Fixing a Superbrain up

Talk about non-Acorn classic computers/hardware/software here (including retro consoles)
StephenM
Posts: 19
Joined: Mon Jun 27, 2016 7:54 pm

Re: Fixing a Superbrain up

Postby StephenM » Sun Jun 11, 2017 2:48 pm

jonb wrote:
hoglet wrote:
What's bothering me here is that PIP is crashing all the time when copying from A: to B:. Yet the memory is fine.. so I am beginning to wonder if it is caused by the ROM (which has all the FDC routines in it) having a 40 track setup but the CP/M BIOS only using 35. These are known things with the Superbrain - from the factory the machine did 35 tracks, but there were 3rd party BIOSes that allowed 40. The Tandon drives are 40T DS DD or SS DD, but still 40 tracks. Hmm.

.


Did you change the EPROM code to allow 40 tracks?

It could be that a mismatch with CP/M is causing a problem but unlikely the EPROM code is over written when the system tracks are read from the boot disk. Try changing the EPROM back to 35 tracks and rule it out.
The problem with the 3rd party BIOSes was the incompatibilty issues between machines a bit like the Kaypro lineup.

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Mon Jun 12, 2017 7:34 am

HI Stephen

I have the original ACT boot ROM and a Standard QD v3.1 ROM (from Bitsavers / VCF) and they both exhibit his odd behaviour. Funnily enough (I may have said this before), it's got a slower seek rate than the ACT ROM. I agree there could be a mismatch between this and the CP/M BIOS. I am using the Compustar 30 CP/M image, which I believe is the same as a Superbrain QD from the disk perspective (35 tracks, DS/DD). I've also found a Superbrain boot disk (boot message says "Superbrain" instead of "Compustar 30") but that doesn't work either.

I have several formatting programs (one of which offers different track counts) and none of them work.

Hmm, now might be a good time to look at that ROM disassembly..

[edit: I have a Superbrain II QD boot image too, although it likes to reset the screen to gibberish with every keypress (It's setting the graphics mode on the CRT chip). It won't format, either...]

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Mon Jun 12, 2017 8:30 am

Update:

See this thread on VCF: http://www.vcfed.org/forum/showthread.p ... -ROM-Dumps

One of the images shows a Superbrain running the Compustar 30 boot disk (same as I have here) with the Bitsavers ROM v3.1 (again, same as I have here). Although he doesn't say if it is writing / formatting disks correctly, I would assume it is because there were no further posts about it.

StephenM
Posts: 19
Joined: Mon Jun 27, 2016 7:54 pm

Re: Fixing a Superbrain up

Postby StephenM » Mon Jun 12, 2017 10:18 am

If you look through the updated zip of Superbrain images that Larry created you will find a SBRAIN32 folder. This is the image for a QD with 64k of RAM. Try that on your hxc drive as boot. This is the one I'm having the most success with.
Have we asertained that yours is a QD or SD?

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Mon Jun 12, 2017 11:45 am

I think it is a SD that was upgraded to QD spec. The floppy drive is double sided (although it doesn't work!)

[edit.. well, it appears that this boot disk (which insists on setting the screen up badly) can copy files, but it cannot format a disk properly. More experiments needed methinks....]

StephenM
Posts: 19
Joined: Mon Jun 27, 2016 7:54 pm

Re: Fixing a Superbrain up

Postby StephenM » Mon Jun 12, 2017 12:18 pm

jonb wrote:I think it is a SD that was upgraded to QD spec. The floppy drive is double sided (although it doesn't work!)


You have fallen into the same trap as I did when identifying what machine you have.

The line up was JR (180k) QD (360k) and SD (720k)

That is the disk capacity per drive, a lot of marketing documentation calls the QD as 720k disk capacity but that is 2 x 360k floppy drives.

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Mon Jun 12, 2017 1:05 pm

The Tandon drive is definitely double sided though. Must be 360k.. maybe I should attempt SD format. The machine is a "Superbrain" (not QD or SD according to the label). [Edit - SD not working either!]

Meanwhile...

I did PIP B:=A:*.* to test copying. It looked like it was working, then crashed on the last file. The directory on B: is correct, but attempting to load MBASIC and do some tests resulted in a crash when I tried to go back to the CCP with SYSTEM. Kind of like a memory issue, but I know the RAM on here is all good (tested it several times).

Then I found that MBASIC is corrupted on the source disk. #-o

Still, PIP is crashing on the last file, and format is still not working.

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Mon Jun 12, 2017 5:52 pm

Stephen

Still trying these images that Larry sent. Latest corrected versions.

