'ix for fido mail already, but he has to read and write on line. The only stumbling block at the moment is getting a good Off Line Reader program done. I suppose a point package would be best, but QWK or Blue Wave would be okay for starters(?). Another thing to do would be LZH(LHA), ZIP archiving though.. People who scoff at the thought of fidonet just don't know how good it really is. Maybe they've never tried it? (What do you reckon Brian?) A lot of people are taking advantage of BBSs that offer internet gateways through point packages. They can do email and newsgroups, although it isn't as fast as internet. It's a lot cheaper for people who have local access to such a BBS though. We need a good Fidonet point package/OLR for SAM. Then, we can think about a SAM fido echo - or just invade and take over the Spectrum echo. :) All the stuff about running before we can walk is right. Get the easy stuff done first, before messing about with TCP and all that. (IMHO) Think home user - instead of free university accounts. I mean, I wonder how many of this list would be f___ed without those! >From what I've heard about Termite, it should be brilliant.. if Simon ever finishes it (VBG Si.. VBG.. Honest! (GD&RFC)) :)). Bye, _ |_)ave (/\)hitmore email - davewhitmore@enterprise.net From owner-sam-users@nvg.unit.no Thu Nov 7 20:57:39 1996 Date: Thu, 7 Nov 1996 20:54:50 GMT From: Keith Turner Message-Id: <9611072054.AA11411@turner.cursci.co.uk> To: sam-users@nvg.unit.no Subject: Multitasking WIMPs Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 753 Lines: 19 Ian Collier said: > Also, it had better be pre-emptively multitasked. None of this Win 3.0 "one > bad app crashes the whole system" business... You need an MMU to avoid the Win 3.0 "one bad app crashes the whole system" business. In fact that's almost a definition of the purpose of an MMU. An MMU protects one applications memory space from another misbehaving application. The scheduling of task-switching can be either "pre-emptive" or "cooperative". Unix is capable of pre-emptive multi-tasking, but that feature might be a bit overblown for Samson. If one application refuses to cooperatively multitask, you can always bind a "Kill apps which are not responding" tool to an NMI interrupt, like the Windows 3.? Ctrl-Alt-Delete widget. / Organization: University of Ulster To: sam-users@nvg.unit.no Date: Thu, 7 Nov 1996 21:01:24 GMT Subject: Re : Killer apps - Internet X-Confirm-Reading-To: "Gavin Smith" X-Pmrqc: 1 Priority: normal X-Mailer: Pegasus Mail for Windows (v2.23) Message-Id: <1C500E4B9F@smserver1.ulst.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 328 Lines: 8 >People who scoff at the thought of fidonet just don't know how good it >really is. Maybe they've never tried it? The point is not whether it is good or not, the point is whether it will sell SAM's - this is where the internet is a priority. If someone wants to try fidonet after we have internet access, fine. Gavin Smith From owner-sam-users@nvg.unit.no Thu Nov 7 21:06:53 1996 Date: Wed, 6 Nov 1996 13:17:51 +0100 Message-Id: <96110613175095@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: SAM-USERS@nvg.unit.no Subject: Problems, problems... X-Vms-To: SAM-USERS@NVG.UNIT.NO Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 242 Lines: 13 Right, there *is* something odd with my e-mail thingy, and lots of messages that I sent earlier in the week have only just arrived, and so on... What had I said? Yeah: SAM Colour - purple It's got to be! And *who* could Samsboss be? Al. From owner-sam-users@nvg.unit.no Fri Nov 8 00:03:21 1996 Date: Thu, 7 Nov 1996 18:08:59 -0500 From: Neville Young <106166.1560@compuserve.com> Subject: PhD To: sam users Message-Id: <199611071811_MC1-BCF-21CC@compuserve.com> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 217 Lines: 10 >Alternatively, we make it all PhD. :-) Is it possible to get a PhD by designing a new >computer? >imc Don't think so but you can get CEng from the BCS (not that its worth much if you're looking for a job) Nev. From owner-sam-users@nvg.unit.no Fri Nov 8 00:03:21 1996 Date: Thu, 7 Nov 1996 23:49:11 GMT Message-Id: <199611072349.XAA01002@pipe1.uk.pipeline.com> To: sam-users@nvg.unit.no Subject: Re: Message understood. From: samsboss@uk.pipeline.com (Samsboss@uk.pipeline.com) Cc: sam-users@nvg.unit.no X-Pipeuser: samsboss X-Pipehub: uk.pipeline.com X-Pipegcos: (Samsboss@uk.pipeline.com) X-Mailer: Pipeline v3.5.0 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 452 Lines: 17 On Nov 07, 1996 20:15:04, 'ee31ag@ee.surrey.ac.uk (Andrew M Gale)' wrote: >> Best idea we could have. Why did Bruce not opt for the 96way? >> > > >Well, there's no point having 96pins if you've only >got 64 signals to put down them! It's just as well he >didn't anyway, because it leaves the middel row free >for us to use... -- Very good point. So, what are we goint to use the other 32 for? That is the next question. Samsboss. From owner-sam-users@nvg.unit.no Fri Nov 8 00:03:21 1996 Date: Thu, 7 Nov 1996 23:55:16 GMT Message-Id: <199611072355.XAA01270@pipe1.uk.pipeline.com> To: sam-users@nvg.unit.no Subject: Re: Re : Killer apps - Internet From: samsboss@uk.pipeline.com (Samsboss@uk.pipeline.com) Cc: sam-users@nvg.unit.no X-Pipeuser: samsboss X-Pipehub: uk.pipeline.com X-Pipegcos: (Samsboss@uk.pipeline.com) X-Mailer: Pipeline v3.5.0 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 829 Lines: 21 On Nov 07, 1996 21:01:24, '"Gavin Smith" ' wrote: >>People who scoff at the thought of fidonet just don't know how good it >>really is. Maybe they've never tried it? > >The point is not whether it is good or not, the point is whether it >will sell SAM's - this is where the internet is a priority. If >someone wants to try fidonet after we have internet access, fine. > >Gavin Smith -- Fidonet is good, because you like it - it does things the way you like them to be done. Other nets all have their good points. But the sad fact is - what people think of as the internet (in fact they are only really looking at the WWW) is what they /think/ they need. Once they have a machine, you are welcome to try and educate them, but it will be the magic word INTERNET that gets them. Samsboss. From owner-sam-users@nvg.unit.no Fri Nov 8 00:03:22 1996 Date: Thu, 7 Nov 1996 18:09:03 -0500 From: Neville Young <106166.1560@compuserve.com> Subject: Nothing To: sam users Message-Id: <199611071810_MC1-BCF-21CD@compuserve.com> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 130 Lines: 3 How many Microsoft employees does it take to screw in a light bulb? 95 if they stand near a window. From owner-sam-users@nvg.unit.no Fri Nov 8 05:00:05 1996 Date: Thu, 7 Nov 1996 10:36:54 -0500 From: BrenchleyR@aol.com Message-Id: <961107103654_1383229712@emout06.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1909 Lines: 43 In a message dated 07/11/96 14:51:27, you write: >[cut] >Unfortunately, from what I understand of the design constraints, there is >a requirement that the Z380 be able to access Sam's internal memory, since >(at first, at least) the display will be generated from existing circuitry >in the Sam, accessing the internal memory. To read from or write to >internal memory, the Z380 will need to run slowly. However, using the >external memory should pose no constraints on speed. So running most >programs will be much faster, even for screen-intensive applications - for >every write to the screen there will normally be one or maybe two reads >from the instruction memory. > >Hopefully, the electronics of this will be not *too* difficult to sort >out. > > >Andrew Does this make sence? For initial development, the Z380 does not need to access the SAM's RAM at all. There will be a 'communication block', possibley of SRAM, that can be accessed by both using hardware flags to control access. It should also be possible to taylor this transfer system so that interrupts are used to indicate when things are ready to use. Example. Z380 wants a block of data from disc. Create a request form in the comm-block and tell the Z80 to work on it. Meanwhile, carry on with something until the Z80 interrupts to tell you the block is ready for you to read. Now I know this slows things down, but this will do for starters - until the new Graphics Board is ready. When that comes along you can start driving the new screen modes direct. In other words, at first you will not be able to display a screen direct from the Z380, but you can dump a screen back into the Z80 for display any time you want to look at it. Bo used a similar system to in the early stage of FLASH, the programming was done on a Spectrum, run on a 'half way SAM' and then the screen was read back into the spectrum for examination. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 05:26:15 1996 Date: Thu, 7 Nov 1996 10:36:43 -0500 From: BrenchleyR@aol.com Message-Id: <961107103642_1881303951@emout20.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Why there was a MODE 2 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 329 Lines: 13 In a message dated 07/11/96 14:34:34, you write: >It was rumoured (by Chris White) that originally the Mode 2 mode was meant >to cater for hardware scrolling -- thus the 8K gap between the screen data >and the attributes. > >Any truth to this? > >Simon No. It was just easier to di it that way in the original hardware. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 07:15:41 1996 Date: Thu, 7 Nov 1996 10:36:56 -0500 From: BrenchleyR@aol.com Message-Id: <961107103655_1450338448@emout07.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Mode 2 for emulation? Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 501 Lines: 19 In a message dated 07/11/96 14:58:31, you write: >Why did no one (to my knowledge) go the whole hog and write a specific >TS emulator? >There were loads of Z80 based machines with circuitary simmilar the the >Speccy >so why didn't we get a Jupiter Ace emulator? There is one I think? > >Jupitor Ace was the Forth based machine I think, >is its ROM in the public domain yet? If so then we may want to put >SAMbasic, SAMGUI, C(++), Java AND FORTH on the flash memory! > > ;o) Numb. > > Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 07:26:52 1996 From: Johnna Teare Organization: University of Central Lancashire To: sam-users@nvg.unit.no Date: Thu, 7 Nov 1996 19:42:31 GMT+0 Subject: Re: Tv or not TV Priority: normal X-Mailer: Pegasus Mail for Windows (v2.33) Message-Id: Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 991 Lines: 28 Hello! > I use a TV which has a composite video input, which is linked up to > my SAM via the SCART socket. The sound goes straight to my stereo, if > anyone's interested. Is ther anything special about your SCART lead or is it just a bog standard one? What I'm trying to say is, can I just bung a SCART lead in my sam and then my telly and expect it to work? > I reckon one of the selling points could (poss?) be: You don't need a > monitor if you don't want one! (Though more catchy) How about...'You don't have to shell out for flash displays, coz tellys are used by the SAM Coupe' or something. We could get Gina G to sing a little jingle like that. Erm, just whilst I'm in the lazy mood: I'm still working on my wep pages and I want some other SAM URL's - please don't tell me to go look for them because my Netscape connection is down at the minute. Thank you! > Comments? Never play with grenades. Johnna Pig Teare (j.d.teare@uclan.ac.uk) "They call me Mad The Swine" From owner-sam-users@nvg.unit.no Fri Nov 8 07:27:06 1996 From: Johnna Teare Organization: University of Central Lancashire To: sam-users@nvg.unit.no Date: Thu, 7 Nov 1996 20:00:32 GMT+0 Subject: Re: WebSite (Please read...) Priority: normal X-Mailer: Pegasus Mail for Windows (v2.33) Message-Id: Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2709 Lines: 76 Hello, It's only me going on about my (bloody?!) website again. The following is a list of names and addresses (gotten from the SAM Users List) which I'm going to put into an address book on the site. Please contact me if: 1) You do not wish your name to be published 2) There are any details which are incorrect 3) You know that one of the people has changed address/removed themselves from the list 4) You are not on the list and would like to be added 5) There are any additional details you would like to have included by your name and address. Heres the list, and thanks for your help... Colin Anderton - plpmyljja@hhn1.nott.ac.uk Matthew Bealing - matthew@bealing.avel.co.uk Will Bowring - 9548777@EIGG.SMS.ED.AC.UK Bob Brenchley - FormatPub@aol.com Andrew Collier - asc25@cam.ac.uk Ian Collier - Ian.Collier@comlab.oxford.ac.uk Simon Cooke - simon.cooke@umist.ac.uk - Simon.Cooke@nessie.mcc.ac.uk Jeff Crawford - samcoupe@cadderly.demon.co.uk Ben Curren - ELA95BEC@sheffield.ac.uk Ian Dalziel - 100717.2266@compuserve.com Dan Doore - DOOREDJ@parliament.uk Stefan Drissen - drissen@pi.net - Stefan_Drissen@nl.coopers.com Brian Gaff - briansam@bgserv.demon.co.uk Andrew Gale - ee31ag@surrey.ac.uk David Gommeren - gommerd@interpac.be Simon Goodwin - simon@studio.woden.com Diggory Gray - graydj@phymat.bham.ac.uk Slawomir Grodkowski - slawek@math.uni-goettingen.de Jon Hampton - Jon_Hampton@iconex.mactel.org Dave Handley - d.handley@lancaster.ac.uk Steve Harding - sh5655@bristol.ac.uk Dave Hooper - d.c.hooper@sms.ed.ac.uk Dean Liversidge - dean@error.demon.co.uk Neil Maynard - mne2@cableol.co.uk David Mundon - davidm@enterprise.net J.K Ogden - se94jko@ex.ac.uk Simon Owen - si@obobo.demon.co.uk Sebastian Palucha - palucha@usctoux1.cto.us.edu.pl Rob Partington - rjp@heffer.demon.co.uk Tim Paveley - unc@dplinux.sund.ac.uk - unc@mono.org Matt Round - malevolent@netwales.co.uk Arne di Russo - ar@RMnet.it Allan Skillman - allan@hpopb1.cern.ch Justin Skists - c93js1@dmu.ac.uk Gavin Smith - smith-gc@ulst.ac.uk Johnna Teare - j.d.teare@uclan.ac.uk Frode Tennebo - ft@edh.ericsson.se Luke Trevorrow - blackadder@orinocco.demon.co.uk Keith Turner - keith@cursci.co.uk Robert van der Veeke - rjvveeke@caiw.nl Ben Versteeg - ben@charm.il.ft.hse.nl Paul Walker - P.R.Walker@csv.warwick.ac.uk - csuan@csv.warwick.ac.uk Tim Wells - tgw1001@cam.ac.uk Lee Willis - l.willis@comp.brad.ac.uk Geoff Winkless - geoffw@jumper.mcc.ac.uk Dave Whitmore - davewhitmore@enterprise.net David Zambonini - D.M.Zambonini@cs.cf.ac.uk Gianni Zamperini - gianni.zamperini@galactica.it - BBKaneda@galactica.it Once again, thanks... Johnna Pig Teare (j.d.teare@uclan.ac.uk) "They call me Mad The Swine" From owner-sam-users@nvg.unit.no Fri Nov 8 08:02:27 1996 Date: Thu, 7 Nov 1996 09:33:35 +0100 Message-Id: <96110709333569@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: sam-users@nvg.unit.no Subject: Re: Son Of Sam X-Vms-To: SMTP%"sam-users@nvg.unit.no" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 352 Lines: 11 Cases: You MUST choose a standard case. IBM compats have no identity. You take the writing off a Speccy, SAM, C64, Amstrad, Amiga, ST, Mac, etc. and you can still tell which is which. Take the writing off a PC and you won't be able to tell who made which or what's inside. Choose ONE standard case, so that when people see it, they say SAMSON! Al. From owner-sam-users@nvg.unit.no Fri Nov 8 08:08:28 1996 Date: Thu, 7 Nov 1996 09:42:36 +0100 Message-Id: <96110709423659@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: SAM-USERS@nvg.unit.no Subject: NSSS 2 X-Vms-To: SAM-USERS@NVG.UNIT.NO Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 68 Lines: 5 To TIM W: Well, you can come to the show during the day, tho? Al. From owner-sam-users@nvg.unit.no Fri Nov 8 08:22:54 1996 Date: Thu, 7 Nov 1996 09:56:48 +0100 Message-Id: <96110709564803@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: sam-users@nvg.unit.no Subject: Re: Tv or not TV (Nev read this) X-Vms-To: SMTP%"sam-users@nvg.unit.no" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 335 Lines: 11 I think people would be likely to like a TV option if it meant the whole package was cheaper. How about an option on startup, like the QL has? Or, if we're going to have the GUI/CLI startup thing, how about doing the same for MONITOR/TV? In early versions, perhaps this could be a manual switch while we get the HD sorted out? Al. From owner-sam-users@nvg.unit.no Fri Nov 8 08:23:05 1996 Date: Thu, 7 Nov 1996 10:01:05 +0100 Message-Id: <96110710010583@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: sam-users@nvg.unit.no Subject: Re: Re : Killer apps. X-Vms-To: SMTP%"sam-users@nvg.unit.no" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 110 Lines: 6 PcW16 > It was the new one I was refering to, SLOWWWWWWWW... Oh. Sorry. Is the new one really that bad? Al. From owner-sam-users@nvg.unit.no Fri Nov 8 08:34:35 1996 Date: Thu, 7 Nov 1996 10:16:50 +0100 Message-Id: <96110710164957@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: sam-users@nvg.unit.no Subject: Re: SAMson marketing X-Vms-To: SMTP%"sam-users@nvg.unit.no" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 193 Lines: 7 Bundled software it has to be. I've asked a few people, and they said that a few games, a word processor, and that would be a big plus. How about a football game? We've not had one yet. Al. From owner-sam-users@nvg.unit.no Fri Nov 8 08:41:23 1996 Date: Fri, 8 Nov 1996 09:41:15 +0100 From: ft@edh.ericsson.se (Frode Tenneboe) Message-Id: <9611080841.AA03939@asmal.edh-net> To: sam-users@nvg.unit.no Subject: Re: SAMson marketing X-Sun-Charset: US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 270 Lines: 8 > Bundled software it has to be. I've asked a few people, and they said that > a few games, a word processor, and that would be a big plus. > > How about a football game? We've not had one yet. Perhaps we can persuade Jon Ritman - for good 'ol times sake? :) -Frode From owner-sam-users@nvg.unit.no Fri Nov 8 08:46:21 1996 Date: Thu, 7 Nov 1996 10:23:50 +0100 Message-Id: <96110710234929@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: sam-users@nvg.unit.no Subject: Re: Re : Killer apps - Internet X-Vms-To: SMTP%"sam-users@nvg.unit.no" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 310 Lines: 12 > Imagine the adverts ... > > Explore the Web and read e-mail using your TV, your existing phone-line, and > a Samson Network Computer retailing at 399.99 pounds > >The demand could be huge. Yeah, that would be great. There's a guy sitting next to me now is saying it would be a big pulling point, too. Al. From owner-sam-users@nvg.unit.no Fri Nov 8 08:50:12 1996 Date: Thu, 7 Nov 1996 10:24:55 +0100 Message-Id: <96110710245555@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: sam-users@nvg.unit.no Subject: Re: Tv or not TV X-Vms-To: SMTP%"sam-users@nvg.unit.no" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 109 Lines: 5 I'm not sure if I said this, but I use a TV with a composite video input connected to my SCART output. Al. From owner-sam-users@nvg.unit.no Fri Nov 8 08:59:10 1996 Date: Thu, 7 Nov 1996 10:32:01 +0100 Message-Id: <96110710320095@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: sam-users@nvg.unit.no Subject: Re: The 8 bit SAMSON X-Vms-To: SMTP%"sam-users@nvg.unit.no" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 102 Lines: 7 >(Why don't we all just shut up and go buy a Pentium PC...) > >imc I'm not like everybody else. Al. From owner-sam-users@nvg.unit.no Fri Nov 8 09:05:33 1996 Date: Thu, 7 Nov 1996 10:39:58 +0100 Message-Id: <96110710395754@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: sam-users@nvg.unit.no Subject: Re: Hello. X-Vms-To: SMTP%"sam-users@nvg.unit.no" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 200 Lines: 6 > MMmmm.. Yes! We'll have the case chocolate orange FLAVOURED! Now, now, James. You know it wouldn't last five minutes before you ate it... and we wouldn't want you to do that now, would we? Al. From owner-sam-users@nvg.unit.no Fri Nov 8 09:06:50 1996 Date: Thu, 7 Nov 1996 10:44:27 +0100 Message-Id: <96110710442616@morse.ntu.ac.uk> From: CC604050@ntu.ac.uk (ALLAN CLARKSON) To: SAM-USERS@nvg.unit.no Subject: SAMSON: ISDN? X-Vms-To: SAM-USERS@NVG.UNIT.NO Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 112 Lines: 5 Someone mentioned ISDN compatibility? Get real - I thought this was aimes at home users not millionaires. Al. From owner-sam-users@nvg.unit.no Fri Nov 8 10:53:30 1996 Date: Fri, 8 Nov 1996 10:42:05 +0000 (GMT) From: Tim Paveley To: sam-users@nvg.unit.no Cc: sam-users@nvg.unit.no Subject: Re: WebSite (Please read...) In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 969 Lines: 23 On Thu, 7 Nov 1996, Johnna Teare wrote: > It's only me going on about my (bloody?!) website again. The > following is a list of names and addresses (gotten from the SAM Users > List) which I'm going to put into an address book on the site. I've got a list of people at http://www.mono.org/~unc/Coupe/net/emails.html which are people who've said they don't mind me putting there email address up on the web so 2 points. 1) You may like to see if there are any on there you don't have, I don't think they are all on the mailing list, though I think some maybe out of date (it's a tad hard to check other than spamming everyone every so often which I don't really want to do) 2) If no-one complains could you let me know and I'll bung them all on my list too ;) Tim ....@/ .........................................................................@/ Unc - Tim Paveley - http://dplinux.sund.ac.uk/~unc/ Owner of a Sam Coupe - http://www.mono.org/~unc/Coupe/ From owner-sam-users@nvg.unit.no Fri Nov 8 11:02:18 1996 Date: Fri, 8 Nov 1996 10:55:40 +0000 (GMT) From: "A.S. Collier" X-Sender: asc25@puce.csi.cam.ac.uk To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message In-Reply-To: <961107103654_1383229712@emout06.mail.aol.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2985 Lines: 62 On Thu, 7 Nov 1996 BrenchleyR@aol.com wrote: > >Unfortunately, from what I understand of the design constraints, there is > >a requirement that the Z380 be able to access Sam's internal memory, since > >(at first, at least) the display will be generated from existing circuitry > >in the Sam, accessing the internal memory. > > Does this make sence? No, but it does make sense. Sorry.... ;) > For initial development, the Z380 does not need to access the SAM's RAM at > all. There will be a 'communication block', possibley of SRAM, that can be > accessed by both using hardware flags to control access. It should also be > possible to taylor this transfer system so that interrupts are used to > indicate when things are ready to use. > > In other words, at first you will not be able to display a screen direct from > the Z380, but you can dump a screen back into the Z80 for display any time > you want to look at it. Bo used a similar system to in the early stage of > FLASH, the programming was done on a Spectrum, run on a 'half way SAM' and > then the screen was read back into the spectrum for examination. Well, yes I suppose that is one way to do it. I'd be disappointed if - given this super-duper processor - we still can't scroll the Sam's screen in one frame because we're relying on the Z80B to do the work.... However I do agree your method does have some advantages, particularly in terms of reading from disk; I can see how sharing the processor load would make some games work quite well - Z80B drawing things up on the screen while Z380 works actually controlling the sprites and collision detection (now THAT, I think, is the hardest part of writing a game!) Will the Z380 be able to access the Sam's ports? Otherwise things like keyboard/mouse reading would have to be done from the communications block, sound chip data would have to be written there too - sample players would be particularly hard on the Z80B if any screen changes were made... Perhaps this is a short-term problem until a sound board arrives? I dunno, I'd prefer the method I described (already prototyped by Simon and Martin, so it does work. Nearly) but if yours was just as easy to implement..... Could both be done, getting the best of both worlds? I don't have the electronics experience to say what could or couldn't be done; I just do the software projects. Does anybody out there really *know*? Couple more points: How do we actually get our programs into the memory of the new processor and start them up? I presume the Z80B will output to some port, which starts up the Z380. Could the communications block be located at the start of Z380's memory, and an almost-boot routine be copied there by the Z80B? The Z380 starts running that routine, and copies more of the relavent code into its own memory... Would that mean we don't (yet) need a ROM-style block? Perhaps that would do for the developers' run, before any system code has actually been written. Andrew From owner-sam-users@nvg.unit.no Fri Nov 8 11:36:49 1996 Message-Id: From: ee31ag@ee.surrey.ac.uk (Andrew M Gale) Subject: Re: Plea! Plea! Plea! + Bob's message To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 11:33:28 +0000 (GMT) In-Reply-To: from "A.S. Collier" at Nov 8, 96 10:55:40 am X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1707 Lines: 44 > I dunno, I'd prefer the method I described (already prototyped by Simon > and Martin, so it does work. Nearly) but if yours was just as easy to > implement..... Could both be done, getting the best of both worlds? I > don't have the electronics experience to say what could or couldn't be > done; I just do the software projects. Does anybody out there really > *know*? > Yes, they could both be done - and think of the advantage: * use the SAM's z80 for text or simple graphic displays, and let the z380 deal with more complicated graphical matters * tell the z80 to load a chunk from disk, the z380 can get on and do something else meanwhile. When the z80 is finished, it interrupts the z380, the z380 pulls the z80's BUSRQL low, and then accesses the SAM's memory directly, probably doing a block transfer to its own memory * When printing, the z380 accesses the SAM's RAM directly and dumps a textfile into it. It then tells the z80 to send the file to a printer, and the z380 can get on with something else meanwhile * The z380 dumps a sound sample to the z80, and the z80 plays that file whilst the z380 does something else * Existing SAM programs could be improved by running them on the z80 and getting the z380 to play samples as appropriate - I'd guess it'd only need a few minor alterations to the code of most SAM programs. I'm sure there are plenty more ideas you could think of... -Andy ps: do you think the extra processor really needs to be a z380? 