USE CURSOR KEYS TO SCROLL TEXT. USE CURSOR KEYS TO SCROLL TEXT. JANET-derived Beeb Information From the BBC Mailing List (bbc-list-request@ohm.york.ac.uk) Supplied to 8-Bit Software by: Daniel Shimmin, Queen's College, OXFORD (qu96001@ox.ac.uk) Some of you will know that most (all?) the universities in this country (and the world...) are connected by electronic networks (Unix and things). Naturally computer enthusiasts tend to exchange information through these networks, and, not surprisingly, there is a hard core of BBC enthusiasts still in existence (they appear to be very interested in the 32016 co-processor as well, as it provides serious programming facilities for your Beeb). So anyway, here is such information as has reached me, hopefully it will be of use to some people out there (it's all public domain by definition of course, since it's been beamed around much of the world) although it's mostly rather technical for me. Very interesting though, there seems to be a wealth of information and knowledgeable people out there. If anyone wants to get in contact with any of these strange and distant people, let me know via 8BS. ================ From : Dave (terros@sideways.welly.gen.nz) I've upgraded my hard drive from 8 megs to 20 megs. However, the new drive has a track step time some 15 times faster than the old! Time for a modified version of ADFS to deal with this! My hard drive boot program makes a copy of the (Master) ADFS rom into a sideways RAM bank and modifies it, adding some code to set up the hard drive specs every time Break is pressed (my hard drive controller board defaults to the correct number of heads, but the wrong number of cylinders). This may be useless to all, as my setup is not quite Acorn specification (the SCSI->MFM controller board is a Seagate device, rather than Adaptec, and has head stepping options that may not even exist on the Adaptec board). It'll really be of most use to people trying to hack their own hard drive systems together. Further to that, Acorn were kind enough to send me the technical details of their interface (1Mhz->SCSI), so of anyone needs the info on an ADFS compatible interface (i.e. the circuit diagram), I have this here. (DGS sez: Does this mean he has the info to build a SCSI interface that can just be plugged into the 1MHz bus? If so, you could build one for next to nothing, then just pick up a cheap second-hand SCSI drive (they're everywhere, very cheap), and you would have a big HD system far cheaper than anything else you could buy...any thoughts anyone?) ================ From : Mik Davis at Aston Uni (davism@uhura.aston.ac.uk) Does anybody out there know anything about writing Econet printer servers software? I've been looking at this recently and I thought of the following points: (1) I can pick up packets from the net and print them but (a) how much data is in each packet, (b) how do I know when I have the last one? (2) What do I do if another station requests service whilst I'm still servicing a previous request? (3) Do I have to respond to a client doing a *PS command, and if so, how? I did post a message asking about Econet printer servers on the net, but all I got was a message from Acorn saying "You can buy one from us for forty pounds". ================ From : Walter Meyer, Physics Dpt., Pretoria Univ., S. Africa (wmeyer@scinet.up.ac.za) A local school has a number of BBC Model B's with Amcom DFS's (using the 8271 FDC from Intel). Recently one of the FDC chips blew. Since they would like to keep the BBC's as similar as possible, and would also like to continue to use their Amcom DFS disks (which seem to be incompatible with anything else), they are not keen to upgrade to a 1770 DFS. I understand the 8271 is no longer manufactured by Intel, but, according to the local Intel agents, an equivalent is manufactured by Intel under a different part number. Does anyone know what this part number is or whether it or any equivalent by another manufacturer exists? Any help would be appreciated. ================= Next, a few words of explanation. Firstly, anyone putting the characters :-) in their message is indicating that whatever comes immediately before it was meant to be sarcastic, ironic, funny, not to be taken literally etc. (Because if you look at those three characters from the side, it looks like a smiling face.) Similarly the characters :-( mean "oh dear" or "how sad" or something. Secondly, many people begin their messages by quoting whichever message they are replying or responding to (because it helps people to remember what the original subject of the message was), assisted by software that provides for this. Normally, a repeated message is indicated by adding ">" at the start of each line, but I've used quotes to make things easier. Thus you can see (hopefully) how one