68000s are cheap and have an obvious upgrade path, the 80386 is really cheap, and again has an obvious upgrade. But the z380 code will be familiar to us all. (Also, a 68000 based one will be seen as a poor Amiga or ST) From owner-sam-users@nvg.unit.no Fri Nov 8 11:58:13 1996 Message-Id: <199611081154.MAA31247@dxmint.cern.ch> From: Allan Skillman Subject: SamSon memory management To: sam-users@nvg.unit.no Date: Fri, 8 Nov 96 12:54:27 MET Mailer: Elm [revision: 70.85] Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1357 Lines: 25 Here is my proposition for the SamSon's memory management The z380 has a linear memory map, from 0x00000000 up to 4Gb The z80b has a paged memory map , from 0x0000 to 0xffff In normal operating mode these two memopry maps are separate. My idea is to allow the 64k Sam map to be paged into the z380 map via an I/O call by the z380. The simplest place to put the 64K if in the z380s zero page (0x00000000 to 0x0000ffff) but if possible it would be nice to be able to page it in at any point (mod 64k). In this mode accessing Sam memory segment would be done at normal (slow) Sam speed. With this model the z380 can access the Sam display memory, without any handshaking ram. Allan +--------------------------------------------------------------------------+ | Now available! XCoupe, the one and only SAM Coupe emulator | | *** http://www.hep.ucl.ac.uk/~ajs/xcoupe *** | +------------------------------+-------------------------------------------+ | Allan Skillman | "There are five flavours of resons, the | | HEPP Group | elementary particles of magic : up, down, | | University College London | sideways, sex-appeal and peppermint." | | Email : ajs@hep.ucl.ac.uk | - Terry Pratchett (Lords and Ladies) | +------------------------------+-------------------------------------------+ From owner-sam-users@nvg.unit.no Fri Nov 8 12:22:47 1996 Message-Id: X-Mailer: Novell GroupWise 4.1 Date: Fri, 08 Nov 1996 12:20:06 +0000 From: Dan Doore To: sam-users@nvg.unit.no Subject: SAMSON: ISDN? - Reply Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 539 Lines: 21 > Someone mentioned ISDN compatibility? Get real - I > thought this was aimes at home users not millionaires. What's wrong with ISDN? Prices are coming down (allegedly) all the time :) There is nothing implicit about "ISDN compatibility": AFAIK all an ISDN TA requires is a serial port and it can act like an ordinary modem. The point I was making was that it would be good to have a nice fast serial port/USB port for a nice fast serial device if you have one. But if you *want* to use a 9600 modem, I'm not stopping you. :)) Dan. From owner-sam-users@nvg.unit.no Fri Nov 8 13:10:55 1996 Date: Thu, 7 Nov 1996 17:21:55 GMT From: Mail Delivery Subsystem Subject: Returned mail: Remote protocol error Message-Id: <199611071721.RFE29731@harrier.fen.bris.ac.uk> To: sh5655 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2921 Lines: 72 The original message was received at Thu, 7 Nov 1996 15:17:45 GMT from hadrian [137.222.90.29] ----- The following addresses had delivery problems ----- (unrecoverable error) ----- Transcript of session follows ----- 554 ... Remote protocol error ----- Original message follows ----- Return-Path: Received: from hadrian.bris.ac.uk by harrier.fen.bris.ac.uk (SMI-8.6/SMI-SVR4) id PAA14145; Thu, 7 Nov 1996 15:17:45 GMT Received: by hadrian.bris.ac.uk (SMI-8.6/SMI-SVR4) id PAA02114; Thu, 7 Nov 1996 15:13:55 GMT From: sh5655@harrier.fen.bris.ac.uk (SL Harding) Message-Id: <199611071513.PAA02114@hadrian.bris.ac.uk> Subject: Re: SOS GUI written in C!? To: sam-users@nvg.unit.no Date: Thu, 7 Nov 1996 15:13:54 +0000 (GMT) In-Reply-To: from "Justin Skists" at Nov 7, 96 01:51:07 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit content-length: 1835 > Yes. I believe that! I'm a fairly decent C and Pascal programmer. If we had > ANSI C, we could easily port things like ML and Prolog across. Web browsers > could be knocked up in a few weeks. Games could be knocked up quickly. The GUI > system would be done by now.... > I don't think writing the opperating system in C is a good idea! True, it may be easier and faster to write but it is also more likely to be seriously bugged and will run slower than a pure assembler one. All of the desktop accessorys you get with the machine had better be done in Z380 assembler. At first the speed of the utils may not seem that important, but when start to multitask a few, the inefficiencies will add up. Without the pageing we need to start thinking about another format for the opperation of the desktop accessories. Where will they live and store their variables? How will they multitask? > I'm a crap Z80 programmer. All of my utility programs are written in BASIC > (with some Z80 to speed up stuff). If we had ANSI C, my SMIDIP program would > be on release 2 by now... (With the rest of the SMIDIT suite). I feared this. With a more powerfull machine will we become lazy and start to write Sam games in C? -THIS FEAR WILL NOT BE AS IMPORTANT IF THE COMPILER WE GET WITH THE MACHINE IS GREAT AT OPTIMISATION. > So, in my opinion, ANSI C for SAM should be a higher priority than SOS... :) > C scares me! > -- > ============================================================================= > |Justin Skists (c93js1@dmu.ac.uk) | Don't take life so seriously, | > |BSc (Hons) Computer Science, Year 4 | you'll never survive it! | > |De Montfort University, Leicester, England | - (C) CamLoRTS 1996 | > ============================================================================= > Numb. From owner-sam-users@nvg.unit.no Fri Nov 8 13:11:21 1996 Date: Fri, 8 Nov 1996 13:09:05 +0000 (GMT) From: Tim Wells To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2394 Lines: 49 In reply to what A.S. Collier wrote, My thoughts on a number of points: The communications block would be simple to arrange, but: No real-time screen updates from z380 - when it writes back that screen, the Z80B has to shift that entire screen from the comms block to its own memory - Slow. No direct access to SAMs ports from Z380 (they use Z80Bs address bus - so accessing SAM's ports is similar to directly accessing memory. Z380 loses all contact to outside world when Z80B does something. (Z80B can only do one thing at once, particularly if that one thing is time dependent). This will include playing samples whilst updating screen, etc. A dual solution is really needed, as there will need to be communication between the two processors, which can be accessed independently - ie so the Z80B can say "I'm busy writing to disk - don't touch my address bus", without the Z380 having to touch the Z80B address bus to find that out. However the Z380 really needs to access the SAM display without going via the Z80B. A write-through cache, as someone suggested? > Couple more points: How do we actually get our programs into the memory of > the new processor and start them up? I presume the Z80B will output to > some port, which starts up the Z380. Could the communications block be > located at the start of Z380's memory, and an almost-boot routine be > copied there by the Z80B? The Z380 starts running that routine, and copies > more of the relavent code into its own memory... Would that mean we don't > (yet) need a ROM-style block? Perhaps that would do for the developers' > run, before any system code has actually been written. The last thing we want is the Z80B to try and use the Z380 memory. This would be possible with a bit of hardware, but a much easier solution would be to have an EPROM in there - just to start things up. Then the EPROMs can be updated as system code is developed. All that EPROM need contain at the start is a little bit of code which examines the communication port for "Upload boot code block", and shifts the code from SAM memory into its own memory, and starts to execute it. Tim W. ------------------------------------------------------------------------------ = Tim Wells -- tgw1001@cam.ac.uk -- http://www2.eng.cam.ac.uk/~93tgw = ------------------------------------------------------------------------------ From owner-sam-users@nvg.unit.no Fri Nov 8 13:18:11 1996 From: Mr P R Walker Message-Id: <22440.199611081314@holly.csv.warwick.ac.uk> Subject: Re: Internet/cost/PC's To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 13:14:18 +0000 (GMT) In-Reply-To: <17281192107@mail-gw.uclan.ac.uk> from "Johnna Teare" at Nov 7, 96 05:29:09 pm X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 417 Lines: 10 > We all boo and hiss at the sound of Microsoft and IBM but for many, > they are the computer industry and anything that doesn't carry their > badge is inferior to them. I know this is a sorry state of affairs Exactly the point I was trying to make earlier :) Although I don't think I put it quite as well, I must be honest. > they'll be retailing for a little less than stlg300. Fair enough, Less than what?? From owner-sam-users@nvg.unit.no Fri Nov 8 13:22:43 1996 Date: Fri, 8 Nov 1996 13:17:15 +0000 (GMT) From: Tim Wells To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1312 Lines: 31 On Fri, 8 Nov 1996, Andrew M Gale wrote: > * The z380 dumps a sound sample to the z80, and the z80 plays > that file whilst the z380 does something else Not if the Z380 wants to display anything - the z80 will start to play, and then find it can't touch its address bus because the Z380 is busy sending screen information down it. > * Existing SAM programs could be improved by running them on > the z80 and getting the z380 to play samples as appropriate > - I'd guess it'd only need a few minor alterations to the code > of most SAM programs. This might well work, but I don't think it'll improve any programs - imagine that the z80 starts to draw a lemming perhaps, and then the z380 grabs control of the bus to start sending the next chunk of its samples... The Z80B's address bus will always be a bottleneck here, > ps: do you think the extra processor really needs to be a z380? > 68000s are cheap and have an obvious upgrade path, the 80386 is > really cheap, and again has an obvious upgrade. But the z380 code > will be familiar to us all. (Also, a 68000 based one will be seen > as a poor Amiga or ST) As you've said - the Z380 code will be familiar - existing programs will only need altering to take account of the different hardware, not an entirely different instruction set. Tim W. From owner-sam-users@nvg.unit.no Fri Nov 8 13:25:10 1996 From: James R Curry Organization: De Montfort University To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 13:25:41 GMT Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: Re: SOS colour Priority: normal X-Mailer: Pegasus Mail for Windows (v2.42a) Message-Id: Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 625 Lines: 15 > I've also written an entirely useless SAM basic (tokenised, etc) to > ASCII converter, for a PC. Anyone care? Arghhhhh! Yes! I'd love a copy! (I started to write one myself, but didn't have a list of all the tokens, so spent days figuring them out..... It worked as well, for three lines of basic. Then the fourth read something like LOAD = PRINT DUMP "dhf0ghed09t43.". Since it was written in BASIC (As a PROC) it was pretty slow too. Ho-hum.) -- James R Curry hc95jc@dmu.ac.uk (Wondering what this craze of putting your current feelings/actions in brackets in your signature file is all about. ;) ) From owner-sam-users@nvg.unit.no Fri Nov 8 13:36:16 1996 Date: Fri, 8 Nov 1996 13:27:22 +0000 (GMT) From: Tim Wells To: sam-users@nvg.unit.no Subject: Re: NSSS 2 In-Reply-To: <96110709423659@morse.ntu.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 446 Lines: 12 On Thu, 7 Nov 1996, ALLAN CLARKSON wrote: > Well, you can come to the show during the day, tho? Love to, but with the state of our railways at the moment, it looks like I would get there about twelve, and have to leave about one. Six hours by train and thirty odd quid for one hour at the show. If it wasn't for the dinner, then I could have stayed with my parents in Leeds overnight, and the trip would be a little more worthwhile. Tim W. From owner-sam-users@nvg.unit.no Fri Nov 8 13:39:29 1996 Date: Fri, 8 Nov 1996 13:34:46 +0000 (GMT) From: Tim Wells To: sam-users@nvg.unit.no Subject: Re: Tv or not TV In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 521 Lines: 12 > Is ther anything special about your SCART lead or is it just a bog > standard one? What I'm trying to say is, can I just bung a SCART lead > in my sam and then my telly and expect it to work? 1) Get a SAM Scart lead - the SAM connections are non-standard. See Bob for details 2) If it doesn't work, change the ends around - the lead isn't symmetrical. 3) Suddenly realise that you can read 85 column text in mode 3 after all :) In general you can expect the SCART connection to work without problems. Tim W. From imc Fri Nov 8 13:57:57 1996 Subject: Re: Multitasking WIMPs To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 13:57:57 +0000 (GMT) In-Reply-To: <9611072054.AA11411@turner.cursci.co.uk> from "Keith Turner" at Nov 7, 96 08:54:50 pm X-Mailer: ELM [version 2.4 PL24 PGP6] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1303 Lines: 31 On Thu, 7 Nov 1996 20:54:50 GMT, Keith Turner said: > Ian Collier said: > > Also, it had better be pre-emptively multitasked. None of this Win 3.0 "one > > bad app crashes the whole system" business... > You need an MMU to avoid the Win 3.0 "one bad app crashes the whole system" > business. Well I believe I already asked for an MMU. :-) But actually I was just talking about infinite loops. > The scheduling of task-switching can be either "pre-emptive" or "cooperative". > Unix is capable of pre-emptive multi-tasking, but that feature might be a bit > overblown for Samson. Why? As far as I can tell, pre-emptive multitasking is no more difficult than cooperative multitasking (and you don't have to provide a "give up control" function and insist the programmer uses it). Whether this will work depends on whether there's something that the program might want to do which absolutely will not work if it gets interrupted. If there is (and I can't think of anything except tape saving, disk accesses and sample playing) then this can be provided by a kernel service routine, or else there could be a "I really don't want to be interrupted" function call. Bear in mind that such things as games will probably prefer to boot up by themselves and not run in the windowing system at all. imc From owner-sam-users@nvg.unit.no Fri Nov 8 14:07:33 1996 Date: Fri, 8 Nov 1996 14:01:18 +0000 (GMT) From: Justin Skists X-Sender: c93js1@galois.cms.dmu.ac.uk To: sam-users@nvg.unit.no Subject: Re: Daddy! In-Reply-To: <199611072055.UAA12690@mail.enterprise.net> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 672 Lines: 16 On 7 Nov 1996, Dave Whitmore wrote: > We can't decide on any names yet though.. > DaveSon, SAM, SAMantha.... :)) ^^^^^^^^ This is what I've called my SAM. (Another useless mail from the guy who names all of his computers) -- ============================================================================= |Justin Skists (c93js1@dmu.ac.uk) | Don't take life so seriously, | |BSc (Hons) Computer Science, Year 4 | you'll never survive it! | |De Montfort University, Leicester, England | - (C) CamLoRTS 1996 | ============================================================================= From owner-sam-users@nvg.unit.no Fri Nov 8 14:27:00 1996 Date: Fri, 8 Nov 1996 14:16:46 +0000 (GMT) From: Justin Skists X-Sender: c93js1@galois.cms.dmu.ac.uk To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1093 Lines: 26 On Fri, 8 Nov 1996, Andrew M Gale wrote: > Yes, they could both be done - and think of the advantage: > [snip - things about z380 and z80 working together] That *is* what I call *true* multiprocessing! > ps: do you think the extra processor really needs to be a z380? > 68000s are cheap and have an obvious upgrade path, the 80386 is > really cheap, and again has an obvious upgrade. But the z380 code > will be familiar to us all. (Also, a 68000 based one will be seen > as a poor Amiga or ST) No.. Don't.. Keep to Z80 or at least something like it... We don't want to clone any other computers other than the SAM! (Or it won't be called SAMson. Instead it would be called SAM-adopted-son) -- ============================================================================= |Justin Skists (c93js1@dmu.ac.uk) | Don't take life so seriously, | |BSc (Hons) Computer Science, Year 4 | you'll never survive it! | |De Montfort University, Leicester, England | - (C) CamLoRTS 1996 | ============================================================================= From owner-sam-users@nvg.unit.no Fri Nov 8 14:31:56 1996 From: Johnna Teare Organization: University of Central Lancashire To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 13:42:40 GMT+0 Subject: Re: Saviour of the universe... Priority: normal X-Mailer: Pegasus Mail for Windows (v2.33) Message-Id: <12896B28C8@mail-gw.uclan.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 304 Lines: 11 > Bo used a similar system to in the early stage of > FLASH, the programming was done on a Spectrum, run on a 'half way SAM' and > then the screen was read back into the spectrum for examination. And boy does it show...! > Bob. Johnna Pig Teare (j.d.teare@uclan.ac.uk) "They call me Mad The Swine" From owner-sam-users@nvg.unit.no Fri Nov 8 14:38:09 1996 Date: Fri, 8 Nov 1996 14:34:01 +0000 (GMT) From: Justin Skists X-Sender: c93js1@galois.cms.dmu.ac.uk To: SAM users list Subject: SOS: Multitasking Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1129 Lines: 25 Just a quicky before I go home for the weekend (and wait for 300-odd email messages):- How is it possible to do pre-emptive multitasking on the Z380? As far as I know, the processor needs to have the ability to take away control from one program and give it to another. I'm pretty sure the Z80B doesn't have this feature. Does the Z380? OK, I know that you could use interrupts, but couldn't a program issue a DI and keep the processor to itself? I suppose you could design a "kernel" that runs p-code type instructions (like Visual Basic and Java) but that would be pretty slow compared to assembler. Personally, I think only co-operative multitasking is possible... (Eeek. A first decent email to this list from me in a while) -- ============================================================================= |Justin Skists (c93js1@dmu.ac.uk) | Don't take life so seriously, | |BSc (Hons) Computer Science, Year 4 | you'll never survive it! | |De Montfort University, Leicester, England | - (C) CamLoRTS 1996 | ============================================================================= From owner-sam-users@nvg.unit.no Fri Nov 8 14:56:57 1996 Message-Id: <1.5.4.32.19961108144948.00932f54@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 14:49:48 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: SOS: Multitasking Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 546 Lines: 14 At 02:34 PM 11/8/96 +0000, you wrote: >As far as I know, the processor needs to have the ability to take away control >from one program and give it to another. I'm pretty sure the Z80B doesn't have >this feature. Does the Z380? > >OK, I know that you could use interrupts, but couldn't a program issue a DI and >keep the processor to itself? I suppose you could design a "kernel" that runs >p-code type instructions (like Visual Basic and Java) but that would be pretty >slow compared to assembler. Just use the NMI :) - we may as well! Simon From owner-sam-users@nvg.unit.no Fri Nov 8 15:11:46 1996 Date: Fri, 8 Nov 1996 10:03:23 -0500 From: BrenchleyR@aol.com Message-Id: <961108100320_1613878155@emout08.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: SAMson marketing Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 3156 Lines: 89 In a message dated 08/11/96 04:06:39, you write: > >On Tue, 5 Nov 1996 BrenchleyR@aol.com wrote: > >> We do not need bundled anything. But we do need a reasonable range of >> software from day one, that is why I'm going the add-it-on-to-SAM route to >> start with. > >OK. I accept that. I was thinking in terms on how we could attack the games >playing market. No-body will by the new snazzy >super-duper-sega-saturn-mach-2000 if it didn't have a game bundled with it.. See your point, but... Games consols are different, you can't do a thing with them without some game to plug in. In the computer market it is a little different. But lets get the hardware underway first, when the launch is getting neaerr we can worry about the software. > >> >As soon as we get a decent ANSI C for the SAM, we could have anything we >> damn >> >well please! :) >> >> If you believe that, can I interest you in a map I've got to some hidden >> treasure - only 10,000 UKP to you. > >Yes. I believe that! I'm a fairly decent C and Pascal programmer. If we had >ANSI C, we could easily port things like ML and Prolog across. Web browsers >could be knocked up in a few weeks. Games could be knocked up quickly. The >GUI >system would be done by now.... When Nev got high super-expensive PC based Ansi C compiler with Z80 source code output (what a mouthful) Him and I spent ages hunting hi and low for C source that could be ported to SAM. Result? Nothing. So where is all this C source that we can desend on an get onto SAM? If anyone can supply something, Nev can look at it for us. > >I'm a crap Z80 programmer. All of my utility programs are written in BASIC >(with some Z80 to speed up stuff). If we had ANSI C, my SMIDIP program would >be on release 2 by now... (With the rest of the SMIDIT suite). > >So, in my opinion, ANSI C for SAM should be a higher priority than SOS... :) We at least have a C at the moment, it may be SMALL C but it is better than nothing. And as C will only appeal to a very small minority, I do think there are other more important things at the moment. > >As for the map: Could I have it on a two week approval? :) No. Cash up front. > >> What you get up to in your own home is none of our business... (grin) > >I should be so lucky! :) > >> >Remember, we can't even think about marketing SAMson without software >> >(especially games) ready for it. *THAT* was, IMHO, *THE* problems with SAM >> in >> >the first place. Being late didn't really have that much to do with it >> >(although it did help)! >> >> Umm, not really true. > >We are all entitled to our own opinions... But, when I was doing my >A-Levels, >we were comparing Amigas, STs, and SAMs together. I managed to defend the SAM >with every argument they threw at me. Hell, I even managed to throw some back >at them to the point when I almost convinced a few people to get one. But >when >they asked me what games were available the only answer I could come up with >was "Ermm.. Defenders of the Earth.. Quizball.. That's about it.." - Argument >lost - Amigas and STs won. > >> >I apologise if I seem agro.... >> >> Accepted. > >Thanks... > Your'e welcome. > Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 15:12:02 1996 Date: Fri, 8 Nov 1996 10:03:54 -0500 From: BrenchleyR@aol.com Message-Id: <961108100353_1417728440@emout14.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: SOS the whole thing... Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 208 Lines: 11 In a message dated 08/11/96 04:07:02, you write: >Is it time to start deciding what parts we want and what don't (on an entry >level machine at least!)? > Numb. > > Yes, time is slipping by. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 15:12:03 1996 Date: Fri, 8 Nov 1996 10:03:53 -0500 From: BrenchleyR@aol.com Message-Id: <961108100351_1182288653@emout10.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: SamSon Interrupts Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 268 Lines: 10 In a message dated 08/11/96 04:06:52, you write: >So am I... how about using the COMMS interface's timer system? Should work >perfectly... (apart from when Frame ints mis-trigger it...) > >Simon Could you explain what you mean by the COMMS interface's timer? Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 15:12:03 1996 Date: Fri, 8 Nov 1996 10:04:02 -0500 From: BrenchleyR@aol.com Message-Id: <961108100358_1781650445@emout04.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Re : Killer apps - Internet Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 598 Lines: 16 In a message dated 08/11/96 04:08:56, you write: >Fidonet is good, because you like it - it does things the way you like them >to be done. >Other nets all have their good points. >But the sad fact is - what people think of as the internet (in fact they >are only really looking at the WWW) is what they /think/ they need. Once >they have a machine, you are welcome to try and educate them, but it will >be the magic word INTERNET that gets them. > >Samsboss. Am I going daft? Or is this guy/gal starting to make sence? No, it must be my brains going soft from too many email messages. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 15:12:36 1996 Date: Fri, 8 Nov 1996 10:03:54 -0500 From: BrenchleyR@aol.com Message-Id: <961108100354_1450530189@emout15.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Proposal: Workpackage Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 453 Lines: 16 In a message dated 08/11/96 04:07:03, you write: >I think the problem with Windows 95 is as follows... On a decent >GUI, everything should be Icon based with a little text as a backup/a >little info which you may need sometimes. Win' '95 uses massive >great columns of text under the 'START' button, with the icons used >secondarily/as a backup method. > >-- >James R Curry hc95jc@dmu.ac.uk > > Could not have put it better myself. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 15:13:10 1996 Date: Fri, 8 Nov 1996 10:03:55 -0500 From: BrenchleyR@aol.com Message-Id: <961108100355_1584117261@emout19.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Tv or not TV Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 341 Lines: 16 In a message dated 08/11/96 04:08:29, you write: >I reckon one of the selling points could (poss?) be: You don't need a >monitor if you don't want one! (Though more catchy) > >Comments? > > >Al. Would it be acceptable to have lower screen res modes (1 to 4 say) goint to a TV, while newer hi-res modes need a monitor? Just asking. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 15:13:12 1996 Date: Fri, 8 Nov 1996 10:03:57 -0500 From: BrenchleyR@aol.com Message-Id: <961108100357_1714541581@emout02.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Daddy! Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 542 Lines: 17 In a message dated 08/11/96 04:08:53, you write: >Yeah, it took ages to get the spec right. We can use a lot of the hardware >from the first model, but if it is a boy this time we're gonna need new >software. I think a bundel will need to be prepared before the launch. Or we >could issue that when the time comes.. >We can't decide on any names yet though.. > DaveSon, SAM, SAMantha.... :)) > > > _ >|_)ave Yes, but do we get a say in which processor it uses and how much memory? Bob. (Who loves kid, until they start to answer back). From owner-sam-users@nvg.unit.no Fri Nov 8 15:13:12 1996 Date: Fri, 8 Nov 1996 10:03:58 -0500 From: BrenchleyR@aol.com Message-Id: <961108100357_1748096141@emout03.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Problems, problems... Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 303 Lines: 16 In a message dated 08/11/96 04:08:54, you write: >Yeah: SAM Colour - purple >It's got to be! Yes, agreed, my sister would then buy one. But what shade of purple? > >And *who* could Samsboss be? Getting close, all these hints are coming together. (s)he will go one step to far one day. > >Al. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 15:13:18 1996 Date: Fri, 8 Nov 1996 15:02:43 +0000 (GMT) From: Justin Skists X-Sender: c93js1@galois.cms.dmu.ac.uk To: sam-users@nvg.unit.no Subject: Re: Returned mail: Remote protocol error In-Reply-To: <199611071721.RFE29731@harrier.fen.bris.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2120 Lines: 56 On Thu, 7 Nov 1996, Mail Delivery Subsystem wrote: [snipped some gunk about mailing errors] > I don't think writing the opperating system in C is a good idea! True, it may be > easier and faster to write but it is also more likely to be seriously bugged > and will run slower than a pure assembler one. Maybe true.. Depends how good the compiler is. Then again, looking at my past assembler aberations, they'll be no different. (And James Curry was wondering why I won't write any games...) > All of the desktop accessorys you get with the machine had better be done in > Z380 assembler. > > At first the speed of the utils may not seem that important, but when start to > multitask a few, the inefficiencies will add up. True - Especially the ones that would be used all the time such as screen clock and editors (ReblexWord won't be much use here). > Without the pageing we need to start thinking about another format for the > opperation of the desktop accessories. Where will they live and store their > variables? How will they multitask? (See me other post about multitasking) > I feared this. With a more powerfull machine will we become lazy and start > to write Sam games in C? I know what you are saying. We'll soon need a SAMgreatgrandson with the power of a PentiumPro. But, to get decent performance, you would write the framework in C and do all the mission critical bits in in-line assembler. And get an equivelent application/game with equivelent performance in appropriate places completed in half the time. Yes? > > -THIS FEAR WILL NOT BE AS IMPORTANT IF THE COMPILER WE GET WITH THE > MACHINE IS GREAT AT OPTIMISATION. AGREED. > C scares me! C's not too bad. Better than BASIC anyday! -- ============================================================================= |Justin Skists (c93js1@dmu.ac.uk) | Don't take life so seriously, | |BSc (Hons) Computer Science, Year 4 | you'll never survive it! | |De Montfort University, Leicester, England | - (C) CamLoRTS 1996 | ============================================================================= From owner-sam-users@nvg.unit.no Fri Nov 8 15:13:18 1996 Date: Fri, 8 Nov 1996 10:03:56 -0500 From: BrenchleyR@aol.com Message-Id: <961108100356_1647432589@emout20.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Clock Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 276 Lines: 11 In a message dated 08/11/96 04:08:31, you write: >Any news in advance Bob? >Could do with one to go with my hard disk interface. > >Samsboss. As I said, news in FORMAT, I have to make my money somehow you know... Ok, once FORMAT goes out I will let the list know. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 15:18:49 1996 Message-Id: <1.5.4.32.19961108151246.0093b300@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 15:12:46 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: Plea! Plea! Plea! Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 545 Lines: 22 At 05:36 PM 11/7/96 GMT, you wrote: >Can't wait for the accelerator to be finished? Then use this >wonderful program to create similar screen effects. > >10 CLS#:MODE 4 >20 CSIZE 8,8 >30 PLOT INT(RND*256),INT(RND*176) >40 GOTO 20 > >Okay, so it's not as fast as the accelerator.. But then if it was it >would lose sales. ;) > >(If anyone doesn't get this, I'm refering to the immense amounts of >white 'fuzz' the accelerator was chucking onto the screen at the >gloucester show which Simon brought it to.) *laughs* Excellent!!!! Simon From owner-sam-users@nvg.unit.no Fri Nov 8 15:18:54 1996 Message-Id: <1.5.4.32.19961108151248.00936c44@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 15:12:48 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: Keyboards.. Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 409 Lines: 13 At 06:20 PM 11/7/96 GMT, you wrote: >>All this talk of keyboards.. why not choose a PC one with those >>'Windows' keys on. When you press one the machine could display "You >>must be joking!" and crash. >Bloody good idea that one. Now how can we program all the PCs out there to >do exactly the same thing? Or rather, how can we program them all to *STOP* doing exactly the same thing... ;) Simon From owner-sam-users@nvg.unit.no Fri Nov 8 15:19:02 1996 Message-Id: <1.5.4.32.19961108151249.0094d684@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 15:12:49 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: Fred 74 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 327 Lines: 13 At 11:48 AM 11/6/96 +0100, you wrote: >Yeah, I can't get RGB Demo to work either. > >Luckily, however, being less than 25 mins walk from Anderton's house >as I am... In that case, get the bugger to explain what's wrong and how to fix it! And post a fix up on NVG!!! (Or even, just put a copy of the demo up there...) Simon From owner-sam-users@nvg.unit.no Fri Nov 8 15:19:20 1996 Message-Id: <1.5.4.32.19961108151251.009428b0@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 15:12:51 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: Message understood. Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 497 Lines: 19 At 11:49 PM 11/7/96 GMT, you wrote: >>Well, there's no point having 96pins if you've only >>got 64 signals to put down them! It's just as well he >>didn't anyway, because it leaves the middel row free >>for us to use... >-- > >Very good point. > >So, what are we goint to use the other 32 for? That is the next question. Z380 control signals, probably... seems like a good idea to me... Or other control signals of course... It all depends on the architecture we end up using. Simon From owner-sam-users@nvg.unit.no Fri Nov 8 15:19:32 1996 Message-Id: <1.5.4.32.19961108151245.00936d34@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 15:12:45 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: Proposal: Workpackage Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 470 Lines: 13 At 05:17 PM 11/7/96 +0000, you wrote: >I doubt it would be worth the effort. After all, we only have 128 colours >in the palette so the image is going to look rough no matter how many >palette changes you put in it. > >Also I know from experience that if you FS-dither the image then suddenly >changing one colour in the palette can make a visible line in the image. Hmmmm.... how about a Bayer dither? Or perhaps we can come up with another algorithm for it? Simon From owner-sam-users@nvg.unit.no Fri Nov 8 15:19:54 1996 From: Gavin Smith Organization: University of Ulster To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 15:14:51 GMT Subject: Re: SAMson marketing X-Confirm-Reading-To: "Gavin Smith" X-Pmrqc: 1 Priority: normal X-Mailer: Pegasus Mail for Windows (v2.23) Message-Id: <2E8A341EC7@smserver1.ulst.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 652 Lines: 16 >>So, in my opinion, ANSI C for SAM should be a higher priority than >>SOS... :) >We at least have a C at the moment, it may be SMALL C but it is better >than nothing. And as C will only appeal to a very small minority, I do >think there are other more important things at the moment. Are you joking?! C (or rather C++ nowadays) is a kind of standard language! Most employers demand it, most universities teach it. I think C++ should be almost as high on our priority list as SOS! By the way, someone else said we should get things going and development of SOS started. Bob, can you make a list of something of what you want done? Gavin Smith From owner-sam-users@nvg.unit.no Fri Nov 8 15:20:14 1996 Date: Fri, 8 Nov 1996 15:15:20 GMT From: Keith Turner Message-Id: <9611081515.AA12594@turner.cursci.co.uk> To: sam-users@nvg.unit.no Subject: Re: Multitasking WIMPs - comparison to Unix Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 3212 Lines: 65 >> Ian Collier said: >>> Also, it had better be pre-emptively multitasked. None of this Win 3.0 >>> "one bad app crashes the whole system" business... > Keith Turner said: >> You need an MMU to avoid the Win 3.0 "one bad app crashes the whole system" >> business. > Ian replied: > Well I believe I already asked for an MMU. :-) > > But actually I was just talking about infinite loops. > Keith went on to say: >> The scheduling of task-switching can be either pre-emptive or cooperative. >> Unix is capable of pre-emptive multi-tasking, but that feature might be >> a bit overblown for Samson. Ian queried: > Why? As far as I can tell, pre-emptive multitasking is no more difficult > than cooperative multitasking (and you don't have to provide a "give up > control" function and insist the programmer uses it). In practice I think it has been shown that cooperative multi-tasking is easier to implement on a machine like the Sam, where most programs access the hardware directly. I'll try to illustrate why this is. The following is likely to be only of interest to OS/GUI programmers. Under Unix, all hardware access by applications is guaranteed to be through OS routines. That is why the multitasking works so well under Unix. Those OS system services are also like cooperative multi-tasking routines in some respects. When a Unix app asks for some disk data, Unix lets some other apps have the CPU until the disk data is in memory. Then when those other apps have asked for OS services, the original app is given it's data. This gives the appearance of pre-emptive multi-tasking. Unix only has to force a task change when an app is working entirely within the CPU. Unix is not particularly good at that. All of this works well when the machine has hardware DMA or buffering and the machine does not allow direct access to hardware or out-of-bounds memory. For a machine like the Sam (and the early incarnations of Samson) pre-emptive multi-tasking will be crippled by the locks applications will need to honour to control exclusive access to hardware. Am I right in saying that GEM uses cooperative multi-tasking on the Atari ST and the plans for multi-tasking in DRiVER were going the same way as GEM? If we go down the pre-emptive multitasking route I think we will also have to restrict applications to using OS routines for all access to screen memory, sound chips, system variables, disk drives, etc. If we are that far removed from the design of the Sam, Simon and the "demo-coders" will have no opportunity to shave T-states off their sprite routines and invent cunning disk protection, and the machine would hardly be a Sam anymore, just a slow version of time-sharing machines available fifteen years ago. On the other hand, cooperative multi-tasking allows all the same creativity when accessing hardware, doesn't require an MMU and DMA or big hardware buffers but puts a little more responsibility on the programmer to be fair to other applications. In essence, I feel that the Sam would be no fun for programmers if they had to program in such a way as to support pre-emptive multitasking. And the one unique aspect of the Sam is that it is fun for programmers. / X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 15:17:58 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: SamSon Interrupts Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 941 Lines: 23 At 10:03 AM 11/8/96 -0500, you wrote: >>So am I... how about using the COMMS interface's timer system? Should work >>perfectly... (apart from when Frame ints mis-trigger it...) >> >>Simon > >Could you explain what you mean by the COMMS interface's timer? Right... The comms interface can generate interrupts (in the SAM Technical manual, the interrupt it generates is falsely labelled the Mouse interrupt) -- provided you put a link over a jumper in the design (some interfaces haven't got these jumpers! But they do have through holes so you can solder your own one in). The Comms interface chip has a built in counter/timer circuit, which can be told to generate an interrupt every so often. The counter will count down and stop, whereas the timer will keep repeatedly interrupting every so often (or is it the other way round?). But it's reasonably easy to use... The only problem is that not everyone has a comms interface... Simon From owner-sam-users@nvg.unit.no Fri Nov 8 15:22:44 1996 Message-Id: <1.5.4.32.19961108151859.0094e940@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 15:18:59 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: Tv or not TV (Nev read this) Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 443 Lines: 14 At 09:56 AM 11/7/96 +0100, you wrote: >How about an option on startup, like the QL has? Or, if we're going to have >the GUI/CLI startup thing, how about doing the same for MONITOR/TV? > >In early versions, perhaps this could be a manual switch while we get >the HD sorted out? > >Al. It should be possible for this to all be automatic... At least I think so. Of course, if an LCD screen were to be used, it might get a bit messier... Simon From owner-sam-users@nvg.unit.no Fri Nov 8 15:28:20 1996 Date: Fri, 8 Nov 1996 16:26:29 +0100 From: ft@edh.ericsson.se (Frode Tenneboe) Message-Id: <9611081526.AA11544@asmal.edh-net> To: sam-users@nvg.unit.no Subject: Re: SAMson marketing X-Sun-Charset: US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1207 Lines: 25 > When Nev got high super-expensive PC based Ansi C compiler with Z80 source > code output (what a mouthful) Him and I spent ages hunting hi and low for C > source that could be ported to SAM. Result? Nothing. So where is all this C > source that we can desend on an get onto SAM? If anyone can supply something, > Nev can look at it for us. You don't need that. What you need is the source of a compiler that can produce Z80 - noting else is really worth anything. lcc is a compiler which with the right grammar would produce (it appears) nice Z80 code. However, the documentation is in a not-easily obtainable book... As long as you have the right grammar, you can use the compiler to compile itself with. But first, you have to decide on 1) the memory structure and 2) the object file standard. > We at least have a C at the moment, it may be SMALL C but it is better than > nothing. And as C will only appeal to a very small minority, I do think there > are other more important things at the moment. There exists about three zilion ready-to-comile-with-ansi-compiler programs which can be portet quite easily, amongst other things are algol-compilers, pascal-compilers, basic-compilers ;) -Frode From owner-sam-users@nvg.unit.no Fri Nov 8 15:33:12 1996 Date: Fri, 8 Nov 1996 16:30:39 +0100 From: ft@edh.ericsson.se (Frode Tenneboe) Message-Id: <9611081530.AA11547@asmal.edh-net> To: sam-users@nvg.unit.no Subject: Re: SAMson marketing X-Sun-Charset: US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 284 Lines: 7 > Are you joking?! C (or rather C++ nowadays) is a kind of standard > language! Most employers demand it, most universities teach it. I > think C++ should be almost as high on our priority list as SOS! As my former boss said: "If it's written in C++, nobody want's it" :) -Frode From owner-sam-users@nvg.unit.no Fri Nov 8 15:33:34 1996 From: Johnna Teare Organization: University of Central Lancashire To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 15:25:38 GMT+0 Subject: Re: Re : Killer apps - Internet Priority: normal X-Mailer: Pegasus Mail for Windows (v2.33) Message-Id: <138FA86911@mail-gw.uclan.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 333 Lines: 14 Hello, > Am I going daft? Or is this guy/gal starting to make sence? No, but I think he's starting to make some sense :). I take it you spellcheck FORMAT before you publish it! > No, it must be my brains going soft from too many email messages. > > Bob. > Johnna Pig Teare (j.d.teare@uclan.ac.uk) "They call me Mad The Swine" From owner-sam-users@nvg.unit.no Fri Nov 8 15:51:25 1996 Date: Fri, 8 Nov 1996 15:48:36 +0000 (GMT) From: Justin Skists X-Sender: c93js1@jasmine.cms.dmu.ac.uk To: sam-users@nvg.unit.no Subject: Re: SOS: Multitasking In-Reply-To: <1.5.4.32.19961108144948.00932f54@nessie.mcc.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1125 Lines: 24 On Fri, 8 Nov 1996, Simon Cooke wrote: > At 02:34 PM 11/8/96 +0000, you wrote: > >As far as I know, the processor needs to have the ability to take away control > >from one program and give it to another. I'm pretty sure the Z80B doesn't have > >this feature. Does the Z380? > > > >OK, I know that you could use interrupts, but couldn't a program issue a DI and > >keep the processor to itself? I suppose you could design a "kernel" that runs > >p-code type instructions (like Visual Basic and Java) but that would be pretty > >slow compared to assembler. > > Just use the NMI :) - we may as well! How will the Kernel issue a NMI when it won't have control? Or are we expected to press the Break key every 50th of a second? :) -- ============================================================================= |Justin Skists (c93js1@dmu.ac.uk) | Don't take life so seriously, | |BSc (Hons) Computer Science, Year 4 | you'll never survive it! | |De Montfort University, Leicester, England | - (C) CamLoRTS 1996 | ============================================================================= From owner-sam-users@nvg.unit.no Fri Nov 8 16:00:58 1996 Date: Fri, 8 Nov 1996 15:56:56 +0000 (GMT) From: Justin Skists X-Sender: c93js1@jasmine.cms.dmu.ac.uk To: sam-users@nvg.unit.no Subject: Re: SAMson marketing In-Reply-To: <961108100320_1613878155@emout08.mail.aol.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2036 Lines: 52 On Fri, 8 Nov 1996 BrenchleyR@aol.com wrote: > See your point, but... Games consols are different, you can't do a thing with > them without some game to plug in. In the computer market it is a little > different. But lets get the hardware underway first, when the launch is > getting neaerr we can worry about the software. You won't get any more argument from me.. :) (I was having a "go against everybody" past few days..) > When Nev got high super-expensive PC based Ansi C compiler with Z80 source > code output (what a mouthful) Him and I spent ages hunting hi and low for C > source that could be ported to SAM. Result? Nothing. So where is all this C > source that we can desend on an get onto SAM? If anyone can supply something, > Nev can look at it for us. What can it cope with? I may have a few tiny C programs floating around that I wrote for GCC (works with Unix and DOS). > We at least have a C at the moment, it may be SMALL C but it is better than > nothing. And as C will only appeal to a very small minority, I do think there > are other more important things at the moment. But can it cope with 32 bit integer and floating point arithmatic? :) I really should look at my own tools before I start demanding stuff. I've got a CP/M Small C on ProDOS which, I think, outputs Z80 assembler. I *SHOULD* be able to adapt the output for my own devious ideas.. :) > >As for the map: Could I have it on a two week approval? :) > > No. Cash up front. Shucks. > Your'e welcome. :) PS. No more "We want ANSI C on SAM" arguments from me. I've ran out of arguments anyway! PPS. Let's all go and get our hands dirty! :) -- ============================================================================= |Justin Skists (c93js1@dmu.ac.uk) | Don't take life so seriously, | |BSc (Hons) Computer Science, Year 4 | you'll never survive it! | |De Montfort University, Leicester, England | - (C) CamLoRTS 1996 | ============================================================================= From owner-sam-users@nvg.unit.no Fri Nov 8 16:10:48 1996 Date: Fri, 8 Nov 1996 16:08:25 +0000 (GMT) From: Justin Skists X-Sender: c93js1@jasmine.cms.dmu.ac.uk To: sam-users@nvg.unit.no Cc: sam-users@nvg.unit.no Subject: Re: Tv or not TV In-Reply-To: <961108100355_1584117261@emout19.mail.aol.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 902 Lines: 19 On Fri, 8 Nov 1996 BrenchleyR@aol.com wrote: > Would it be acceptable to have lower screen res modes (1 to 4 say) goint to a > TV, while newer hi-res modes need a monitor? But that would make the programmer write, basically, two versions of the application: the nice one that will use Mode 5 (or whatever) and one to work on Mode 3/4 for the TV.... We need to let the user go slowly up the upgrade path. No point saying that SAMson will work on a TV if none of the new applications will run on the TV... -- ============================================================================= |Justin Skists (c93js1@dmu.ac.uk) | Don't take life so seriously, | |BSc (Hons) Computer Science, Year 4 | you'll never survive it! | |De Montfort University, Leicester, England | - (C) CamLoRTS 1996 | ============================================================================= From owner-sam-users@nvg.unit.no Fri Nov 8 16:11:05 1996 Message-Id: <199611081609.RAA03180@dxmint.cern.ch> From: Allan Skillman Subject: Re: SAMson marketing To: sam-users@nvg.unit.no Date: Fri, 8 Nov 96 17:09:05 MET In-Reply-To: <9611081526.AA11544@asmal.edh-net>; from "Frode Tenneboe" at Nov 8, 96 4:26 pm Mailer: Elm [revision: 70.85] Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 3153 Lines: 60 > > You don't need that. What you need is the source of a compiler that can > produce Z80 - noting else is really worth anything. lcc is a compiler > which with the right grammar would produce (it appears) nice Z80 code. > However, the documentation is in a not-easily obtainable book... > As long as you have the right grammar, you can use the compiler to compile > itself with. > > But first, you have to decide on 1) the memory structure and 2) > the object file standard. And - if we go down the standard compiler route, a linker and library format. All we get out of the compiler is assembly code. We will also need a good assembler capable of assembling multiple source files, external references and macros. Oh yes we also need the the C library writing. > > > We at least have a C at the moment, it may be SMALL C but it is better than > > nothing. And as C will only appeal to a very small minority, I do think there > > are other more important things at the moment. C is probably the most used programming language in the world today. C++ is nice, but just one of the many OO languages, much too 'high level' for the average C programmer (hacker). When it comes down to it C is really just a nice macro assembler BTW I've been searching for a C compiler with a z80 backend to no avail. I've contacted the writers of c68 (the oen teh QL guys use), bcc, lcc and the guy who did the z80 version of small C (more about that in a mo). Noone of these guys knows of a public version of Ansi C for Z80. lcc is the easiest to build a backend too, although there may be a problem with the way it uses the stack, as z80 has no equivalent to the indexed stack segement access via the BP register (ala x86 MOV AX,[BP+DI]) - lcc uses this for a frame pointer. However there is a rumoured version of lcc with a 6502 backend - so an 8 bit version can be done. The only C compiler available is from Hitech, they sent me some details. They produce cross compilers for DOS and Unix which compiler Ansi C to Z80 assembly. If we had some money to offer they might help us - I think they are quite a small firm. As I said, I now have the source for teh Z80 version of small C - the compiler I assume was the basis of SAM C. The package contains an assembler and linker (but not library linking). I can certainly put together a SAM runtime file and use the package to cross compile Sam software. I was thinking about using it to put together a Driver Application builder. Allan +--------------------------------------------------------------------------+ | Now available! XCoupe, the one and only SAM Coupe emulator | | *** http://www.hep.ucl.ac.uk/~ajs/xcoupe *** | +------------------------------+-------------------------------------------+ | Allan Skillman | "There are five flavours of resons, the | | HEPP Group | elementary particles of magic : up, down, | | University College London | sideways, sex-appeal and peppermint." | | Email : ajs@hep.ucl.ac.uk | - Terry Pratchett (Lords and Ladies) | +------------------------------+-------------------------------------------+ From owner-sam-users@nvg.unit.no Fri Nov 8 16:26:09 1996 From: sh5655@harrier.fen.bris.ac.uk (SL Harding) Message-Id: <199611081618.QAA09925@reed.bris.ac.uk> Subject: Re: Tv or not TV To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 16:18:21 +0000 (GMT) In-Reply-To: from "Justin Skists" at Nov 8, 96 04:08:25 pm X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1046 Lines: 26 > > On Fri, 8 Nov 1996 BrenchleyR@aol.com wrote: > > > Would it be acceptable to have lower screen res modes (1 to 4 say) goint to a > > TV, while newer hi-res modes need a monitor? > I don't know if anyone recieved it but a couple of days ago I suggested that we send the full signal of the new modes to the monitor, but break up the resolution that is sent to the modulator so that it will fit on a TV. Vertically we can take advantage of the interlacing effect of normal TV and reconfigure the output so that it actually uses the in-between lines. In the horizontal direction we will just have to miss the odd pixel out. We could have a sub-mode whereby we alter the pixels that we miss out horizontaly but this feature should be turn-off-able to avoid horrible flickery displays on some TVs. It is important that all people can see a representation of all the initial modes. In the original post I drew parallels with the way ST owners developed software utils to do this. Did any one not recieve the original post? Numb. From owner-sam-users@nvg.unit.no Fri Nov 8 16:26:09 1996 From: sh5655@harrier.fen.bris.ac.uk (SL Harding) Message-Id: <199611081618.QAA09925@reed.bris.ac.uk> Subject: Re: Tv or not TV To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 16:18:21 +0000 (GMT) In-Reply-To: from "Justin Skists" at Nov 8, 96 04:08:25 pm X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1046 Lines: 26 > > On Fri, 8 Nov 1996 BrenchleyR@aol.com wrote: > > > Would it be acceptable to have lower screen res modes (1 to 4 say) goint to a > > TV, while newer hi-res modes need a monitor? > I don't know if anyone recieved it but a couple of days ago I suggested that we send the full signal of the new modes to the monitor, but break up the resolution that is sent to the modulator so that it will fit on a TV. Vertically we can take advantage of the interlacing effect of normal TV and reconfigure the output so that it actually uses the in-between lines. In the horizontal direction we will just have to miss the odd pixel out. We could have a sub-mode whereby we alter the pixels that we miss out horizontaly but this feature should be turn-off-able to avoid horrible flickery displays on some TVs. It is important that all people can see a representation of all the initial modes. In the original post I drew parallels with the way ST owners developed software utils to do this. Did any one not recieve the original post? Numb. From owner-sam-users@nvg.unit.no Fri Nov 8 16:31:57 1996 Message-Id: <32835F20.151C@warwick.ac.uk> Date: Fri, 08 Nov 1996 16:26:08 +0000 From: Paul Walker X-Mailer: Mozilla 3.0 (X11; I; SunOS 5.5.1 sun4m) Mime-Version: 1.0 To: sam-users@nvg.unit.no Subject: Help! Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 250 Lines: 9 Apologies to all others, but... What was the reference for the funny files given out not long ago (by Dave Hooper?)? I did make a note of it, but I've lost it now and the files looked good! A quick reply would be good, if anyone's on line... Paul From owner-sam-users@nvg.unit.no Fri Nov 8 16:31:57 1996 Date: Fri, 8 Nov 1996 16:27:32 +0000 (GMT) From: "A.S. Collier" X-Sender: asc25@indigo.csi.cam.ac.uk To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1652 Lines: 47 On Fri, 8 Nov 1996, Andrew M Gale wrote: > * Existing SAM programs could be improved by running them on > the z80 and getting the z380 to play samples as appropriate > - I'd guess it'd only need a few minor alterations to the code > of most SAM programs. Hmmm. > I'm sure there are plenty more ideas you could think of... Yes, but ideas tend to come only with practice. So just as soon as somebody gets the circuit designed and built, I'd really like to have a look at what I can do with it... There have been somewhat over 200 messages since I asked: Is it worth continuing with Z80B WinDOS, when I may soon be starting Z380 WinDOS? Nobody's replied yet... > ps: do you think the extra processor really needs to be a z380? > 68000s are cheap and have an obvious upgrade path, the 80386 is > really cheap, and again has an obvious upgrade. But the z380 code > will be familiar to us all. (Also, a 68000 based one will be seen > as a poor Amiga or ST) Absolutlely, Z380 all the way. And not just for sentimental reasons either... Question to anybody who knows a bit more about the Z380 that I do: For a Windows-system application, of which there may be many in memory at one time; will code need to be 100% relocatable? Or is there some system where an address, when offset by (say) 64K, will look exactly the same to the processor? I heard something about limiting values to 16 bits instead of 32. IE. Will it be possible to do self-modifying 'Windowsish' applications? (That is one advantage of the paging system, every program can think that it's the only one in memory) Andrew PS What would I use as an assembler for the Z380? From owner-sam-users@nvg.unit.no Fri Nov 8 16:43:36 1996 Message-Id: <1.5.4.32.19961108163939.00936f80@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 16:39:39 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: SOS: Multitasking Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 339 Lines: 11 At 03:48 PM 11/8/96 +0000, you wrote: >> Just use the NMI :) - we may as well! > >How will the Kernel issue a NMI when it won't have control? Or are we expected >to press the Break key every 50th of a second? :) Quite easy - if we're designing an accelerator board, what's to stop us wiring a programmable timer to the NMI line? Simon From owner-sam-users@nvg.unit.no Fri Nov 8 16:51:11 1996 Date: Fri, 8 Nov 1996 11:43:02 -0500 From: BrenchleyR@aol.com Message-Id: <961108114302_1115195966@emout18.mail.aol.com> To: sam-users@nvg.unit.no Subject: A few answers. Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2475 Lines: 53 In a message dated 08/11/96 10:57:38, you write: >On Thu, 7 Nov 1996 BrenchleyR@aol.com wrote: >[cut] >> In other words, at first you will not be able to display a screen direct >from >> the Z380, but you can dump a screen back into the Z80 for display any time >> you want to look at it. Bo used a similar system to in the early stage of >> FLASH, the programming was done on a Spectrum, run on a 'half way SAM' and >> then the screen was read back into the spectrum for examination. > >Well, yes I suppose that is one way to do it. I'd be disappointed if - >given this super-duper processor - we still can't scroll the Sam's screen >in one frame because we're relying on the Z80B to do the work.... However >I do agree your method does have some advantages, particularly in terms of >reading from disk; I can see how sharing the processor load would make >some games work quite well - Z80B drawing things up on the screen while >Z380 works actually controlling the sprites and collision detection (now >THAT, I think, is the hardest part of writing a game!) No, what I was trying to say was this. To allow the Z830 programmers to get a head start, the first version of the board will only communicate with the internal SAM processor (the Z80B) via a comms-block. Programmers can set up test screen withing the Z380's memory and run 'imaginary screens' to be going on with. These screen can be copied back into the Z80's space and either displayed in a lower-res form on the SAM mode 4 screen, or printed out in some way that the programmer can understand what is being shown. I think the internal SAM chippery should handle modes 1 to 4 and display them using the existing ASIC. New modes (to be agreed later) will be handled by the new graphics card. which the Z380 will be able to write to/read from, as soon as we get it built. > >[Cut] > > >Couple more points: How do we actually get our programs into the memory of >the new processor and start them up? I presume the Z80B will output to >some port, which starts up the Z380. Could the communications block be >located at the start of Z380's memory, and an almost-boot routine be >copied there by the Z80B? The Z380 starts running that routine, and copies >more of the relavent code into its own memory... Would that mean we don't >(yet) need a ROM-style block? Perhaps that would do for the developers' >run, before any system code has actually been written. That still need a bit of thought. > > > >Andrew Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 16:51:11 1996 Date: Fri, 8 Nov 1996 11:43:04 -0500 From: BrenchleyR@aol.com Message-Id: <961108114303_1182299966@emout20.mail.aol.com> To: sam-users@nvg.unit.no Subject: Serial Interface Questions. Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 593 Lines: 17 I have been going back over some of the recent messages and there has been several referances to RS232/Serial i/o, and this has led me to the following short list of questions. 1) What is considered wrong with the current COMMS Interface? 2) If speed is a problem, can the existing chip(s) go faster? 3) If we have to change to another design, what chip should it be based on? 4) Again, if we change, should the new interface have one or two RS232 ports? Look forward to your answers. And if you just use the numbering system instead of quoting each time - we will all save space. Bob, From owner-sam-users@nvg.unit.no Fri Nov 8 16:51:57 1996 Message-Id: <1.5.4.32.19961108164210.0093ea58@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 16:42:10 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: Plea! Plea! Plea! + Bob's message Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1437 Lines: 36 At 04:27 PM 11/8/96 +0000, you wrote: >There have been somewhat over 200 messages since I asked: Is it worth >continuing with Z80B WinDOS, when I may soon be starting Z380 WinDOS? >Nobody's replied yet... Yep -- go for it. It won't be *that* hard to convert to Z380 versions later. I promise :) >> ps: do you think the extra processor really needs to be a z380? >> 68000s are cheap and have an obvious upgrade path, the 80386 is >> really cheap, and again has an obvious upgrade. But the z380 code >> will be familiar to us all. (Also, a 68000 based one will be seen >> as a poor Amiga or ST) > >Absolutlely, Z380 all the way. And not just for sentimental reasons >either... > >Question to anybody who knows a bit more about the Z380 that I do: For a >Windows-system application, of which there may be many in memory at one >time; will code need to be 100% relocatable? Or is there some system where >an address, when offset by (say) 64K, will look exactly the same to the >processor? I heard something about limiting values to 16 bits instead of >32. > >IE. Will it be possible to do self-modifying 'Windowsish' applications? >(That is one advantage of the paging system, every program can think that >it's the only one in memory) Ummm... I don't think so... I think the code will need to be 100% relocatable... >PS What would I use as an assembler for the Z380? COMET would do for now... Zilog sell them too, AFAIK. Simon From owner-sam-users@nvg.unit.no Fri Nov 8 17:11:36 1996 Message-Id: <1.5.4.32.19961108170524.0093c880@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 17:05:24 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: Serial Interface Questions. Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 770 Lines: 24 At 11:43 AM 11/8/96 -0500, you wrote: >1) What is considered wrong with the current COMMS Interface? Interrupts aren't on as standard, the voltages are slightly screwy, causing chip errors, the DTR line is wired up rather than the RTS line (I pin swap them on people's boxes if they send them to me), and there's not enough signals (CD, DTR, DSR would be handy... RI would be a bonus). > >2) If speed is a problem, can the existing chip(s) go faster? Nope - but it can go at 38400 -- above which you start hitting the limitations of the SAM's speed. >3) If we have to change to another design, what chip should it be based on? IM26C92. >4) Again, if we change, should the new interface have one or two RS232 ports? Have you got a postscript Printer, BOB? Simon From owner-sam-users@nvg.unit.no Fri Nov 8 17:25:56 1996 Message-Id: <1.5.4.32.19961108172202.0093da20@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 08 Nov 1996 17:22:02 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Comms Interface Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 279 Lines: 12 There is a comms interface design available at: ftp.nvg.unit.no, under: pub/sam-coupe/docs/gemini.zip (or gemini.ps) If you've got a postscript printer, load it up and print it out ;) BTW: If you want to do anything commercial with it, ask me first. It's only polite. Simon From owner-sam-users@nvg.unit.no Fri Nov 8 17:39:38 1996 Message-Id: From: ee31ag@ee.surrey.ac.uk (Andrew M Gale) Subject: Re: Plea! Plea! Plea! + Bob's message To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 17:34:34 +0000 (GMT) In-Reply-To: from "Tim Wells" at Nov 8, 96 01:17:15 pm X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 433 Lines: 11 > This might well work, but I don't think it'll improve any programs - > imagine that the z80 starts to draw a lemming perhaps, and then the z380 > grabs control of the bus to start sending the next chunk of its samples... > The Z80B's address bus will always be a bottleneck here, > No, the dac would have to be on the z380 side in this case, and the sample would be in the z380's memory, so there would be no contention. -Andy From owner-sam-users@nvg.unit.no Fri Nov 8 17:45:19 1996 X-Authentication-Warning: dcsun4.comp.brad.ac.uk: lwillis owned process doing -bs Date: Fri, 8 Nov 1996 17:41:43 +0000 (GMT) From: Lee To: SAM users list Subject: Re: SOS: Multitasking In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 825 Lines: 23 On Fri, 8 Nov 1996, Justin Skists wrote: ;>How is it possible to do pre-emptive multitasking on the Z380? ;> ;>As far as I know, the processor needs to have the ability to take away control ;>from one program and give it to another. I'm pretty sure the Z80B doesn't have ;>this feature. Does the Z380? ;> No, multi-tasking in this sense is handled by the OS, _NOT_ the processor. We're talking about whether the GUI we're (maybe) going to have running several windows with different programs running in each (Correct me if I've misunderstood!). The programs are written to a specified style so that the GUI can control them. The processor has nought to do with it! Lee. How many Microsoft employees does it take to screw in a light bulb? None, they just make darkness the industry standard! From owner-sam-users@nvg.unit.no Fri Nov 8 17:48:15 1996 Message-Id: From: ee31ag@ee.surrey.ac.uk (Andrew M Gale) Subject: Re: SOS: Multitasking To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 17:43:35 +0000 (GMT) In-Reply-To: from "Justin Skists" at Nov 8, 96 03:48:36 pm X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 580 Lines: 13 > How will the Kernel issue a NMI when it won't have control? Or are we expected > to press the Break key every 50th of a second? :) > I don't know about you, but I'm not particularly desperate for multitasking. I'd like to have more than one program running, each in its own window, and then be able to flick between them, but them appearing to run simultaneously doesn't really appeal to me... in which case using the break key would be fine! Mind you, all of us who got used to Daley Thompson's Decathlon joystick-waggling could probably press break at least 50 times a sec! From owner-sam-users@nvg.unit.no Fri Nov 8 17:48:44 1996 Message-Id: From: ee31ag@ee.surrey.ac.uk (Andrew M Gale) Subject: Re: Tv or not TV To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 17:45:19 +0000 (GMT) In-Reply-To: from "Justin Skists" at Nov 8, 96 04:08:25 pm X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 257 Lines: 6 > But that would make the programmer write, basically, two versions of the > application: the nice one that will use Mode 5 (or whatever) and one to work Well, we'll just have to have a GUI with really nice graphic calls so that the upgrade is painless.. From owner-sam-users@nvg.unit.no Fri Nov 8 17:54:50 1996 Message-Id: From: ee31ag@ee.surrey.ac.uk (Andrew M Gale) Subject: Re: Plea! Plea! Plea! + Bob's message To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 17:49:01 +0000 (GMT) In-Reply-To: from "A.S. Collier" at Nov 8, 96 04:27:32 pm X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 358 Lines: 11 > PS What would I use as an assembler for the Z380? > I guess your first application would have to be an assembler for the SAM, then! Or we could sweet-talk Steve Nutting into writing one! (Incidentally, when we get going, it would probably be a good idea to get Steve in on the act since he does seem to be pretty reliable at churning things out). -Andy From owner-sam-users@nvg.unit.no Fri Nov 8 17:54:53 1996 Message-Id: From: ee31ag@ee.surrey.ac.uk (Andrew M Gale) Subject: Re: Serial Interface Questions. To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 17:52:26 +0000 (GMT) In-Reply-To: <961108114303_1182299966@emout20.mail.aol.com> from "BrenchleyR@aol.com" at Nov 8, 96 11:43:04 am X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 229 Lines: 8 > 4) Again, if we change, should the new interface have one or two RS232 ports? > Two would be nice because it wouldn't cost more and would mean that a PC mouse could be plugged in - much cheaper than a standard SAM one. -Andy From owner-sam-users@nvg.unit.no Fri Nov 8 18:13:55 1996 From: sh5655@harrier.fen.bris.ac.uk (SL Harding) Message-Id: <199611081805.SAA10037@reed.bris.ac.uk> Subject: Re: Plea! Plea! Plea! + Bob's message To: sam-users@nvg.unit.no Date: Fri, 8 Nov 1996 18:05:32 +0000 (GMT) In-Reply-To: from "Andrew M Gale" at Nov 8, 96 05:34:34 pm X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 941 Lines: 26 > > > This might well work, but I don't think it'll improve any programs - > > imagine that the z80 starts to draw a lemming perhaps, and then the z380 > > grabs control of the bus to start sending the next chunk of its samples... > > The Z80B's address bus will always be a bottleneck here, > > > > No, the dac would have to be on the z380 side in this case, > and the sample would be in the z380's memory, so there would be > no contention. > -Andy > > I thought we intended to put it right on the RISK graphics/sound board eventually. There will be no need for the Z80B to play any part in the graphics or sound of the final device (except perhaps for graphics in modes 1 to 4). Whoo! I have just been playing on a Silicon graphics O2! The GUI is built around Netscape! 1 handy thing though, as well as the scroll bars on the windows there is a scale bar so you can make the icons bigger or smaller to taste! C9 Numb. From owner-sam-users@nvg.unit.no Fri Nov 8 18:18:59 1996 Date: Fri, 8 Nov 1996 13:11:32 -0500 From: BrenchleyR@aol.com Message-Id: <961108131131_1847965388@emout17.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: SAMson marketing Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1166 Lines: 34 In a message dated 08/11/96 15:25:19, you write: > [cut] >You don't need that. What you need is the source of a compiler that can >produce Z80 - nothing else is really worth anything. lcc is a compiler >which with the right grammar would produce (it appears) nice Z80 code. >However, the documentation is in a not-easily obtainable book... >As long as you have the right grammar, you can use the compiler to compile >itself with. By 'not easily obtainable, does this mean you managed to get it after a big effort? Or are you still trying - in which case email me details an I will try a couple of my contacts. > >But first, you have to decide on 1) the memory structure and 2) >the object file standard. Memory structure for Z380 board looks like it will be flat. And by object file do you mean something to ge into a standard linker - which is what I would love to see done. > [cut] > >There exists about three zilion ready-to-compile-with-ansi-compiler >programs which can be portet quite easily, amongst other things are >algol-compilers, pascal-compilers, basic-compilers ;) Yes, but everyone says there is lots of source code - but where? > > -Frode Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 18:19:05 1996 Date: Fri, 8 Nov 1996 13:11:34 -0500 From: BrenchleyR@aol.com Message-Id: <961108131133_1915074124@emout19.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Tv or not TV Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 261 Lines: 11 In a message dated 08/11/96 16:10:37, you write: >We need to let the user go slowly up the upgrade path. No point saying that >SAMson will work on a TV if none of the new applications will run on the >TV... > >-- Good point, back to the drawing board. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 18:19:14 1996 Date: Fri, 8 Nov 1996 13:11:35 -0500 From: BrenchleyR@aol.com Message-Id: <961108131134_1948628556@emout20.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 362 Lines: 10 In a message dated 08/11/96 16:30:03, you write: >There have been somewhat over 200 messages since I asked: Is it worth >continuing with Z80B WinDOS, when I may soon be starting Z380 WinDOS? >Nobody's replied yet... Yes. Go on and finish. At least some experiance can be gained from using it even if lots more features get included in the Z380 version. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 18:19:29 1996 Date: Fri, 8 Nov 1996 13:11:26 -0500 From: BrenchleyR@aol.com Message-Id: <961108131125_1483476556@emout12.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: SAMson marketing Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 613 Lines: 17 In a message dated 08/11/96 15:18:19, you write: >By the way, someone else said we should get things going and >development of SOS started. Bob, can you make a list of something of >what you want done? > >Gavin Smith I think that what I am hoping will happen is people will start saying thins like "Oh I know haow to do that, have it done in a few days (weeks, or even months)." It would be very halpful if everyone could email in just a few lines say what they think they would be good at doing - so we can have an idea of what resource there is. In the meantime the spec is coming together slowly. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 18:19:38 1996 Date: Fri, 8 Nov 1996 13:11:31 -0500 From: BrenchleyR@aol.com Message-Id: <961108131130_1748119884@emout16.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Multitasking WIMPs - comparison to Unix Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 336 Lines: 14 In a message dated 08/11/96 15:18:56, you write: >In essence, I feel that the Sam would be no fun for programmers if they had >to program in such a way as to support pre-emptive multitasking. And the one >unique aspect of the Sam is that it is fun for programmers. > >/ > 110% agree, especially with the last sentence. Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 18:19:38 1996 Date: Fri, 8 Nov 1996 13:11:32 -0500 From: BrenchleyR@aol.com Message-Id: <961108131132_1881519692@emout18.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Re : Killer apps - Internet Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 341 Lines: 10 In a message dated 08/11/96 15:31:38, you write: >No, but I think he's starting to make some sense :). I take it you >spellcheck FORMAT before you publish it! No, that is Jenny's job. And anyway, there is a whole month to get each issue right, with the rate these emails fly these days, I can't be blamed for a fwe mitsakse cna I? Bob. From owner-sam-users@nvg.unit.no Fri Nov 8 23:59:43 1996 Date: Fri, 8 Nov 1996 23:57:48 +0000 (GMT) From: "A.S. Collier" X-Sender: asc25@indigo.csi.cam.ac.uk To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message In-Reply-To: <961108131134_1948628556@emout20.mail.aol.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 5475 Lines: 115 On Fri, 8 Nov 1996 BrenchleyR@aol.com wrote: > In a message dated 08/11/96 16:30:03, you write: > > >There have been somewhat over 200 messages since I asked: Is it worth > >continuing with Z80B WinDOS, when I may soon be starting Z380 WinDOS? > >Nobody's replied yet... > > Yes. Go on and finish. At least some experiance can be gained from using it > even if lots more features get included in the Z380 version. > > Bob. It isn't a matter of adding new features, it's a matter of completely redesigning the screen drivers, memory access (I'll probably have to rethink that anyway - I've only accounted for 512K (I have no ext. memory and didn't think I'd ever need to get it) so all that code flies straight out of the WinDO) and a whole load of other things. I'm telling you I can't just port it over from one machine to the next. But... at least I'll get some idea of how I can organize the programs, so it is worth a go. Don't expect me to be able to do much work on it until the Christmas vac though. When will the Z380 boards be ready? Guesses? I think the developers' version should be approximately the same as the final one - don't want to have to write everything twice. But we do want to be able to generate some sort of screen before the high-res-graphics boards are developed, so one of the discussed methods should do. Please somebody start working on the hardware! I'm not sure I would envy the task of writing the 'official' GUI operating system... with the wildly divergent ideas I've seen over the past few months, I doubt that it will be possible to keep everybody (or even, anybody) completely happy. But, here's a list of the general ideas which I hope to get into WinDOS: (At first) Mode 4, 8 standard colours, plus eight which depend on the currently selected window. Background pattern. Sound samples (probably disrupt working of applications on Z80B, but don't necessarily need to for the Z380). Event-driven control architecture. (At stage 2: Frame interrupt based, time-slice multitasking. All applications have windows on the one screen.) Several Window types, easiest one for programmer is: draw the contents of the window in some misc area of memory, and inform the OS when you want that copied to the actual screen. Yes, I know it introduces a software overhead but the advantages to the apps programmer are: No need to remove mouse pointer before altering contents of window, no need to bother whether your window is selected, covered up or whatever; no need to bother about what the user has done to the size or position of the window; to draw the contents slowly does not mean the screen changes slowly and there's no flickering, the mouse pointer doesn't have to be taken off screen for any long periods, so although the changes take slightly longer the user doesn't notice anyway. Plus there will be the usual text windows, call this address "xxx" when the window gets drawn, dialogue boxes etc etc. Moving, resizing and scroll bars are handled by the OS without bothering the application. IE WinDOS will hopefully shield the applications programmer from all the low level GUI operations, in exactly the way which DRiVER doesn't. Menu bar is at the top of screen (menus available depend on currently selected window). Click on menu name with mouse, and either hold down or reclick on chosen option. Also... CNTRL+first letter of menu name to bring up menu, and use cursor to select option. Also... Function keys as short-cut to specific options. I'd suggest that some convention is kept to, eg F1 Save as, F2 Save, F3 Load; F7 Copy, F8 Cut, F9 Paste Info bar at bottom of screen, gives infomation depending on whatever the mouse is pointing to. Eg Details of menu options, copyright messages, or something controlled by the application. (Inspired by Mac's bubble help, but much faster and less obtrusive; it first came about because I needed the background to be repeated in 16-pixel squares, and I had 8 pixels left over after the menu bar) Provision for a proper screen-saver, with adjustable inactivity time, plus count use of mouse as activity :) Ability to quick-start the screensaver by moving mouse pointer to bottom corner of screen. Copy text or graphic contents of any Window, to be printed out or used by another application. The original plan was to release the program in two stages, the first being the WIMP system but offering no DOS support, released as PD or similar. (Other programmers can start to write and test their apps at this stage, loading their code in place of whatever application is released with WinDOS). Second stage, (released probably as shareware at about 15UKP) = above + all file support. DOS will be integrated into the same code, and supports SamDOS, MSDOS, Hard Disk, and anything else which seems important. Choice of words/icons for files! NB A user-definable graphic for folders will be difficult on a Sam format disk (reason being the folder is not allocated any disk space, apart from the directory entry itself, so there is nowhere to store the icon) The Windowing code will be the same between the first two stages, so all the applications people have written should be immediately compatible. That plan, of course, I will have to rethink now that an extended Sam is on the horizon. As for the distribution arrangements, I think it should certainly be made possible for upgrades to be readily available, there will probably be a lot of minor changes. Andrew From owner-sam-users@nvg.unit.no Sat Nov 9 01:43:05 1996 From: SL Harding Message-Id: <199611090140.BAA04449@harrier.fen.bris.ac.uk> Subject: Re: Saviour of the universe... To: sam-users@nvg.unit.no Date: Sat, 9 Nov 1996 01:40:37 +0000 (GMT) In-Reply-To: <12896B28C8@mail-gw.uclan.ac.uk> from "Johnna Teare" at Nov 8, 96 01:42:40 pm X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 5419 Lines: 121 > > > Bo used a similar system to in the early stage of > > FLASH, the programming was done on a Spectrum, run on a 'half way SAM' and > > then the screen was read back into the spectrum for examination. > > And boy does it show...! > > > Bob. > Johnna Pig Teare (j.d.teare@uclan.ac.uk) > "They call me Mad The Swine" > The use of the PC SAM emulator for the development of software has already been suggested! Why worry about building a physical 'half way SAMSON'? I think we DO need to think more about the inner workings of the graphics card as well as the main Z380 chunk. We should give the graphics card some power! If we do give it a 33MHz RISK processor of its own then I need to know more about how powerfull this actually is when dealing with screens of around the 1-2 meg mark. I can't see how we can think about the visual appearance of the GUI for example until this has been specified. If this does happen it is clear that it would be very difficult for both the RISK and the Z380 to use the same memory at the same time. Here are a few possible arrangements: i,The Z380 uses the graphics processor as a slave system by sending it commands to do and waiting for them to be completed before it continues. Commands such as 'plot pixel at X,Y'. This would be powerfull as the slave processor could use the same command to plot the pixel at any resolution or colour depth, the way it actually goes about plotting the pixel depends on the current screen mode. This would solve the I-can't-run-any-applications-as-I-only-have-an-old-T.V problem. It could also be used to give nice scalling effects within a mode, and to ensure all work is kept within the relevent GUI window independently of the scale the window was designed to have. As the Z380 is disabled while the RISK chip is doing its stuff It should be relatively easy to keep all of the video memory in the same place as the main memory. i.e. saving the 'I have 2 spare meg of main memory but need more memory for the graphics mode I want to run' sort of problem PC people have. The disadvantage is obvious as the Z380 can do nothing while the RISK is at work. ii, As i, but once a job has been put on the RISKs command stream don't kill the Z380. Until the job is done but let them run in parallel. There are a few ways I can see to do this: a, Have a seperate and completely distinct main and video memory. So in effect the two systems can run at the same time and be considered as different computers networked together. This would probably be the simplest of the parallel systems to implement. It is good in that there can never be any memory contention, But bad in that you are forced to use the RISK chip for all (but mode 1 to 4) graphic routeens, even if the Z380 may be better at a particular effect. It also suffers from the 'all memory is one of two types' problem mentioned above, Note that this lack of video memory problem will be increased if hardware scrolling is required within the video memory. b,Give each computer N memory chips for their own use, where N can be allocated dynamically. i.e. surpose we use 4 1Meg simms. If you want a small resolution with a limited number of colours you tell the Z380 that it can use memory chips 1 to 3 and the RISK that it can use chip 4. If later on in the application you want a better quality static picture you ensure the OS is not using 3rd chip then tell the RISK that it can now use it and the Z380 that it can no longer use the chip. So the Z380 will be using chips 1 to 2 and the graphics can be stored on chips 3 and 4. This is a compromise option. It also means that if, surpose it was faster to use the RISK chip to draw the vector graphics in our version of Elite and the Z380 to draw the contol pannel, you may be able to change the chip allocation when the screen is updating the top of the picture to give the Z380 the use of chip 3 then swap it back to the RISK before it tries to display it -For super fast games only! A bad idea for the OS to try this! c,Allow both to use the memory bank, but if the two processors try to use the same memory chip (or part of memory chip) then give one chip priority to interupt and make the other wait. This would give a very versitile system but it must be ensured that the programs are written to keep the program as far away from the graphics as possible to avoid one of the processors from locking the other up excessively. d,Similar to c but give each chip its own cache to limit acesses to the main memory. d,Don't give either system direct memory access to the 1 big chunk but make them ask an MMU (A new term for me too!) to process the information for them. This may seem slow but it is what Silicon graphics do in their new (super fast and super expensive) O2 systems! Please use this space to moan about the continual suggestion of co-processor machines....................................................... ............................................................................ ............................................................................ I would lean towards option iia or iib. iia should be the most practical. assuming the programable RISK is powerfull enough. We must have hardware texture mapping capability in the SON of SAMSON! (It will be necessary to view the incarnation of the www about 3 years down the road methinks!) Please ignore me Numb. From owner-sam-users@nvg.unit.no Sat Nov 9 01:46:16 1996 From: SL Harding Message-Id: <199611090143.BAA04555@harrier.fen.bris.ac.uk> Subject: Re: SOS box colour To: sam-users@nvg.unit.no Date: Sat, 9 Nov 1996 01:43:50 +0000 (GMT) In-Reply-To: <96110613175095@morse.ntu.ac.uk> from "ALLAN CLARKSON" at Nov 6, 96 01:17:51 pm X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 278 Lines: 12 > Yeah: SAM Colour - purple > It's got to be! > > And *who* could Samsboss be? > > Al. > Why not coloured indigo? Then we could put a sticker on the side saying 'Silicon Graph1cs'. We would shift millions of them if we priced them at 299 pounds each! C9 :o) Numb. From owner-sam-users@nvg.unit.no Sat Nov 9 05:00:31 1996 Date: Fri, 8 Nov 1996 19:00:54 -0500 From: Neville Young <106166.1560@compuserve.com> Subject: my C compiler Cc: sam users Message-Id: <199611082359_MC1-BC1-48E8@compuserve.com> To: unlisted-recipients:;@nvg.unit.no (no To-header on input) Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 690 Lines: 50 Very brief notes on my C compiler for your info. HighTech C by Avocet. Primitives char 8 bits short 16 bits long 32 bits int 16 bits float 32 bits double 32 bits supports Enumerations. Type qualifiers: far near interrupt fast interrupt port static volatile persistent Rom development models: small z80 small z180 large z80 large z180 cp/m models cp/m z80 cp/m z180 I have written a simple startup module for sam supports Z80 Z180 & 64180 chips Full assembler in line Baised on K&R C with ANSI extensions. Generates: cpp -> cc-> (optimise->) asm -> link-> bin Loads of standard libs with sources. Not a small C but a full implimentation. Anything else you want to know. Nev. From owner-sam-users@nvg.unit.no Sat Nov 9 11:34:04 1996 Message-Id: <199611091134.LAA12993@mail.enterprise.net> From: davewhitmore@enterprise.net (Dave Whitmore) To: sam-users@nvg.unit.no Subject: Re : Killer apps - Internet Date: 08 Nov 1996 23:37:42 References: <1C500E4B9F@smserver1.ulst.ac.uk> X-Newsreader: Spot 1.3 X-Mailer: NetGate 1.3 (Amiga; TCP/IP) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1746 Lines: 37 * In a message of 07 Nov 96 Gavin Smith wrote to sam-users@nvg.unit.no: Hi Gavin, >> People who scoff at the thought of fidonet just don't know how good it >> really is. Maybe they've never tried it? GS> The point is not whether it is good or not, the point is whether it GS> will sell SAM's - this is where the internet is a priority. If someone GS> wants to try fidonet after we have internet access, fine. No. My point is that people should be able to do fidonet *now!* Without any acceleration or extra gizmos. I'm not saying "Hey, buy this, it can do fido!", I'm saying that Comm'ix, (and Termite, when it has it's Xfer protocols sorted) can upload and download fidonet mail packets. The machine can do it now. Before we can do anything, a new comms interface will be needed. The current ones can be bodged (or modified, as we like to say). If/when this comms interface is sorted, general BBSing, fidonet and all the other amateur networks that can be accessed and will be available to SAM users, without needing to subscribe to internet; which will also cost people more. Internet will be a lot harder to implement than a bog standard terminal program. It is ridiculous to say internet first, when so much work has been done with Termite & Comm'ix. Either way, an Off Line Reader program will be necessary, so why not make one that handles Internet Email, Newsgroups and Fidonet in one package. Or, we could do the Fido first and bolt on internet email later. Whatever happens with comms on SAM. we'll need a ZIPer and an UnZIPer, LHA, GIF & JPEG decoders, proper ASCII text editors.. oh the list goes on and gets more scary the more I think of it. :/ Bye, _ |_)ave (/\)hitmore email - davewhitmore@enterprise.net From owner-sam-users@nvg.unit.no Sat Nov 9 11:34:04 1996 Message-Id: <199611091134.LAA13008@mail.enterprise.net> From: davewhitmore@enterprise.net (Dave Whitmore) To: sam-users@nvg.unit.no Subject: Internet = sad Date: 08 Nov 1996 23:56:20 References: <199611072355.XAA01270@pipe1.uk.pipeline.com> X-Newsreader: Spot 1.3 X-Mailer: NetGate 1.3 (Amiga; TCP/IP) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1577 Lines: 38 In a message of 07 Nov 96 Samsboss@uk.pipeline.com wrote to sam-users@nvg.unit.no: Sup> Fidonet is good, because you like it - it does things the way you like Sup> them to be done. Nope. I like internet. I pay for internet and I get fidonet free (well, there's the price of a phone call). Sup> Other nets all have their good points. Sup> But the sad fact is - what people think of as the internet (in fact Sup> they are only really looking at the WWW) is what they /think/ they Sup> need. Once they have a machine, you are welcome to try and educate Sup> them, but it will be the magic word INTERNET that gets them. The fact that SAM can't do fidonet properly now is quite sad, when most other machines can do it without any bother.. The sad fact is that WWW is one hell of a lot further away (for SAM) than amateur networks are. WWW is impossible to do on the current SAM, Fidonet isn't. Okay, someone might say we can get the text and leave the graphics out, but none of those said people will see anything magic about that. It's all very well thinking about a new accelerated SAM with a new processor, graphic capabilities, multitasking, CLI, great GUI and internet access, etc. If we're not careful, we'll end up with nothing more than an Amiga 1200 surfer pack.. and just look how popular that is (not)! And I've already got all that anyway. It sure is a pity that internet seems to be the only reason people want a computer now.. In fact it is very sad. :) but I'll smile instead. Bye, _ |_)ave (/\)hitmore email - davewhitmore@enterprise.net From owner-sam-users@nvg.unit.no Sat Nov 9 11:48:16 1996 Date: Sat, 9 Nov 1996 06:46:20 -0500 From: BrenchleyR@aol.com Message-Id: <961109064619_1383566078@emout07.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Serial Interface Questions. Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 263 Lines: 14 In a message dated 08/11/96 17:08:28, you write: >>4) Again, if we change, should the new interface have one or two RS232 >ports? > >Have you got a postscript Printer, BOB? > >Simon > > No, but one for a modem and one for a (well something else anyway). Bob. From owner-sam-users@nvg.unit.no Sat Nov 9 13:15:55 1996 X-Authentication-Warning: dcsun4.comp.brad.ac.uk: lwillis owned process doing -bs Date: Sat, 9 Nov 1996 13:13:41 +0000 (GMT) From: Lee To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message In-Reply-To: <199611081805.SAA10037@reed.bris.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 387 Lines: 15 On Fri, 8 Nov 1996, SL Harding wrote: ;>Whoo! I have just been playing on a Silicon graphics O2! The GUI is built ;>around Netscape! I presume this is a sick joke, since that would kind of be a LARGE step down for a SG machine !? Lee. How many Microsoft employees does it take to screw in a light bulb? None, they just make darkness the industry standard! From owner-sam-users@nvg.unit.no Sat Nov 9 13:56:03 1996 X-Authentication-Warning: dcsun4.comp.brad.ac.uk: lwillis owned process doing -bs Date: Sat, 9 Nov 1996 13:53:48 +0000 (GMT) From: Lee To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 644 Lines: 18 On Fri, 8 Nov 1996, A.S. Collier wrote: ;>reclick on chosen option. Also... CNTRL+first letter of menu name to bring ;>up menu, and use cursor to select option. Also... Function keys as ;>short-cut to specific options. I'd suggest that some convention is kept ;>to, eg F1 Save as, F2 Save, F3 Load; F7 Copy, F8 Cut, F9 Paste Can't we have sensible things like CNTRL+s is save, the sort of thing that almost _every_ GUI I've used uses ?. Save Function keys for application-specific uses. Lee. How many Microsoft employees does it take to screw in a light bulb? None, they just make darkness the industry standard! From owner-sam-users@nvg.unit.no Sat Nov 9 16:55:25 1996 From: Johnna Teare Organization: University of Central Lancashire To: sam-users@nvg.unit.no Date: Sat, 9 Nov 1996 16:46:24 GMT+0 Subject: Re: Re : Killer apps - Internet Priority: normal X-Mailer: Pegasus Mail for Windows (v2.33) Message-Id: <2CEF03143C@mail-gw.uclan.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 445 Lines: 15 > In a message dated 08/11/96 15:31:38, you write: > > >No, but I think he's starting to make some sense :). I take it you > >spellcheck FORMAT before you publish it! > > No, that is Jenny's job. And anyway, there is a whole month to get each issue > right, with the rate these emails fly these days, I can't be blamed for a fwe > mitsakse cna I? > > Bob. > **grins** Johnna Pig Teare (j.d.teare@uclan.ac.uk) "They call me Mad The Swine" From owner-sam-users@nvg.unit.no Sat Nov 9 17:46:48 1996 Message-Id: <3285338C.FA3@csv.warwick.ac.uk> Date: Sat, 09 Nov 1996 17:44:44 -0800 From: Paul Walker Organization: University of Warwick X-Mailer: Mozilla 2.0 (Win16; I) Mime-Version: 1.0 To: sam-users@nvg.unit.no Subject: Re: We Must Have A Modem So There... References: <199611061651.QAA10278@pipe1.uk.pipeline.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 327 Lines: 7 Samsboss@uk.pipeline.com wrote: > I think a modem should be VERY high on the list of projects to add onto the > back of SAM and to put in the SAMSON case. Personally, I think a fast serial port should be very high on the list - then there won't be any need to design a modem, as you can just use the many already available. From owner-sam-users@nvg.unit.no Sat Nov 9 20:25:29 1996 Date: Sat, 9 Nov 1996 15:23:12 -0500 From: BrenchleyR@aol.com Message-Id: <961109152311_1316612899@emout09.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: We Must Have A Modem So There... Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 555 Lines: 17 In a message dated 09/11/96 17:46:28, Paul Walker wrote: >Personally, I think a fast serial port should be very high on the list - then >there won't be any >need to design a modem, as you can just use the many already available. > > I've always thought that a dedicated modem would be cheaper to produce, especially if it did not go through the BT certification process. These days there must be an 'off the shelf' chip-set we could drop onto a board. The problem with an external modem is the high cost in relation to the cost of the machine. Bob. From owner-sam-users@nvg.unit.no Sun Nov 10 12:33:59 1996 Date: Sun, 10 Nov 1996 12:27:13 GMT Message-Id: <199611101227.MAA19884@pipe1.uk.pipeline.com> To: sam-users@nvg.unit.no Subject: New Screen Modes. From: samsboss@uk.pipeline.com (Samsboss@uk.pipeline.com) Cc: sam-users@nvg.unit.no X-Pipeuser: samsboss X-Pipehub: uk.pipeline.com X-Pipegcos: (Samsboss@uk.pipeline.com) X-Mailer: Pipeline v3.5.0 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 431 Lines: 12 There has been several mailinks about the number of colours on the screen and sow on, but has anybods given thoughts to the screen layout? Where duz us go beyond modes 1,2,3 & 4? I think the start should be Mode 5, as mode 3 but 16 colours. Then Mode 6, as Mode 5 but double the vertical resolution. Samsboss. Its Sunday, there was not new postings, so I thought to meself, go on - post it anyway, someone may read it. From owner-sam-users@nvg.unit.no Sun Nov 10 12:47:56 1996 From: Johnna Teare Organization: University of Central Lancashire To: sam-users@nvg.unit.no Date: Sun, 10 Nov 1996 12:43:10 GMT+0 Subject: Re: New Screen Modes. Priority: normal X-Mailer: Pegasus Mail for Windows (v2.33) Message-Id: <40DBC12C77@mail-gw.uclan.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1022 Lines: 29 Hello > There has been several mailinks about the number of colours on the screen > and sow on, but has anybods given thoughts to the screen layout? > > Where duz us go beyond modes 1,2,3 & 4? > > I think the start should be Mode 5, as mode 3 but 16 colours. Then Mode 6, > as Mode 5 but double the vertical resolution. How about a mode with hardware scrolling? And I've got some stupid girl sitting next to me who keeps swearing at the computer really loudly and hammering away at the keys. It's a delight to see the PC pushed to it's limits! > > Samsboss. > Its Sunday, there was not new postings, so I thought to meself, go on - > post it anyway, someone may read it. Well, I did, and I've been waiting for new mailing since I logged in at 10am. Expecting a ton of new mails (it's usually about 70 or so every time I clock in) I got 5 and three of those were mail delivery errors! Is everybody hungover, or having Sunday lunch? Johnna Pig Teare (j.d.teare@uclan.ac.uk) "They call me Mad The Swine" From owner-sam-users@nvg.unit.no Sun Nov 10 12:56:45 1996 Message-Id: <199611101254.NAA04154@mailserv.caiw.nl> From: Robert van der Veeke To: sam-users Subject: Re: New Screen Modes. Date: Sun, 10 Nov 1996 13:57:39 +0100 X-Msmail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1155 Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 462 Lines: 12 > Well, I did, and I've been waiting for new mailing since I logged in > at 10am. Expecting a ton of new mails (it's usually about 70 or so > every time I clock in) I got 5 and three of those were mail delivery > errors! > > Is everybody hungover, or having Sunday lunch? yawn, goodmorning, grat grat, mmmmmhhhhh 1.57 PM already Robert van der Veeke AKA RJV Graphics [rjvveeke@caiw.nl] Red leader here: watch out we are running out of diskspace real soon! From owner-sam-users@nvg.unit.no Sun Nov 10 14:02:22 1996 X-Authentication-Warning: dcsun4.comp.brad.ac.uk: lwillis owned process doing -bs Date: Sun, 10 Nov 1996 13:59:32 +0000 (GMT) From: Lee To: sam-users@nvg.unit.no Subject: Re: New Screen Modes. In-Reply-To: <40DBC12C77@mail-gw.uclan.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 521 Lines: 20 On Sun, 10 Nov 1996, Johnna Teare wrote: ;>Is everybody hungover, or having Sunday lunch? Neither, but as far as I'm concerned any time before 1:30 on a sunday afternoon simply doesn't exist! ;>Johnna Pig Teare (j.d.teare@uclan.ac.uk) ;>"They call me Mad The Swine" 'Scuse me being ignorant, but are you a bloke or girly, Johnna's a name I ain't heard before .. Lee. How many Microsoft employees does it take to screw in a light bulb? None, they just make darkness the industry standard! From imc Sun Nov 10 14:27:46 1996 Subject: Re: New Screen Modes. To: sam-users@nvg.unit.no Date: Sun, 10 Nov 1996 14:27:46 +0000 (GMT) In-Reply-To: <199611101227.MAA19884@pipe1.uk.pipeline.com> from "Samsboss@uk.pipeline.com" at Nov 10, 96 12:27:13 pm X-Mailer: ELM [version 2.4 PL24 PGP6] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 447 Lines: 12 On Sun, 10 Nov 1996 12:27:13 GMT, Samsboss@uk.pipeline.com said: > There has been several mailinks about the number of colours on the screen > and sow on, but has anybods given thoughts to the screen layout? > I think the start should be Mode 5, as mode 3 but 16 colours. But you didn't mention the layout! I think rectangular pixels such as those of mode 3 are rather limited in use. Most modern displays have square pixels, I believe. imc From imc Sun Nov 10 15:12:52 1996 Subject: Re: Plea! Plea! Plea! + Bob's message To: sam-users@nvg.unit.no Date: Sun, 10 Nov 1996 15:12:52 +0000 (GMT) In-Reply-To: from "A.S. Collier" at Nov 8, 96 11:57:48 pm X-Mailer: ELM [version 2.4 PL24 PGP6] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 5679 Lines: 117 On Fri, 8 Nov 1996 23:57:48 +0000 (GMT), A.S. Collier said: > (At first) Mode 4, 8 standard colours, plus eight which depend on the > currently selected window. Why such an arbitrary arrangement? Surely you could allow each window to choose how many "private" colours it wants. (In X, an application can have an ordinary colormap or a private one. Most windows use the ordinary one so that colours don't flash when you move in and out of windows. A window will ask for a particular colour and be given back an ink number that was either already defined to be that colour or was unused and has newly been defined to be that colour. This fails when all the ink numbers have been used up. Some windows then create a private colormap instead. They request colours in the same manner, but this time all the inks start off unused. The palette is remapped every time a window with a private colormap is selected.) > Background pattern. Sound samples (probably > disrupt working of applications on Z80B, but don't necessarily need to for > the Z380). Why does a windowing system need sound samples anyway? > Event-driven control architecture. Marketing nonsense? :-) > All applications have windows on the one > screen.) Of course, given that you are planning to do all this on a 256x192 screen it might help to have a virtual desktop (in other words, all the windows are not on the same screen). > Several Window types, easiest one for programmer is: draw the contents of > the window in some misc area of memory, and inform the OS when you want > that copied to the actual screen. Yes, I know it introduces a software > overhead but the advantages to the apps programmer are: No need to remove > mouse pointer before altering contents of window, no need to bother > whether your window is selected, covered up or whatever; Each of these things would (or at least should) always be true, no matter what mechanism the programmer uses for drawing on a window. > no need to bother > about what the user has done to the size or position of the window; How do you create the window contents then if you don't know what the size of the window is? > to > draw the contents slowly does not mean the screen changes slowly and > there's no flickering, the mouse pointer doesn't have to be taken off > screen for any long periods, so although the changes take slightly longer > the user doesn't notice anyway. These should perhaps be under the control of the programmer. In other words, even if you give her several methods to draw the window she could still decide to draw it all out first and then send the whole thing to the windowing system. I don't actually know how windows get drawn in X, but I do know that something very similar to your description exists, because that's how xz80 draws the screen. Whenever the application wants to change the screen, or receives a signal from the windowing system that part of it has been uncovered, it sends a bit image of (part of) the window to the windowing system, and the windowing system puts it on the screen (taking notice of which bits of the window are actually visible). I believe there are also other methods, such as telling the windowing system to draw some graphics in your window. Also, when you send a bit of your bitmap you are not restricted to placing it in any one position in the window. > Plus there will be the usual text windows, Who draws a text window then? Is this done as above or differently? > call this address "xxx" when the window