message follows another. (Read the following, which starts with a reply to someone else's message (quoting it first), then continues with someone else adding further comment to the reply (quoting it first), etc. and you might understand) ================== From : Ian Stephenson (ian@ohm.york.ac.uk) "After revivig an old project, I hit upon the solution (made easier by the fact that I now have a Master with Shadow Ram, which I didn't have before). I've successfully doubled the vertical resolution of the bitmapped screen modes, giving a maximum of 640x512 in Mode 0. "Having done this, I haven't any idea what to do with it! :-) "The refresh rate is effectively halved, which seems to be too slow to stop a certain amount of flicker under the wrong conditions (isolated pixels flicker). But it's not bad. Text looks good. "Anyone got any ideas for this?" Excellent! I figure this works by switching the banks between the interlace, so lines from each bank get interleaved(?). I had wondered if this was possible, but don't have shadow RAM to try it with (if this works well enough, I might get some!). Once selected how well does it sit with the rest of the system (i.e. is it transparent?), or are special drivers needed? Will it work with any shadow board? The obvious thing would be to develop some sort of windowing environment (the old resolution wasn't really enough for this). Once I get BCPL running I was planning to add some O-O (DGS: I think that might stand for Object-Orientated) (as in Objective C) - I consider such things essential for windowing. It would then be pretty easy to get something like the original Mac finder up and running. Have you tried converting the GIF viewer to it? Mode 0 stuff looked pretty good - twice the resolution would be excellent. Keep the text the same physical size, but write some font drivers. Double resolution would allow you to add serifs and other cute features without overly crowding things. Acorn failed to see the graphics revolution, and hence supprt for the above stuff is limited. Therefore it's probably something worth investigating further. ============= From : Martin Ebourne, Electronics, S'hampton Uni (mje@soton.ac.uk) "Excellent!...(DGS - he then repeats the above paragraph)...Will it work with any shadow board?" Some bad news for you...this is Master only. Beeb shadow RAM boards (at least, the ones I've seen) are completely different in operation. The problem is that the 6845 is wired into the main RAM, and needs to be because it is that which carries out the refresh. Hence the screen is always displayed from main RAM. If you put shadow mode in, the main RAM is paged out and shadow RAM paged in, so all your data is stored in there. Then OSWRCH is trapped and the main RAM paged in before and out again after. There is no physical way of displaying the screen from the shadow RAM. I know this because I tried to do it, and then had to disassemble the controlling softwareto find out why it didn't work. Obvious really. What confused me is that the manual stated clearly that the screen was displayed from shadow RAM, but they are definitely lying :-( On the bright side, Master 128's are quite cheap now... Ask Richard Whitehand (who should be listening in to this...), he's got half a dozen and they are currently in his bad books, so he's bound to be keen to get rid of them. :-) ============== From : Dmitry Petrov, Moscow State Uni, via Russian ARM Support Group (DPetrov@misis.msk.su) Hello! I have a DOS filing system for BBC B+/Master. It allows you to read/write standard 360K PC disks. (DGS: wow! :-) ) If anybody interested I could send you the DOSFS by e-mail. Also I have some other utilities for Econet, RAM FS. I am interested in getting communication programs for BBC/Master (allowing 8-bit character set and X, Y, Zmodem or Kermit protocols). ============== From : Richard Whitehand (R.W.F.Whitehand@uk.ac.loughborough), Challenger BBS, Fast Viewdata, v23/22/22bis/MNP2-5 - 24hrs - 021 445 3913 "I am interested in getting communications programs for BBC/Master" I have a whole set of these on my bulletin board. Your best bet is the "Aterm" series by Gareth Babb that do black&white ANSI or 4-colour ANSI terminal emulation, Xmodem, Xmodem-1k, Ymodem, Ymodem-batch and Ymodem-batch-g protocols...the terminal is pretty fast too...should keep up to at least 9600 baud without problems... I don't think anyone has written Zmodem for the Beeb yet...besides, if you've got a reliable link (e.g. an MNP modem) then Ymodem-batch-g is just as fast in most circumstances. ============== OK that is about enough for now. If there is any interest in this I have lots more on a whole variety of subjects (about twelve times as much text as that above, and getting more all the time), which can be included in future 8BS issues. Daniel Shimmin.