gets drawn, dialogue boxes etc > etc. Moving, resizing and scroll bars are handled by the OS without > bothering the application. Bad idea. The application may want to know when it's been resized. It may even forbid the user to resize it (it doesn't make sense for xz80 to be any size except the size it starts up as). The resizing itself is of course handled by the window manager, but once it's done the application should be informed. About scroll bars: how does the windowing system know what the scroll bars are for? It doesn't. It can only tell the window "someone just clicked on the scroll bar, 5 pixels from the top" or "someone is moving the slider of the scroll bar". > Info bar at bottom of screen, gives infomation depending on whatever the > mouse is pointing to. Eg Details of menu options, copyright messages, or > something controlled by the application. (Inspired by Mac's bubble help, > but much faster and less obtrusive; it first came about because I needed > the background to be repeated in 16-pixel squares, and I had 8 pixels left > over after the menu bar) Ah, so it's actually only a 256x176 windowing system. You didn't instead think "Well why don't we save some screen space and get rid of the menu bar" then... :-) > Ability to quick-start the screensaver by moving mouse pointer to bottom > corner of screen. Rather non-intuitive. > NB A user-definable graphic > for folders will be difficult on a Sam format disk (reason being the > folder is not allocated any disk space, apart from the directory entry > itself, so there is nowhere to store the icon) There's nothing to stop you from allocating disk space though, is there? A "directory" directory entry still has a "used sectors" map. imc PS The one thing I hate about PC windowing systems is click-to-type. I like to be able to type in which ever window contains the mouse pointer. From imc Sun Nov 10 15:17:19 1996 Subject: Re: Plea! Plea! Plea! + Bob's message To: sam-users@nvg.unit.no Date: Sun, 10 Nov 1996 15:17:19 +0000 (GMT) In-Reply-To: from "A.S. Collier" at Nov 8, 96 04:27:32 pm X-Mailer: ELM [version 2.4 PL24 PGP6] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 892 Lines: 20 On Fri, 8 Nov 1996 16:27:32 +0000 (GMT), A.S. Collier said: > Absolutlely, Z380 all the way. And not just for sentimental reasons > either... > Question to anybody who knows a bit more about the Z380 that I do: For a > Windows-system application, of which there may be many in memory at one > time; will code need to be 100% relocatable? Or is there some system where > an address, when offset by (say) 64K, will look exactly the same to the > processor? You need an MMU for this, which is why I'm hesitant to base a new computer on any processor that doesn't support MMUs. I don't quite know how the MMU arranges this, but in Unix each process has a block of memory that starts at address 0. The only code that needs to be relocateable therefore is code in shared libraries (which will be used by more than one process and may be paged in at a different address in each process). imc From owner-sam-users@nvg.unit.no Sun Nov 10 15:19:52 1996 Date: Sun, 10 Nov 1996 15:13:31 GMT Message-Id: <199611101513.PAA25631@pipe1.uk.pipeline.com> To: sam-users@nvg.unit.no Subject: Re: New Screen Modes. From: samsboss@uk.pipeline.com (Samsboss@uk.pipeline.com) Cc: sam-users@nvg.unit.no X-Pipeuser: samsboss X-Pipehub: uk.pipeline.com X-Pipegcos: (Samsboss@uk.pipeline.com) X-Mailer: Pipeline v3.5.0 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1200 Lines: 37 On Nov 10, 1996 12:43:10, 'Johnna Teare ' wrote: >Hello > >> There has been several mailinks about the number of colours on the screen >> and sow on, but has anybods given thoughts to the screen layout? >> >> Where duz us go beyond modes 1,2,3 & 4? >> >> I think the start should be Mode 5, as mode 3 but 16 colours. Then Mode 6, >> as Mode 5 but double the vertical resolution. > >How about a mode with hardware scrolling? > >And I've got some stupid girl sitting next to me who keeps swearing >at the computer really loudly and hammering away at the keys. It's a >delight to see the PC pushed to it's limits! >> >> Samsboss. >> Its Sunday, there was not new postings, so I thought to meself, go on - >> post it anyway, someone may read it. > >Well, I did, and I've been waiting for new mailing since I logged in >at 10am. Expecting a ton of new mails (it's usually about 70 or so >every time I clock in) I got 5 and three of those were mail delivery >errors! > >Is everybody hungover, or having Sunday lunch? -- Sum bugger's gone and organised a party and not invited us - that's what's happened. Samsboss. Sad and alone... From owner-sam-users@nvg.unit.no Sun Nov 10 15:22:33 1996 Date: Sun, 10 Nov 1996 15:15:44 GMT Message-Id: <199611101515.PAA25725@pipe1.uk.pipeline.com> To: sam-users@nvg.unit.no Subject: Re: New Screen Modes. From: samsboss@uk.pipeline.com (Samsboss@uk.pipeline.com) Cc: sam-users X-Pipeuser: samsboss X-Pipehub: uk.pipeline.com X-Pipegcos: (Samsboss@uk.pipeline.com) X-Mailer: Pipeline v3.5.0 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 506 Lines: 16 On Nov 10, 1996 13:57:39, '"Robert van der Veeke" ' wrote: >> Is everybody hungover, or having Sunday lunch? > >yawn, goodmorning, grat grat, mmmmmhhhhh 1.57 PM already > >Robert van der Veeke AKA RJV Graphics >[rjvveeke@caiw.nl] >Red leader here: watch out we are running out of diskspace real soon! -- Look Robert, You's been ignoring that Red Leader for too long. Doooo sumit about your lack of disky space, please, before him get really upset. Samsboss. From owner-sam-users@nvg.unit.no Sun Nov 10 15:24:09 1996 Date: Sun, 10 Nov 1996 15:18:13 GMT Message-Id: <199611101518.PAA25907@pipe1.uk.pipeline.com> To: sam-users@nvg.unit.no Subject: Re: New Screen Modes. From: samsboss@uk.pipeline.com (Samsboss@uk.pipeline.com) Cc: sam-users@nvg.unit.no X-Pipeuser: samsboss X-Pipehub: uk.pipeline.com X-Pipegcos: (Samsboss@uk.pipeline.com) X-Mailer: Pipeline v3.5.0 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 337 Lines: 15 On Nov 10, 1996 13:59:32, 'Lee ' wrote: >'Scuse me being ignorant, but are you a bloke or girly, Johnna's a name I >ain't heard before .. > >Lee. -- Johnna? Johnna!!! Him old cricket man aint he. And I don't want no other little girly invading my patch :) Samsboss. Still got em on de run... From owner-sam-users@nvg.unit.no Sun Nov 10 15:29:02 1996 Date: Sun, 10 Nov 1996 15:20:43 GMT Message-Id: <199611101520.PAA26017@pipe1.uk.pipeline.com> To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message From: samsboss@uk.pipeline.com (Samsboss@uk.pipeline.com) Cc: sam-users@nvg.unit.no X-Pipeuser: samsboss X-Pipehub: uk.pipeline.com X-Pipegcos: (Samsboss@uk.pipeline.com) X-Mailer: Pipeline v3.5.0 Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1153 Lines: 31 On Nov 10, 1996 15:17:19, 'Ian.Collier@comlab.ox.ac.uk' wrote: >On Fri, 8 Nov 1996 16:27:32 +0000 (GMT), A.S. Collier said: >> Absolutlely, Z380 all the way. And not just for sentimental reasons >> either... > >> Question to anybody who knows a bit more about the Z380 that I do: For a >> Windows-system application, of which there may be many in memory at one >> time; will code need to be 100% relocatable? Or is there some system where >> an address, when offset by (say) 64K, will look exactly the same to the >> processor? > >You need an MMU for this, which is why I'm hesitant to base a new computer >on any processor that doesn't support MMUs. > >I don't quite know how the MMU arranges this, but in Unix each process >has a block of memory that starts at address 0. The only code that needs >to be relocateable therefore is code in shared libraries (which will be >used by more than one process and may be paged in at a different address >in each process). -- Its like paging the 64K area on SAM. The MMU just puts the area into the address range 0 to whateversneeded - but transparant like. (I think) Samsboss. From owner-sam-users@nvg.unit.no Sun Nov 10 15:44:30 1996 From: Mr P R Walker Message-Id: <25106.199611101542@lily.csv.warwick.ac.uk> Subject: Re: We Must Have A Modem So There... To: sam-users@nvg.unit.no Date: Sun, 10 Nov 1996 15:42:43 +0000 (GMT) In-Reply-To: <961109152311_1316612899@emout09.mail.aol.com> from "BrenchleyR@aol.com" at Nov 9, 96 03:23:12 pm X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 563 Lines: 12 > I've always thought that a dedicated modem would be cheaper to produce, > especially if it did not go through the BT certification process. These days I'm against that for a couple of reasons : a) that modem would then be illegal to use on the phone system, and b) it would lock the owner of the machine to /your/ modem (your in the sense of the company), whereas using a serial port you could use any modem you liked, already had. If I already have a modem (which I do), I'd be pissed off if I had to buy another which I could /only/ use with the sam! Paul From owner-sam-users@nvg.unit.no Sun Nov 10 17:05:50 1996 Message-Id: <199611101704.RAA20563@mail.enterprise.net> From: davewhitmore@enterprise.net (Dave Whitmore) To: sam-users@nvg.unit.no Subject: We Must Have A Modem So There... Date: 10 Nov 1996 16:51:17 References: <25106.199611101542@lily.csv.warwick.ac.uk> X-Newsreader: Spot 1.3 X-Mailer: NetGate 1.3 (Amiga; TCP/IP) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1197 Lines: 26 In a message of 10 Nov 96 Mr P R Walker wrote to sam-users@nvg.unit.no: >> I've always thought that a dedicated modem would be cheaper to produce, >> especially if it did not go through the BT certification process. These >> days MPR> I'm against that for a couple of reasons : MPR> a) that modem would then be illegal to use on the phone system, and b) MPR> it would lock the owner of the machine to /your/ modem (your in the MPR> sense of the company), whereas using a serial port you could use any MPR> modem you liked, already had. If I already have a modem (which I do), MPR> I'd be pissed off if I had to buy another which I could /only/ use with MPR> the sam! Me too! TBH, I wouldn't touch one if that were the case. The only reason I can see for having a dedicated modem would be to monopolise the situation. No chance! I can't think of any other computer that has it's own modem. It's a bit like going back to the days when you had to buy the Commodore tape machine for the C64. Build a new comms interface by all means, but don't close the doors to a choice of modem. People wouldn't like that at all. _ |_)ave (/\)hitmore email - davewhitmore@enterprise.net From owner-sam-users@nvg.unit.no Sun Nov 10 17:57:50 1996 Date: Sun, 10 Nov 1996 12:55:45 -0500 From: BrenchleyR@aol.com Message-Id: <961110125544_1982459426@emout14.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: We Must Have A Modem So There... Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2139 Lines: 51 In reply to both a message from P.R.Walker and one from D.Whitmore, both dated 10/11/96. In my original message I said: I've always thought that a dedicated modem would be cheaper to produce, especially if it did not go through the BT certification process. These days there must be an 'off the shelf' chip-set we could drop onto a board. The problem with an external modem is the high cost in relation to the cost of the machine. Paul replied: > I'm against that for a couple of reasons : > a) that modem would then be illegal to use on the phone system, Very few of us would be prevented from using an 'un-approved' modem because I would imagine that most of us now have the modern style Telephone plugs. These isolate the internal wiring/equipment from BT's lines. It is only illigal to connect unapproved equipment DIRECT to the BT line (dispite what BT will tell you). >and b) >it would lock the owner of the machine to /your/ modem (your in the >sense of the company), whereas using a serial port you could use any > modem you liked, already had. If I already have a modem (which I do), > I'd be pissed off if I had to buy another which I could /only/ use with > the sam! > Dave also said: >Me too! TBH, I wouldn't touch one if that were the case. > >The only reason I can see for having a dedicated modem would be to monopolise >the situation. No chance! I can't think of any other computer that has it's >own modem. It's a bit like going back to the days when you had to buy the >Commodore tape machine for the C64. Well I, and millions of others use a dedicated modem every day. Internal PC modems out sell external by about 6 to 1. Reason? No need for a case, power supply, and in some cases simpler design. > >Build a new comms interface by all means, but don't close the doors to a >choice of modem. People wouldn't like that at all. Given a choice I think that most people would go for internal, even if there was not a money saving involved. Yes, by all means, lets have a fast serial interface for those that want it. But if we want to keep prices down then we should not ignore the internal dedicated route. Bob. From owner-sam-users@nvg.unit.no Sun Nov 10 18:38:12 1996 Date: Sun, 10 Nov 1996 18:35:48 +0000 (GMT) From: "A.S. Collier" X-Sender: asc25@indigo.csi.cam.ac.uk To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message In-Reply-To: <9611101512.AA20039@boothp2.ecs.ox.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 9068 Lines: 189 On Sun, 10 Nov 1996 Ian.Collier@comlab.ox.ac.uk wrote: > On Fri, 8 Nov 1996 23:57:48 +0000 (GMT), A.S. Collier said: > > (At first) Mode 4, 8 standard colours, plus eight which depend on the > > currently selected window. > > Why such an arbitrary arrangement? Surely you could allow each window to > choose how many "private" colours it wants. > > (In X, an application can have an ordinary colormap or a private one. > Most windows use the ordinary one so that colours don't flash when you > move in and out of windows. A window will ask for a particular colour > and be given back an ink number that was either already defined to be > that colour or was unused and has newly been defined to be that colour. > This fails when all the ink numbers have been used up. Some windows then > create a private colormap instead. They request colours in the same manner, > but this time all the inks start off unused. The palette is remapped every > time a window with a private colormap is selected.) It will be *much* easier for the application, if it always knows what ink colours it is using. Say you want to print a sprite - do you always need a look-up table? "Today, yellow is colour number, er, 11." I wouldn't like to write programs like that at all. Note also that, for the moment, we're running this in sixteen colours, so with your arrangement the CLUT would get full up very quickly indeed. With mine, although the contents of the windows may change colour, the background and borders never will. And because the first five colours are a convenient scale, most applications will want to use those. > Why does a windowing system need sound samples anyway? You might as well ask, why does it need graphics? It makes things nicer for the user, to be given some audible response when things happen. > > Event-driven control architecture. > > Marketing nonsense? :-) When the user clicks somwehere in the window, this is an EVENT. The OS calls a routine, in the application, which is associated with that event. IE the event DRIVES the application. Hence, EVENT DRIVEN. > > Several Window types, easiest one for programmer is: draw the contents of > > the window in some misc area of memory, and inform the OS when you want > > that copied to the actual screen. Yes, I know it introduces a software > > overhead but the advantages to the apps programmer are: No need to remove > > mouse pointer before altering contents of window, no need to bother > > whether your window is selected, covered up or whatever; > > Each of these things would (or at least should) always be true, no matter what > mechanism the programmer uses for drawing on a window. Of course it isn't! If you're application just writes staright to the screen without bothering over the above conditions, you will a) corrupt the contents of the other windows, b) be unable to change the screen underneath the mouse pointer. > > no need to bother > > about what the user has done to the size or position of the window; > > How do you create the window contents then if you don't know what the size > of the window is? Ian, READ this please. (See Bob, I argue with *everybody*) Your application writes to an area of general memory. This is unaffectted by the mouse pointer or the other windows, so you can write to it whenever you like, at whatever speed you like. The amount of memory allocatted to a window corresponds to the largest size that the window can have (and is chosen by the application when the window was opened). When the screen needs to be updated, you call a jump table routine in the OS, saying which parts of the window have been updated. The GUI then removes the mouse pointer if necessary, checks that the relavent area of that window is still being displayed, and if so, copies those areas to the screen. Reasons it may not be displayed include: - it is covered up by another window - the window has been resized, and that section is not in view. So as I said, the application doesn't need to worry about any of those things. It can if it wants to, so you WILL be able to find things like: what size the window is, where the window is being displayed. etc etc. But using the above method of drawing the screen, apps programmers will not need to do that. > These should perhaps be under the control of the programmer. In other > words, even if you give her several methods to draw the window she could > still decide to draw it all out first and then send the whole thing to the > windowing system. Now that sounds familiar. Draw it out first... Draw things in some area of memory. send the whole thing to the windowing system... call that jump table routine. > > Plus there will be the usual text windows, > > Who draws a text window then? Is this done as above or differently? The OS, using the fast text-placing routine used in the rest of the program. Basically you give it a text file and its length, and that is printed in the window. Word-wrapping, scrolling down etc is done automatically by the text routine, so your text will fit into the window whatever size the user makes it. > > call this address "xxx" when the window gets drawn, dialogue boxes etc > > etc. Moving, resizing and scroll bars are handled by the OS without > > bothering the application. > > Bad idea. The application may want to know when it's been resized. It may > even forbid the user to resize it (it doesn't make sense for xz80 to be any > size except the size it starts up as). The resizing itself is of course Set the variable maximum size and minimum size with the same values. > handled by the window manager, but once it's done the application should > be informed. About scroll bars: how does the windowing system know what > the scroll bars are for? It doesn't. It can only tell the window "someone > just clicked on the scroll bar, 5 pixels from the top" or "someone is moving > the slider of the scroll bar". YES IT DOES Because the area of memory which contains the bitmap data for the window is of a known size, and the part being displayeds if of a known size and known offset. So FOR THIS TYPE OF WINDOW (which was all I was talking about at that stage) the OS knows exactly how to interpret clicks on the scroll bar. > > Info bar at bottom of screen, gives infomation depending on whatever the > > mouse is pointing to. Eg Details of menu options, copyright messages, or > > something controlled by the application. (Inspired by Mac's bubble help, > > but much faster and less obtrusive; it first came about because I needed > > the background to be repeated in 16-pixel squares, and I had 8 pixels left > > over after the menu bar) > Ah, so it's actually only a 256x176 windowing system. You didn't instead > think "Well why don't we save some screen space and get rid of the menu > bar" then... :-) We have to remember that not everybody has even got a mouse, and those who don't will want to get at their menus by a system of key presses, rather than trying to move the pointer around. (Which is why CNTRL+letter brings up the menu instead of a specific option, which is something the function keys will need to do. After all, they're there so why not use them? Several PC programs do this sort of thing, like Turbo Pascal's IDE for example. Admittedly this is a DOS program using its own inbuilt windowsy environment, rather like SamMines does.) And we need a menu bar because nobody has got a three button mouse. Many applications will want to use a second button for some alternative function, and not to bring up the menu. If you're playing mines, you wouldn't want one of the menu options to be "place a marker here" - but you would need it to be unless the menus were in a bar at the top. > > Ability to quick-start the screensaver by moving mouse pointer to bottom > > corner of screen. > > Rather non-intuitive. So? It's a really nice feature on Mac system 7, and will be useful 'cos you know about it. > > NB A user-definable graphic > > for folders will be difficult on a Sam format disk (reason being the > > folder is not allocated any disk space, apart from the directory entry > > itself, so there is nowhere to store the icon) > > There's nothing to stop you from allocating disk space though, is there? > A "directory" directory entry still has a "used sectors" map. Maybe, but will MasterDos or SamDos notice that the space has been used? I don't want any normal-format disks to be backwards incompatible. > imc You don't say ;-) > PS The one thing I hate about PC windowing systems is click-to-type. I like > to be able to type in which ever window contains the mouse pointer. Oh right, so you accidently nudge the mouse and all your keyboard input gets thrown away until you notice. Lovely! In fact, that's what I hate MOST about Linux. See what I mean about not being able to please any of the people for even some of the time? Andrew From owner-sam-users@nvg.unit.no Sun Nov 10 18:41:15 1996 Date: Sun, 10 Nov 1996 18:38:17 +0000 (GMT) From: "A.S. Collier" X-Sender: asc25@indigo.csi.cam.ac.uk To: sam-users@nvg.unit.no Subject: Re: We Must Have A Modem So There... In-Reply-To: <961110125544_1982459426@emout14.mail.aol.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 307 Lines: 12 On Sun, 10 Nov 1996 BrenchleyR@aol.com wrote: > These isolate the internal wiring/equipment from BT's lines. It is only > illigal to connect unapproved equipment DIRECT to the BT line (dispite what > BT will tell you). What about BT's lawyers? You may well be right, but they've got more money. Andrew From owner-sam-users@nvg.unit.no Sun Nov 10 18:51:26 1996 Date: Sun, 10 Nov 1996 18:50:20 +0000 (GMT) From: "A.S. Collier" X-Sender: asc25@indigo.csi.cam.ac.uk To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message In-Reply-To: <9611101517.AA20064@boothp2.ecs.ox.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1412 Lines: 32 On Sun, 10 Nov 1996 Ian.Collier@comlab.ox.ac.uk wrote: > You need an MMU for this, which is why I'm hesitant to base a new computer > on any processor that doesn't support MMUs. It will be no *fun* to program a machine if (like the big operating systems) it starts to trap you whenever you try to do anything clever or out of the ordinary. And being 'fun to program' might be one of the big ways to attract programmers (apart from a better chance of fame than on the over-crowded PC systems). > I don't quite know how the MMU arranges this, but in Unix each process > has a block of memory that starts at address 0. The only code that needs > to be relocatable therefore is code in shared libraries (which will be > used by more than one process and may be paged in at a different address > in each process). I thought I'd remembered something about switching the Z380's PC and SP into 16 bit mode (with the upper sixteen being held constant) which would work fine if we've got a reasonable amount of memory. Working normally, do instructions like JP nnnnnnnn and LD (nnnnnnnn),A always need a 32-bit address? I need one of those manuals! Bob, Mr Organiser man, do you want to order a dozen or so for all the programmers, and we'll pay for the postage when you pass them on? Saves the trouble of thinking up a better excuse than fifteen Z380-controlled voice-operated telephone exchanges ;-) Andrew From owner-sam-users@nvg.unit.no Sun Nov 10 19:22:22 1996 Message-Id: From: ee31ag@ee.surrey.ac.uk (Andrew M Gale) Subject: Re: Plea! Plea! Plea! + Bob's message To: sam-users@nvg.unit.no Date: Sun, 10 Nov 1996 19:21:04 +0000 (GMT) In-Reply-To: <9611101517.AA20064@boothp2.ecs.ox.ac.uk> from "Ian.Collier@comlab.ox.ac.uk" at Nov 10, 96 03:17:19 pm X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 585 Lines: 13 > I don't quite know how the MMU arranges this, but in Unix each process > has a block of memory that starts at address 0. The only code that needs > to be relocateable therefore is code in shared libraries (which will be > used by more than one process and may be paged in at a different address > in each process). > So am I right in thinking that eacgh application thinks its running on its own unique machine with a contiguous block of memory stretching from 0 to .... er, some big number? If so, I don't think that would be too hard to implement in some limited fashion... -Andy From owner-sam-users@nvg.unit.no Sun Nov 10 20:53:21 1996 Subject: Re: Plea! Plea! Plea! + Bob's message To: sam-users@nvg.unit.no Date: Sun, 10 Nov 1996 20:37:19 +0000 (GMT) Cc: Simon.Cooke@nessie.mcc.ac.uk (Simon Cooke) In-Reply-To: from "Andrew M Gale" at Nov 10, 96 07:21:04 pm X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit From: Simon Cooke Message-Id: <"runix.runi.540:10.11.96.20.37.35"@runix.runit.sintef.no> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 327 Lines: 12 Memory manager... Hmm.... if all its got to do is give a contiguous memory map from 0 to eeek... The only problem left is demand-paging... which could be sorted if we had a support processor -- just halt the other one while the support processor goes off and gets the memory. It'd still slow things down though... Simon From owner-sam-users@nvg.unit.no Mon Nov 11 07:13:06 1996 Date: Mon, 11 Nov 1996 08:14:48 +0100 From: ft@edh.ericsson.se (Frode Tenneboe) Message-Id: <9611110714.AA12726@asmal.edh-net> To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message X-Sun-Charset: US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 520 Lines: 18 [bla bla bla] (Why not just port fvwm?) > > PS The one thing I hate about PC windowing systems is click-to-type. I like > > to be able to type in which ever window contains the mouse pointer. (I agree with Ian) > > Oh right, so you accidently nudge the mouse and all your keyboard input > gets thrown away until you notice. Lovely! In fact, that's what I hate > MOST about Linux. See what I mean about not being able to please any of > the people for even some of the time? Why not make it configurable? -Frode From owner-sam-users@nvg.unit.no Mon Nov 11 07:30:38 1996 Date: Mon, 11 Nov 1996 08:31:30 +0100 From: ft@edh.ericsson.se (Frode Tenneboe) Message-Id: <9611110731.AA12742@asmal.edh-net> To: sam-users@nvg.unit.no Subject: Re: SAMson marketing X-Sun-Charset: US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 901 Lines: 30 > By 'not easily obtainable, does this mean you managed to get it after a big > effort? > Or are you still trying - in which case email me details an I will try a > couple of my contacts. Alternative 2). :) http://www.cs.princeton.edu/software/lcc/ > > > >But first, you have to decide on 1) the memory structure and 2) > >the object file standard. > > Memory structure for Z380 board looks like it will be flat. Yes, but will it be with or without a MMU? > And by object file do you mean something to ge into a standard linker - which > is what I would love to see done. So do I. :) > >There exists about three zilion ready-to-compile-with-ansi-compiler > >programs which can be portet quite easily, amongst other things are > >algol-compilers, pascal-compilers, basic-compilers ;) > > Yes, but everyone says there is lots of source code - but where? Take a look at any gnu mirror. -Frode From owner-sam-users@nvg.unit.no Mon Nov 11 07:39:03 1996 Date: Mon, 11 Nov 1996 08:40:02 +0100 From: ft@edh.ericsson.se (Frode Tenneboe) Message-Id: <9611110740.AA12749@asmal.edh-net> To: sam-users@nvg.unit.no Subject: Re: We Must Have A Modem So There... X-Sun-Charset: US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2412 Lines: 58 > In reply to both a message from P.R.Walker and one from D.Whitmore, both > dated 10/11/96. > > In my original message I said: > I've always thought that a dedicated modem would be cheaper to produce, > especially if it did not go through the BT certification process. These days > there must be an 'off the shelf' chip-set we could drop onto a board. The > problem with an external modem is the high cost in relation to the cost of > the machine. > > Paul replied: > > I'm against that for a couple of reasons : > > a) that modem would then be illegal to use on the phone system, > > Very few of us would be prevented from using an 'un-approved' modem because I > would imagine that most of us now have the modern style Telephone plugs. > These isolate the internal wiring/equipment from BT's lines. It is only > illigal to connect unapproved equipment DIRECT to the BT line (dispite what > BT will tell you). And it will be difficult to adapt this for exports. > > >and b) > >it would lock the owner of the machine to /your/ modem (your in the > >sense of the company), whereas using a serial port you could use any > > modem you liked, already had. If I already have a modem (which I do), > > I'd be pissed off if I had to buy another which I could /only/ use with > > the sam! > > > Dave also said: > >Me too! TBH, I wouldn't touch one if that were the case. > > > >The only reason I can see for having a dedicated modem would be to > monopolise > >the situation. No chance! I can't think of any other computer that has it's > >own modem. It's a bit like going back to the days when you had to buy the > >Commodore tape machine for the C64. > > Well I, and millions of others use a dedicated modem every day. Internal PC > modems out sell external by about 6 to 1. Reason? No need for a case, power > supply, and in some cases simpler design. 1) Again an ISA bus would help, but 2) Intenal modems stinks - generally. > > > >Build a new comms interface by all means, but don't close the doors to a > >choice of modem. People wouldn't like that at all. > > Given a choice I think that most people would go for internal, even if there > was not a money saving involved. Yes, by all means, lets have a fast serial > interface for those that want it. But if we want to keep prices down then we > should not ignore the internal dedicated route. Not ignore, but carefully not consider it. ;) -Frode From imc Mon Nov 11 10:54:33 1996 Subject: Re: Plea! Plea! Plea! + Bob's message To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 10:54:33 +0000 (GMT) In-Reply-To: <9611110714.AA12726@asmal.edh-net> from "Frode Tenneboe" at Nov 11, 96 08:14:48 am X-Mailer: ELM [version 2.4 PL24 PGP6] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2680 Lines: 52 On Mon, 11 Nov 1996 08:14:48 +0100, Frode Tenneboe said: > (Why not just port fvwm?) Couldn't have put it better myself. :-) Unfortunately I don't know how window managers work, but it seems an excellent way to modularise the system and make it configurable at the same time. [For anyone who doesn't know what we are talking about, the window manager is the program which details things like where the windows pop up on the screen, how the user moves, resizes and selects them, what the window frame looks like, and so on. X has the concept of "a window" which can be any size, any position, iconised or not iconised, and may or may not have the focus, but X by itself does not give very much control to the user. The user runs a window manager to do this. The window manager can also decide such things as "if I press left function key 5 (which on a Sun keyboard has the word "Front" written on it) then the window raises itself. This is very convenient, since the window manager does not intercept mouse clicks inside windows (many windows require the mouse clicks for themselves so the click can't be used to raise the window as it is on certain other windowing systems: xterm uses the mouse for cutting and pasting, for example). It also decides what buttons there are on the window's title bar and what menus you get when you click on the title bar or on one of the buttons. It can also implement a virtual desktop so that for example if I move the mouse off the side of the screen it reappears on a different screen with different windows.] > > Oh right, so you accidently nudge the mouse and all your keyboard input > > gets thrown away until you notice. Lovely! In fact, that's what I hate > > MOST about Linux. See what I mean about not being able to please any of > > the people for even some of the time? > Why not make it configurable? In the absence of a window manager, this seems like the best option. Incidentally, accidentally nudging the mouse happens to me much less frequently than you seem to think. And if it does happen it isn't long before I notice that nothing is happening. It isn't as if I type a whole paragraph of text without looking at the screen! What does sometimes happen is that the mouse pointer isn't where I think it is, so my input goes to another window, with potentially harmful results. However, this happens far more frequently in the "click to type" model because then it still happens even when the mouse pointer is in the correct place, simply because I forgot to click it first. The ideal solution is thus to add a device which interprets which window I'm thinking of and then sends all the input to it. :-) imc From owner-sam-users@nvg.unit.no Mon Nov 11 11:07:13 1996 From: Mr P R Walker Message-Id: <3520.199611111102@skegger.csv.warwick.ac.uk> Subject: Re: We Must Have A Modem So There... To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 11:02:33 +0000 (GMT) In-Reply-To: <961110125544_1982459426@emout14.mail.aol.com> from "BrenchleyR@aol.com" at Nov 10, 96 12:55:45 pm X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 984 Lines: 24 > Very few of us would be prevented from using an 'un-approved' modem because I > would imagine that most of us now have the modern style Telephone plugs. BTW, the last time I read that warning it said "direct or indirect connection", I think. I'm not /sure/ about that tho. > Dave also said: > >Me too! TBH, I wouldn't touch one if that were the case. Same here. > Well I, and millions of others use a dedicated modem every day. In which case, all I can say is bad luck. Aside from everything else, Think of the number of people manufacturing PC modems. ANd then think of the number of people who'd be making the Sam ones. That's another reason that I wouldn't want a dedicated one. > Given a choice I think that most people would go for internal, even if there > was not a money saving involved. Yes, by all means, lets have a fast serial Have you asked them? Have you pointed out that with an external modem, they could also use it on any other computer they might have? From owner-sam-users@nvg.unit.no Mon Nov 11 11:21:41 1996 Message-Id: <1.5.4.32.19961111111635.0094a7bc@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 11 Nov 1996 11:16:35 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: We Must Have A Modem So There... Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 853 Lines: 25 At 11:02 AM 11/11/96 +0000, you wrote: >> Given a choice I think that most people would go for internal, even if there >> was not a money saving involved. Yes, by all means, lets have a fast serial > >Have you asked them? Have you pointed out that with an external modem, >they could also use it on any other computer they might have? One thing I think I should point out here: A SAM specific modem is going to cost ~40 UKP minimum to implement. Maybe more in the ~60 UKP range. For that money, you're going to get a modem which runs at 2400 baud -- if you're lucky. More likely it'll run at 1200 baud. For a similar price you can get a 9600 baud modem. For 30 pounds more, you can get a 14400 baud one. Why go for a custom-made modem, when surely the best option all-round is an external one? All we need now is a /GOOD/ comms interface. Simon From owner-sam-users@nvg.unit.no Mon Nov 11 11:27:09 1996 Date: Mon, 11 Nov 1996 11:20:04 +0000 (GMT) From: Justin Skists X-Sender: c93js1@klein.cms.dmu.ac.uk To: sam-users@nvg.unit.no Subject: Re: SOS: Multitasking In-Reply-To: <1.5.4.32.19961108163939.00936f80@nessie.mcc.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1190 Lines: 27 On Fri, 8 Nov 1996, Simon Cooke wrote: > At 03:48 PM 11/8/96 +0000, you wrote: > >> Just use the NMI :) - we may as well! > > > >How will the Kernel issue a NMI when it won't have control? Or are we expected > >to press the Break key every 50th of a second? :) > > Quite easy - if we're designing an accelerator board, what's to stop us > wiring a programmable timer to the NMI line? You know what? I had thought of this as soon as I left the computer lab.. Something done in hardware.. :) But what about (hardware) applications that need NMI? Such as the Break button? I take it things like the Break button will need to be on a I/O port and when the kernel detects an NMI, it would look at the I/O port and decide whether it is a Multitask swap NMI or another NMI and act accordingly... -- ============================================================================= |Justin Skists (c93js1@dmu.ac.uk) | Don't take life so seriously, | |BSc (Hons) Computer Science, Year 4 | you'll never survive it! | |De Montfort University, Leicester, England | - (C) CamLoRTS 1996 | ============================================================================= From owner-sam-users@nvg.unit.no Mon Nov 11 11:27:13 1996 Message-Id: X-Mailer: Novell GroupWise 4.1 Date: Mon, 11 Nov 1996 11:20:32 +0000 From: Dan Doore To: sam-users@nvg.unit.no Subject: Nothing to do with SamSon Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 470 Lines: 17 Dear All, I'm currently playing around in Sam C and I want to do something that I could do in Basic, but currently cannot in Sam C. In Sam C there are functions for open_scr, close_scr and display, but no function for the SCREEN command. The library calls for display (and such) are simple LD c,l and then CALL &xxx (not that I understand what it's doing - but I can guess :) Soo... Is there a ROM system call that I can call to act like the SCREEN command? Dan. From imc Mon Nov 11 11:41:08 1996 Subject: Re: Nothing to do with SamSon To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 11:41:08 +0000 (GMT) In-Reply-To: from "Dan Doore" at Nov 11, 96 11:20:32 am X-Mailer: ELM [version 2.4 PL24 PGP6] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 165 Lines: 6 On Mon, 11 Nov 1996 11:20:32 +0000, Dan Doore said: > Is there a ROM system call that I can call to act like the SCREEN command? Yes, I believe it's at 0100. imc From imc Mon Nov 11 11:52:05 1996 Subject: Re: SOS: Multitasking To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 11:52:05 +0000 (GMT) In-Reply-To: from "Justin Skists" at Nov 11, 96 11:20:04 am X-Mailer: ELM [version 2.4 PL24 PGP6] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 892 Lines: 17 On Mon, 11 Nov 1996 11:20:04 +0000 (GMT), Justin Skists said: > But what about (hardware) applications that need NMI? Such as the Break button? Well you could change this into a software key that the OS checks whenever it receives a timer interrupt. (Alternatively you could just use the frame interrupts for multitasking and decree that any program that disables the interrupts without it being absolutely necessary is a "badly behaved" app. (On OS/2 a "badly behaved" app doesn't process windowing events while it's busy; this means that the focus can't leave the window so the user is prevented from using any other app. IBM's rules for its programmers say that the programmer should take steps to avoid this. This clearly isn't the case for Apple since the scanning program that I used the other week locked the machine for a couple of minutes while the scanner was working)). imc From imc Mon Nov 11 11:54:02 1996 Subject: Re: We Must Have A Modem So There... To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 11:54:02 +0000 (GMT) In-Reply-To: <3520.199611111102@skegger.csv.warwick.ac.uk> from "Mr P R Walker" at Nov 11, 96 11:02:33 am X-Mailer: ELM [version 2.4 PL24 PGP6] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 388 Lines: 10 On Mon, 11 Nov 1996 11:02:33 +0000 (GMT), Mr P R Walker said: > Have you asked them? Have you pointed out that with an external modem, > they could also use it on any other computer they might have? If they already have another computer to use it on, why don't they just use that computer and not bother with the Sam?... (Not that I'm arguing for a proprietary modem, of course). imc From imc Mon Nov 11 11:56:54 1996 Subject: Re: Plea! Plea! Plea! + Bob's message To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 11:56:54 +0000 (GMT) In-Reply-To: <"runix.runi.540:10.11.96.20.37.35"@runix.runit.sintef.no> from "Simon Cooke" at Nov 10, 96 08:37:19 pm X-Mailer: ELM [version 2.4 PL24 PGP6] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 504 Lines: 12 On Sun, 10 Nov 1996 20:37:19 +0000 (GMT), Simon Cooke said: > The only problem left is demand-paging... which could be sorted if we had > a support processor -- just halt the other one while the support > processor goes off and gets the memory. OK as long as the processor isn't doing the refreshing. :-) So you are presumably saying that the support processor should be given all the details of how to find the required bit of memory, including if necessary which bit of disk to load it from. imc From owner-sam-users@nvg.unit.no Mon Nov 11 12:04:12 1996 Date: Mon, 11 Nov 1996 11:55:02 +0000 (GMT) From: Justin Skists X-Sender: c93js1@klein.cms.dmu.ac.uk To: SAM users list Subject: SNA-RUNner: IFF2 Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2541 Lines: 62 Hi there, At the moment, I've been trying to write a program on the SAM that would run .sna (48K Spectrum snapshots) on my SAM. And guess what, I've come across a small problem... According to the specs (that came with JPP), one of the bytes contain a bit that tell the state of IFF2. Correct me if I'm wrong, but this is how I understand NMI works:- Z80 detects NMI and sticks current PC onto stack. IFF1 gets copied into IFF2 (IFF1 says whether interupts are enabled when set) IFF1 is set to 0 (means disable interrupts) Z80 jumps to 66h and then jumps to where ever the routine is. RETN is decoded by Z80 at end of the NMI service routine. RETN puts IFF2 back into IFF1 and then returns using the last value on stack. I have absolutely no idea of how to set IFF2! So this is how I thought I could get around it:- di ; start of my program in Speccy rom at 3780h ; ; bits and bobs of getting registers how they should be before snapshot ; button was pressed; ; ld a, iff2_offset bit 2, a jr nz, no_ei ei ; assuming this sets IFF1 no_ei ; ; other bits in getting the right page in C&D, the right SP and other ; registers to the way the .sna says they should be. ; reti ; instead of retn so hopefully it would keep ; IFF1 set if the iff2_offset is set and reset ; if iff2_offset is reset. Guess what, it doesn't seem to work! Why do I get the feeling I'm missing something? About 60% games tested it works fine. The other 40% didn't do anything other than stay on the same screen as it was when it was snapped. Checking the other 40%, I discovered 90% of those has IFF2 set according to the specs. And the other 60% didn't. (Empire was the exception, but I could never even get it loaded on my SAM when I had the tape! So, I don't exactly trust it. Manic Miner was the other exception. Willy was jumping and jumping and jumping for no reason - reminded me of playing a game on teh speccy set to Kempston with out a joystick!) So.. Any ideas? I kinda need this because when I go away to do my Masters, I don't really think I'll have my own PC to play speccy games. It's my parent's, you see... :/ -- ============================================================================= |Justin Skists (c93js1@dmu.ac.uk) | Don't take life so seriously, | |BSc (Hons) Computer Science, Year 4 | you'll never survive it! | |De Montfort University, Leicester, England | - (C) CamLoRTS 1996 | ============================================================================= From owner-sam-users@nvg.unit.no Mon Nov 11 12:04:20 1996 Date: Mon, 11 Nov 1996 11:58:28 +0000 (gmt) From: "A.S. Collier" To: sam-users@nvg.unit.no Subject: Re: SOS: Multitasking In-Reply-To: <9611111152.AA00774@booth11.ecs.ox.ac.uk> Message-Id: X-X-Sender: ASC25@imap.hermes.cam.ac.uk Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 444 Lines: 15 On Mon, 11 Nov 1996 Ian.Collier@comlab.ox.ac.uk wrote: > (Alternatively you could just use the frame interrupts for multitasking > and decree that any program that disables the interrupts without it being > absolutely necessary is a "badly behaved" app. I've said I'd be doing something like that. But what do you do, if you think an app is badly behaved? (Apart from writing to the author and telling him the error of his ways) Andrew From owner-sam-users@nvg.unit.no Mon Nov 11 12:04:24 1996 Date: Mon, 11 Nov 1996 11:57:52 +0000 (GMT) From: Justin Skists X-Sender: c93js1@klein.cms.dmu.ac.uk To: sam-users@nvg.unit.no Subject: Re: Comms Interface In-Reply-To: <1.5.4.32.19961108172202.0093da20@nessie.mcc.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 936 Lines: 27 On Fri, 8 Nov 1996, Simon Cooke wrote: > There is a comms interface design available at: > > ftp.nvg.unit.no, > under: pub/sam-coupe/docs/gemini.zip (or gemini.ps) I've got this file. Does it work? (I mean, most hardware things I do from magazines, etc, don't work..) > If you've got a postscript printer, load it up and print it out ;) Or any Windows compatible printer id you use something like Ghostscript or PSView. > BTW: If you want to do anything commercial with it, ask me first. It's only > polite. Don't worry, we will.. :) -- ============================================================================= |Justin Skists (c93js1@dmu.ac.uk) | Don't take life so seriously, | |BSc (Hons) Computer Science, Year 4 | you'll never survive it! | |De Montfort University, Leicester, England | - (C) CamLoRTS 1996 | ============================================================================= From owner-sam-users@nvg.unit.no Mon Nov 11 12:04:54 1996 X-Authentication-Warning: dcsun4.comp.brad.ac.uk: lwillis owned process doing -bs Date: Mon, 11 Nov 1996 11:58:46 +0000 (GMT) From: Lee To: sam-users@nvg.unit.no Subject: Re: click-to-type? In-Reply-To: <9611110714.AA12726@asmal.edh-net> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 735 Lines: 24 On Mon, 11 Nov 1996, Frode Tenneboe wrote: ;>> > PS The one thing I hate about PC windowing systems is click-to-type. I like ;>> > to be able to type in which ever window contains the mouse pointer. ;>(I agree with Ian) Me too! ;>> ;>> Oh right, so you accidently nudge the mouse and all your keyboard input ;>> gets thrown away until you notice. Lovely! In fact, that's what I hate ;>> MOST about Linux. See what I mean about not being able to please any of ;>> the people for even some of the time? ;> ;>Why not make it configurable? Hmm, sounding like X-Windows again ... Lee. How many Microsoft employees does it take to screw in a light bulb? None, they just make darkness the industry standard! From owner-sam-users@nvg.unit.no Mon Nov 11 12:13:17 1996 Message-Id: X-Mailer: Novell GroupWise 4.1 Date: Mon, 11 Nov 1996 11:55:00 +0000 From: Dan Doore To: sam-users@nvg.unit.no Subject: Re: Nothing to do with SamSon - Reply Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 637 Lines: 20 > > Is there a ROM system call that I can call to act like the SCREEN command? > > Yes, I believe it's at 0100. Hmmm, according to the libraries [digs out scrap of paper], the display command calls &100, open_scr calls &157 (with a lot of stack pushing/poping beforehand) and close_scr calls &180. If this is the case then it would seem the display command does the equivelent of a SCREEN and DISPLAY at the same time, or is the library just wrong? :) The reason for all this is that I am writing something where I want the graphics to live on a hidden screen so that I can grab them and put them on the current screen. Dan. From owner-sam-users@nvg.unit.no Mon Nov 11 12:26:58 1996 From: Mr P R Walker Message-Id: <21341.199611111213@lupin.csv.warwick.ac.uk> Subject: Re: We Must Have A Modem So There... To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 12:13:48 +0000 (GMT) In-Reply-To: <9611111154.AA00795@booth11.ecs.ox.ac.uk> from "Ian.Collier@comlab.ox.ac.uk" at Nov 11, 96 11:54:02 am X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 400 Lines: 10 > If they already have another computer to use it on, why don't they just use > that computer and not bother with the Sam?... Nice, constructive remark there, Ian. So... I have a PC with a modem already. Should I not bother with Sam, then? HOw about Dave Whitmore? Dean liversidge? Even Bob? > (Not that I'm arguing for a proprietary modem, of course). (Not that I particularly care, of course.) From owner-sam-users@nvg.unit.no Mon Nov 11 12:46:13 1996 Message-Id: <1.5.4.32.19961111122746.0094c4c0@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 11 Nov 1996 12:27:46 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: Plea! Plea! Plea! + Bob's message Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 523 Lines: 13 At 11:56 AM 11/11/96 +0000, you wrote: >OK as long as the processor isn't doing the refreshing. :-) > >So you are presumably saying that the support processor should be given >all the details of how to find the required bit of memory, including if >necessary which bit of disk to load it from. Yep, that's pretty much the length and breadth of it... otherwise, I can't really see any way to get the paging done -- other than having some kind of paging command which is called, which would really slow things down. Simon From owner-sam-users@nvg.unit.no Mon Nov 11 12:48:09 1996 Message-Id: <1.5.4.32.19961111122747.0094861c@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 11 Nov 1996 12:27:47 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: Comms Interface Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 394 Lines: 14 At 11:57 AM 11/11/96 +0000, you wrote: >> There is a comms interface design available at: >> >> ftp.nvg.unit.no, >> under: pub/sam-coupe/docs/gemini.zip (or gemini.ps) > >I've got this file. Does it work? (I mean, most hardware things I do from >magazines, etc, don't work..) Strictly it should. I've not prototyped it though, so try it at your peril. Theoretically it's fine though. Simon From owner-sam-users@nvg.unit.no Mon Nov 11 12:51:54 1996 Message-Id: <1.5.4.32.19961111122749.00947f88@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 11 Nov 1996 12:27:49 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: We Must Have A Modem So There... Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 547 Lines: 16 At 11:54 AM 11/11/96 +0000, you wrote: >Status: > >On Mon, 11 Nov 1996 11:02:33 +0000 (GMT), Mr P R Walker said: >> Have you asked them? Have you pointed out that with an external modem, >> they could also use it on any other computer they might have? > >If they already have another computer to use it on, why don't they just use >that computer and not bother with the Sam?... Because (in my case) the comms software on my PC has a nasty habit of hanging every few minutes. Using my software on the SAM, there's no problem at all :) Simon From owner-sam-users@nvg.unit.no Mon Nov 11 12:57:39 1996 X-Authentication-Warning: dcsun4.comp.brad.ac.uk: lwillis owned process doing -bs Date: Mon, 11 Nov 1996 12:52:06 +0000 (GMT) From: Lee To: sam-users@nvg.unit.no Subject: Re: RCPT: What I would like in the SOS In-Reply-To: <87324BC7326@fs2.ee.umist.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 590 Lines: 23 On Mon, 11 Nov 1996, M.G. Smith wrote: ;>Confirmation of reading: your message - ;> ;> Date: 7 Nov 96 20:32 ;> To: sam-users@nvg.unit.no ;> Subject: What I would like in the SOS ;> ;>Was read at 12:45, 11 Nov 96. ;> Do I bloody care .... Why doesn't the mail program at least send this sort of message to the originator not the reply-to address!!!. (Not that I'm doing a project on mail formats and behaviour ...) Lee. How many Microsoft employees does it take to screw in a light bulb? None, they just make darkness the industry standard! From imc Mon Nov 11 13:02:46 1996 Subject: Re: Plea! Plea! Plea! + Bob's message To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 13:02:46 +0000 (GMT) In-Reply-To: from "Andrew M Gale" at Nov 10, 96 07:21:04 pm X-Mailer: ELM [version 2.4 PL24 PGP6] MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 656 Lines: 14 On Sun, 10 Nov 1996 19:21:04 +0000 (GMT), Andrew M Gale said: > So am I right in thinking that eacgh application thinks its > running on its own unique machine with a contiguous block > of memory stretching from 0 to .... er, some big number? Yes. Except in Unix the memory doesn't have to be contiguous. The stack grows down from F7FFFFFF and the program grows upward from 2000. Libraries seem to be loaded in just below the lowest possible position of the stack (which is calculated from the limit on stack size). Areas of memory between the program and the stack, and from 0-2000 remain invalid and trying to access them results in a signal. imc From owner-sam-users@nvg.unit.no Mon Nov 11 13:03:40 1996 Message-Id: X-Mailer: Novell GroupWise 4.1 Date: Mon, 11 Nov 1996 12:59:16 +0000 From: Dan Doore To: sam-users@nvg.unit.no Subject: RCPT: Re : Killer apps - Internet - Reply Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 97 Lines: 11 > Confirmation of reading: your message Please turn this off before we kill you. :) Dan. From owner-sam-users@nvg.unit.no Mon Nov 11 13:13:22 1996 From: sh5655@harrier.fen.bris.ac.uk (SL Harding) Message-Id: <199611111131.LAA06309@marsh.bris.ac.uk> Subject: Re: New Amigas To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 11:31:44 +0000 (GMT) In-Reply-To: <199610311825.SAA19481@pipe1.uk.pipeline.com> from "Samsboss@uk.pipeline.com" at Oct 31, 96 06:25:20 pm X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 694 Lines: 23 > > On Oct 31, 1996 17:18:30, 'Johnna Teare ' wrote: > > > > > >> Don't you think that SAM itself was designed and released too late in > the > >game? > >> A few years earlier and it would have been much more popular than Amigas > > >even. > > The question I would raise is "does SAM or SAMSON have to be as popular as > the Amigas?" > I think that part of the charm of SAM is that not everone has one - some of > us want to be different from the norm. > > > Samsboss > The One And Only > Should we start getting worried about the new Amiga that may (or may not) be released sometime early next year? I hear it looks like a toaster... Numb From owner-sam-users@nvg.unit.no Mon Nov 11 13:21:42 1996 From: James R Curry Organization: De Montfort University To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 13:23:13 GMT Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: Re: Tv or not TV Priority: normal X-Mailer: Pegasus Mail for Windows (v2.42a) Message-Id: <13C59330BB8@sahara.cms.dmu.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 387 Lines: 12 > It is important that all people can see a representation of all the initial > modes. In the original post I drew parallels with the way ST owners developed > software utils to do this. Did any one not recieve the original post? > Numb. I got it! -- James R Curry hc95jc@dmu.ac.uk "Augh! I can't think of any good quotes to put in this signature file" - James Curry. From owner-sam-users@nvg.unit.no Mon Nov 11 13:24:48 1996 From: James R Curry Organization: De Montfort University To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 13:26:39 GMT Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: Re: Plea! Plea! Plea! + Bob's message Priority: normal X-Mailer: Pegasus Mail for Windows (v2.42a) Message-Id: <13C67F76049@sahara.cms.dmu.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 458 Lines: 13 > continuing with Z80B WinDOS, when I may soon be starting Z380 WinDOS? > Nobody's replied yet... Hold on a second.. If the Z380 uses virtualy the same instructions as the Z80B but faster, couldn't you simply finish the Z80B version, and then upgrade this code to do more on the faster Z380? Or am I missing something here? -- James R Curry hc95jc@dmu.ac.uk "Augh! I can't think of any good quotes to put in this signature file" - James Curry. From owner-sam-users@nvg.unit.no Mon Nov 11 13:57:49 1996 From: James R Curry Organization: De Montfort University To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 13:37:43 GMT Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: Re: Tv or not TV Priority: normal X-Mailer: Pegasus Mail for Windows (v2.42a) Message-Id: <13C96CB0CFE@sahara.cms.dmu.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 493 Lines: 17 > >We need to let the user go slowly up the upgrade path. No point saying that > >SAMson will work on a TV if none of the new applications will run on the > >TV... > > > >-- > > Good point, back to the drawing board. Surely it would be possible for applications to run in all modes (Well, maybe not modes 1+2) and just work BETTER in the better graphics modes.. -- James R Curry hc95jc@dmu.ac.uk "Augh! I can't think of any good quotes to put in this signature file" - James Curry. From owner-sam-users@nvg.unit.no Mon Nov 11 14:07:56 1996 Date: Mon, 11 Nov 1996 14:47:17 +0100 From: ft@edh.ericsson.se (Frode Tenneboe) Message-Id: <9611111347.AA13373@asmal.edh-net> To: sam-users@nvg.unit.no Subject: Re: Tv or not TV X-Sun-Charset: US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 238 Lines: 8 > Surely it would be possible for applications to run in all modes > (Well, maybe not modes 1+2) and just work BETTER in the better > graphics modes.. Yes! Why don't we design the video-modes based on vector display system? ;) -Frode From owner-sam-users@nvg.unit.no Mon Nov 11 14:11:31 1996 Date: Mon, 11 Nov 1996 15:01:45 +0100 From: ft@edh.ericsson.se (Frode Tenneboe) Message-Id: <9611111401.AA14149@asmal.edh-net> To: sam-users@nvg.unit.no Subject: Unsub. and a quote X-Sun-Charset: US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 668 Lines: 22 Hi fellow SAM'ers. I'm sorry to inform you that David Gommeren (LI, of Tetris-fame) just wrote me saying that he has unsubscribed from the list due to the heavy traffic recently. Specifickly he told me to say: "A decent successor of the Sam needs: An ARM 7100/7500 because It's fast It's a one chip computer (add memory and it will work) It's easy to program It's cheap Throw away the old hardware, nobody needs it People complaining about the need for a Z80/Z380 to make it a real Son of Sam, should now this: to be a real Son of Sam it should feel and work (from the outside) like a Sam, who cares about what's inside?" -Frode From owner-sam-users@nvg.unit.no Mon Nov 11 14:18:03 1996 From: Johnna Teare Organization: University of Central Lancashire To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 14:05:35 GMT+0 Subject: Re: WinDOS Priority: normal X-Mailer: Pegasus Mail for Windows (v2.33) Message-Id: <5A4639000E@mail-gw.uclan.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 585 Lines: 14 > > PS The one thing I hate about PC windowing systems is click-to-type. I like > > to be able to type in which ever window contains the mouse pointer. > > Oh right, so you accidently nudge the mouse and all your keyboard input > gets thrown away until you notice. Lovely! In fact, that's what I hate > MOST about Linux. See what I mean about not being able to please any of > the people for even some of the time? > As long as you can toggle between the text windows with the TAB key, I don't see a problem! Johnna Pig Teare (j.d.teare@uclan.ac.uk) "They call me Mad The Swine" From owner-sam-users@nvg.unit.no Mon Nov 11 14:32:40 1996 From: Johnna Teare Organization: University of Central Lancashire To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 14:22:18 GMT+0 Subject: Re: Return to Gender Priority: normal X-Mailer: Pegasus Mail for Windows (v2.33) Message-Id: <5A865713AD@mail-gw.uclan.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 296 Lines: 12 Hello again, > 'Scuse me being ignorant, but are you a bloke or girly, Johnna's a name I > ain't heard before .. Not sure. We havn't got around to studying anatomy yet. We're still learning teh primary colours :) > Lee. Johnna Pig Teare (j.d.teare@uclan.ac.uk) "They call me Mad The Swine" From owner-sam-users@nvg.unit.no Mon Nov 11 14:33:21 1996 From: Johnna Teare Organization: University of Central Lancashire To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 14:26:18 GMT+0 Subject: Re: Thoughts on release Priority: normal X-Mailer: Pegasus Mail for Windows (v2.33) Message-Id: <5A86570D74@mail-gw.uclan.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1867 Lines: 41 From: Johnna Teare Organization: University of Central Lancashire To: nvg.unit.org!sam-users@unit.edu - erm this bit is wrong methinks! Date sent: Sun, 10 Nov 1996 12:38:22 GMT+0 Subject: SOS: Deadlines Priority: normal Just a thought, but if we plan to create any kind of commercial market for the SAMSon then we must really have it out by next christmas - maybe not the completed model, but certainly some of teh upgrade boards for SAM. The longer we leave it, the more users we'll lose and the more the computer industry will have moved on. And I think SAMSon is a pretty cool name...although there is bound to be legal implication with electronics giant SAMSUNG because they sound pretty similar. Any other suggestions for names? Sooner or later we'll need a dedicated specification - perhaps Bob could write it down. And soon after that we need to start writing software, maybe just on the SAM at first to test ideas. I can see a lot of good coming out of this project already - it's certainly inspired the list in the last few weeks. Bob, do you intend to announce the plans to the SAM public on the whole yet, or are you going to wait for something to emerge first? The reason I'm asking is that I believe the SAM scene needs a little kick at the moment - something to convince users who might be vernturing towards buying PC's this Christmas to keep the faith with the SAM for just a short while longer. Not many people who pay stlg1000 for a machine this christmas are going to want to spend money on new boards for their SAM in the Summer and we might lose some of our market here. Oh, and Congratulations to Evander Holyfield. Johnna Pig Teare (j.d.teare@uclan.ac.uk) "They call me Mad The Swine" Johnna Pig Teare (j.d.teare@uclan.ac.uk) "They call me Mad The Swine" From owner-sam-users@nvg.unit.no Mon Nov 11 14:36:09 1996 Message-Id: <3287A9AF.17F2@csv.warwick.ac.uk> Date: Mon, 11 Nov 1996 14:33:19 -0800 From: Paul Walker Organization: University of Warwick X-Mailer: Mozilla 2.0 (Win16; I) Mime-Version: 1.0 To: sam-users@nvg.unit.no Subject: Re: Re : Killer apps - Internet References: <1C500E4B9F@smserver1.ulst.ac.uk> <199611091134.LAA12993@mail.enterprise.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 488 Lines: 12 Dave Whitmore wrote: > Fidonet in one package. Or, we could do the Fido first and bolt on internet > email later. This would be better, and probably more attainable as well. When we finally get a decent version of C for the Sam, I've got the source code for LHA, so once someone's altered the assembler we've got one compressor straight away. > GIF & JPEG decoders, proper ASCII text editors.. oh the list goes on and > gets more scary the more I think of it. :/ Daunting, isn't it? From owner-sam-users@nvg.unit.no Mon Nov 11 14:40:15 1996 From: Johnna Teare Organization: University of Central Lancashire To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 14:30:34 GMT+0 Subject: Re: We Must Have A Modem So There... Priority: normal X-Mailer: Pegasus Mail for Windows (v2.33) Message-Id: <5AA3EC7166@mail-gw.uclan.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1299 Lines: 30 > > Very few of us would be prevented from using an 'un-approved' modem because I > > would imagine that most of us now have the modern style Telephone plugs. > > BTW, the last time I read that warning it said "direct or indirect connection", > I think. I'm not /sure/ about that tho. > > > Dave also said: > > >Me too! TBH, I wouldn't touch one if that were the case. > > Same here. Basic demand and supply should dictate that producing a SAMSon specific modem is completely impractical if it is possible to use a standard PC one. There are millions of PC's out there and eventually, most of tehm will have modems, thus creating a large competitive market for PC Modems. If we produce a SAM specific one, we need to fund developemtn, prodcution and marketing for it ourselves. Why bother? Why not just buy a standard PC one off the shelf and leave all that nonsense to them? If we need to set a stadnard SAM baud rate then make it V32 if it could cope with this. If tehre is a large demand for an internal modem, why can't we just use one of the internal PC ones? Sorry, all questions and no answers, but I really think that a SAM specific one is just creating work for us that we don't need to or have to do. Johnna Pig Teare (j.d.teare@uclan.ac.uk) "They call me Mad The Swine" From owner-sam-users@nvg.unit.no Mon Nov 11 15:01:22 1996 Date: Mon, 11 Nov 1996 09:56:49 -0500 From: BrenchleyR@aol.com Message-Id: <961111095648_1915477591@emout11.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Plea! Plea! Plea! + Bob's message Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 387 Lines: 15 In a message dated 10/11/96 18:51:52, you write: >I need one of those manuals! Bob, Mr Organiser man, do you want to order a >dozen or so for all the programmers, and we'll pay for the postage when >you pass them on? Saves the trouble of thinking up a better excuse than >fifteen Z380-controlled voice-operated telephone exchanges ;-) > > >Andrew > > Will see what we can do. Bob. From owner-sam-users@nvg.unit.no Mon Nov 11 15:12:46 1996 From: James R Curry Organization: De Montfort University To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 15:13:56 GMT Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: RE: something or another. Priority: normal X-Mailer: Pegasus Mail for Windows (v2.42a) Message-Id: <13E31EA1F52@sahara.cms.dmu.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 537 Lines: 14 > Whatever happens with comms on SAM. we'll need a ZIPer and an > UnZIPer, LHA, GIF & JPEG decoders, proper ASCII text editors.. oh > the list goes on and gets more scary the more I think of it. :/ Hmm, aren't there plenty of those that have been written in C..? That could be compiled on, oh, for instance, just off the top of my head... an ANSI C COMPILER? *James sits back and awaits the arguments...* -- James R Curry hc95jc@dmu.ac.uk "Augh! I can't think of any good quotes to put in this signature file" - James Curry. From owner-sam-users@nvg.unit.no Mon Nov 11 15:17:51 1996 From: Gavin Smith Organization: University of Ulster To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 15:11:29 GMT Subject: Re: New Amigas X-Confirm-Reading-To: "Gavin Smith" X-Pmrqc: 1 Priority: normal X-Mailer: Pegasus Mail for Windows (v2.23) Message-Id: <46961A317E@smserver1.ulst.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 662 Lines: 12 > Should we start getting worried about the new Amiga that may (or may not) be > released sometime early next year? I hear it looks like a toaster... Numb The "toaster" shaped Amiga won't be released now, since Escom went bankrupt. In case you haven't heard, Viscorp bought the rights to the Amiga during the summer, mainly for their "set-top internet boxes". Viscorp apparently have plans for a Power-PC based Amiga, but are taking so long about it that the Amiga community are beginning to get worried if it will come out at all...Anyway, if it does come out, it will priced around 1200 quid so I wouldn't worry too much about it. Gavin Smith From owner-sam-users@nvg.unit.no Mon Nov 11 15:58:50 1996 Date: Mon, 11 Nov 1996 10:41:25 -0500 From: BrenchleyR@aol.com Message-Id: <961111104123_1418307185@emout20.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Nothing to do with SamSon Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 610 Lines: 27 In a message dated 11/11/96 11:23:44, you write: >Dear All, > >I'm currently playing around in Sam C and I want to do something that I could >do in >Basic, but currently cannot in Sam C. > >In Sam C there are functions for open_scr, close_scr and display, but no >function >for the SCREEN command. > >The library calls for display (and such) are simple LD c,l and then CALL &xxx >(not >that I understand what it's doing - but I can guess :) > >Soo... > >Is there a ROM system call that I can call to act like the SCREEN command? > >Dan. The command 'display (int s); makes screen 's' the active one. Bob. From owner-sam-users@nvg.unit.no Mon Nov 11 15:58:50 1996 Date: Mon, 11 Nov 1996 10:41:28 -0500 From: BrenchleyR@aol.com Message-Id: <961111104127_1584553613@emout02.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: Thoughts on release Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 2270 Lines: 50 In a message dated 11/11/96 14:29:31, you write: >Just a thought, but if we plan to create any kind of commercial >market for the SAMSon then we must really have it out by next >christmas - maybe not the completed model, but certainly some of teh >upgrade boards for SAM. The longer we leave it, the more users we'll >lose and the more the computer industry will have moved on. Agreed, if there is not big movement by next christmas then a lot of the 'early' potential buyers just may fall to the glitz of the Pentium. > >And I think SAMSon is a pretty cool name...although there is bound to >be legal implication with electronics giant SAMSUNG because they sound >pretty similar. Any other suggestions for names? > >Sooner or later we'll need a dedicated specification - perhaps Bob >could write it down. And soon after that we need to start writing >software, maybe just on the SAM at first to test ideas. I can see a >lot of good coming out of this project already - it's certainly >inspired the list in the last few weeks. > >Bob, do you intend to announce the plans to the SAM public on the >whole yet, or are you going to wait for something to emerge first? The >reason I'm asking is that I believe the SAM scene needs a little kick >at the moment - something to convince users who might be vernturing >towards buying PC's this Christmas to keep the faith with the SAM for >just a short while longer. Not many people who pay stlg1000 for a >machine this christmas are going to want to spend money on new boards >for their SAM in the Summer and we might lose some of our market here. Some news in the November issue (going out late this week) More news in the December issue. But no, for the time being, I don't want to mention SAMSON. If I did it would totaly stop sales of the Elite and recon Coupe. That would kill the project dead befor it even got started. I want to start pushing the new boards as soon as we can agree on the specs. The Clock/Calender board is now commited to. There is a 'full buffer' board in design (all address, data and most of the control lines buffered). What comes next will be up to you lot. Bob. > >Oh, and Congratulations to Evander Holyfield. > >Johnna Pig Teare (j.d.teare@uclan.ac.uk) >"They call me Mad The Swine" > > From owner-sam-users@nvg.unit.no Mon Nov 11 15:59:20 1996 Date: Mon, 11 Nov 1996 10:41:24 -0500 From: BrenchleyR@aol.com Message-Id: <961111104122_1383858061@emout19.mail.aol.com> To: sam-users@nvg.unit.no Subject: Re: We Must Have A Modem So There... Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1498 Lines: 45 In a message dated 11/11/96 11:19:23, you write: >At 11:02 AM 11/11/96 +0000, you wrote: >>> Given a choice I think that most people would go for internal, even if >there >>> was not a money saving involved. Yes, by all means, lets have a fast >serial >> >>Have you asked them? Have you pointed out that with an external modem, >>they could also use it on any other computer they might have? > You don't haveto ask people, you look at sales. Where the choice exists (in the PC field) the big majority are going for internal modems - that has to say something to you. >One thing I think I should point out here: > >A SAM specific modem is going to cost ~40 UKP minimum to implement. Maybe >more in the ~60 UKP range. > >For that money, you're going to get a modem which runs at 2400 baud -- if >you're lucky. More likely it'll run at 1200 baud. What makes you say that? > >For a similar price you can get a 9600 baud modem. For 30 pounds more, you >can get a 14400 baud one. I would have thought that 14400 would be the minimum worth implimenting. Better still, 28800. > >Why go for a custom-made modem, when surely the best option all-round is an >external one? The only good reason given so far for using an external is the ability to connect it to another computer. A valid reason I agree, but not a good enough reason not to produce the internal one. > >All we need now is a /GOOD/ comms interface. Can't look at your design Simon, I have nothing that will look at a .PS file. > >Simon From owner-sam-users@nvg.unit.no Mon Nov 11 15:59:55 1996 From: James R Curry Organization: De Montfort University To: sam-users@nvg.unit.no Date: Mon, 11 Nov 1996 15:47:18 GMT Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: Re: Plea! Plea! Plea! Priority: normal X-Mailer: Pegasus Mail for Windows (v2.42a) Message-Id: <13EBFB11386@sahara.cms.dmu.ac.uk> Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 532 Lines: 23 . > >Can't wait for the accelerator to be finished? Then use this > >wonderful program to create similar screen effects. > > > >10 CLS#:MODE 4 > >20 CSIZE 8,8 > >30 PLOT INT(RND*256),INT(RND*176) > >40 GOTO 20 > > > >Okay, so it's not as fast as the accelerator.. But then if it was it > >would lose sales. ;) > *laughs* > > Excellent!!!! > > Simon Going in the next issue of BOAI then, is it? :) -- James R Curry hc95jc@dmu.ac.uk "Augh! I can't think of any good quotes to put in this signature file" - James Curry. From owner-sam-users@nvg.unit.no Mon Nov 11 16:19:05 1996 Message-Id: <1.5.4.32.19961111161213.00944214@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 11 Nov 1996 16:12:13 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: We Must Have A Modem So There... Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 1500 Lines: 41 At 10:41 AM 11/11/96 -0500, you wrote: >>A SAM specific modem is going to cost ~40 UKP minimum to implement. Maybe >>more in the ~60 UKP range. >> >>For that money, you're going to get a modem which runs at 2400 baud -- if >>you're lucky. More likely it'll run at 1200 baud. > >What makes you say that? Well, looking at the available components in the electronics manuals I have around, I can't find anything which can output signals above 2400 baud. And believe me, the design of a circuit to do anything faster isn't trivial. >>For a similar price you can get a 9600 baud modem. For 30 pounds more, you >>can get a 14400 baud one. > >I would have thought that 14400 would be the minimum worth implimenting. >Better still, 28800. This is the thing; you can buy 14400 baud modems off the shelf for much cheaper than you'd be able to build them yourself... >>Why go for a custom-made modem, when surely the best option all-round is an >>external one? > >The only good reason given so far for using an external is the ability to >connect it to another computer. A valid reason I agree, but not a good enough >reason not to produce the internal one. Okay, in that case: cost. It'll cost you far far more to produce an internal SAM modem (not that it /would/ be internal) than it would to buy one off the shelf. >>All we need now is a /GOOD/ comms interface. > >Can't look at your design Simon, I have nothing that will look at a .PS file. Have printed it out and will pop it in the post. Simon From owner-sam-users@nvg.unit.no Mon Nov 11 16:19:06 1996 Message-Id: <3287C0B9.6FFD@csv.warwick.ac.uk> Date: Mon, 11 Nov 1996 16:11:37 -0800 From: Paul Walker Organization: University of Warwick X-Mailer: Mozilla 2.0 (Win16; I) Mime-Version: 1.0 To: sam-users@nvg.unit.no Subject: Re: something or another. References: <13E31EA1F52@sahara.cms.dmu.ac.uk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 373 Lines: 13 James R Curry wrote: > > Hmm, aren't there plenty of those that have been written in C..? That > could be compiled on, oh, for instance, just off the top of my head... > an ANSI C COMPILER? Yes there are, but until we get a compiler which implements files properly, they'll all need major rewriting. Paul ... He's dead, Jim. You get his phaser, I'll get his wallet. From owner-sam-users@nvg.unit.no Mon Nov 11 16:33:24 1996 Message-Id: <1.5.4.32.19961111162708.0095096c@nessie.mcc.ac.uk> X-Sender: scooke@nessie.mcc.ac.uk X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 11 Nov 1996 16:27:08 +0000 To: sam-users@nvg.unit.no From: Simon Cooke Subject: Re: Thoughts on release Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 433 Lines: 12 At 10:41 AM 11/11/96 -0500, you wrote: >I want to start pushing the new boards as soon as we can agree on the specs. >The Clock/Calender board is now commited to. There is a 'full buffer' board >in design (all address, data and most of the control lines buffered). What >comes next will be up to you lot. ARGGGHHHH!!!!! PLEASE TELL ME THAT THIS BOARD USES THE DBDIR LINE!!! Otherwise, we all may as well stop now, folks... Simon From owner-sam-users@nvg.unit.no Mon Nov 11 16:46:37 1996 Date: Mon, 11 Nov 1996 16:35:03 +0000 (GMT) From: Tim Wells To: sam-users@nvg.unit.no Subject: Re: SNA-RUNner: IFF2 In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 493 Lines: 14 > At the moment, I've been trying to write a program on the SAM that would run > .sna (48K Spectrum snapshots) on my SAM. And guess what, I've come across a > small problem... I'm sure someone can sort out that problem, but do you need to write a new program to do this - because: There is a program on nvg which will convert .sna to the +D/SAM snapshot format. That will then load into most of the emulators around. I think the program is call Convrt.pak, and is in \utils\misc Tim W. From owner-sam-users@nvg.unit.no Mon Nov 11 17:17:46 1996 X-Authentication-Warning: dcsun4.comp.brad.ac.uk: lwillis owned process doing -bs Date: Mon, 11 Nov 1996 17:15:30 +0000 (GMT) From: Uncle Bulgaria To: sam-users@nvg.unit.no Subject: Re: Return to Gender In-Reply-To: <5A865713AD@mail-gw.uclan.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: sam-users@nvg.unit.no Status: RO Content-Length: 335 Lines: 16 On Mon, 11 Nov 1996, Johnna Teare wrote: ;>Not sure. We havn't got around to studying anatomy yet. We're still ;>learning teh primary colours :) And spelling presumably ... ;) Lee. How many Microsoft employees does it take to screw in a light bulb? None, they just make darkness the industry standard!