There are various BIOSes present:

  • CBOOT: Compustar Model 30 v0.2 - 35 tracks, DS
  • CSR30ENH: Compustar Model 30 v0.2 - 42 tracks, SS
  • CSR30NON: No boot track - 42 tracks, SS
  • CSR-COMM: Compustar Model 30 v0.3 - 40 tracks, DS
  • CSYSGEN: Compustar Model 30 v0.2 - 35 tracks, DS
  • CUTILS: Compustar Model 30 v0.2 - 35 tracks, DS
  • QDBIOS4: BIOS v4.0x17 (UNRELEASED!) - 40 tracks, DS
  • QDHDBIOS: CMC QD Super5 DOS v1.1 - 40 tracks, DS *** This contains information and BIOSes for the ACT HDD interface ***
  • QD-UTILS: CMC Superbrain / Targa DOS v1.0 - 40 tracks, DS
  • QD-ZCPR3: ZCPR - 40 tracks, DS
  • ZMAST: ZCPR OS - 40 tracks, DS
  • SBRAIN32: 64K Superbrain DOS v3.0 - 40 tracks, DS
The drive geometry is as reported by the HxC emulator software. None of these images run properly on a 50Hz machine, and quite a few of them set the CRT8002 to a graphic character set. To correct this we need a short program that restores the CRT settings after boot by doing OUT 68,3 (I'm using the ICE for this at the moment).

I'm pretty sure that the image with 42 tracks is corrupt.

StephenM
Posts: 19
Joined: Mon Jun 27, 2016 7:54 pm

Re: Fixing a Superbrain up

Postby StephenM » Tue Jun 13, 2017 7:45 am

StephenM wrote:
Did you change the EPROM code to allow 40 tracks?

It could be that a mismatch with CP/M is causing a problem but unlikely the EPROM code is over written when the system tracks are read from the boot disk. Try changing the EPROM back to 35 tracks and rule it out.
The problem with the 3rd party BIOSes was the incompatibilty issues between machines a bit like the Kaypro lineup.


I may have been a little hasty with the above.

I was correct in implying the EPROM code is over written but only the top half is lost. The bottom half is where the FDC routines reside and where CP/M calls them.

So an EPROM (the boot loader and FDC routines) must have a corresponding BIOS(containing the disk parameters and disk block map) from the first two tracks where CP/M resides.

I may not have explained it well but it's like having odd socks, not good!

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Tue Jun 13, 2017 9:06 am

Yeah, I knew that.. ;)

I believe we have a disassembly of the disk routines from the chap who did it long ago.

I haven't looked at the ACT ROM to see what it supports in terms of track counts, but I'd think it is a pretty good bet that the 3.1 ROM can do 35 track, DS. It is supposed to be from a QD. There are some others out there, but I think they are all standard Intertec ones. You know the VCF thread, but I'll post a link anyway: http://www.vcfed.org/forum/showthread.p ... -ROM-Dumps

Perhaps I should try the 3.05 ROM? But dmempis states that the 3.1 ROM works in his QD...

Update. I have looked at the ACT ROM and it appears to support 35 tracks only. Has no patch for 40 track disks. So in theory I should be able to use one of the 35 track boot disks (say, CBOOT) to format a copy of itself using the 35 track formatter (that is on the CBOOT image - F.COM, the Intertec formatter v3.1). However, it doesn't work, like before. Looks like it formats, but no changes on the target disk.

A test with PIP gives good results with these 35 track images. Doing ERA b:*.* clears down the disk as expected (consider it a "fast format") and a full copy of A:*.* to B: worked. No crashes or other oddness, which I attribute that to the ACT BIOS being 35 track. So that's one mystery solved, but what about the formatting problem, I wonder?

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Wed Jun 14, 2017 11:09 am

Ok.. have looked at the commented disassembly of the boot ROM (which also contains the formatting code that is used by CPU2). I was trying to see if there was any difference between a write sector and format command (as implemented by CPU2) and it seems that write uses the FDC1791 WRITE SECTOR command, as you'd expect, but format is using WRITE TRACK.

When it formats a track, it waits for the index pulse, lays down a gap, then the sector address block, then another gap, then the data block, then a final gap. Each block is followed by a 2 byte CRC that the FDC calculates and generates. Once it finishes all the sectors, it lays down a trailing gap.

It uses the same register to pass the bytes of the track format as the write sector routine. During the format, I can see the HxC display switching from R to W (indicates when it is reading or writing data) and of course the track / side indicators incrementing as the format progresses. And as stated previously, it actually does the verify, too...

I've read the FDC datasheet (below) but I can't see that a write sector would differ from a write track from the electronic point of view (I was hoping it would say a certain line is used that wasn't used for writing a sector, then I could check that line with a probe), but I can't see anything..

What am I missing?


Western Digital FD1791.pdf
FD1971 datasheet
(1.64 MiB) Downloaded 5 times

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Wed Jun 14, 2017 1:09 pm

Hmm, seems like I am not missing much.

  • Formatting works on a BBC Micro drive unit if I specify single sided. I think the drive must be SS.
  • The Tandon drive isn't broken. It just refuses to read the boot disk I have. I have formatted it successfully, double sided, and laid down a boot track. It's now copying all the files from my (HxC) boot disk quite happily.
  • And it boots!
  • What we seem to have here is an incompatibility with the HxC device. That, as they say, is a bit of a bummer...

StephenM
Posts: 19
Joined: Mon Jun 27, 2016 7:54 pm

Re: Fixing a Superbrain up

Postby StephenM » Wed Jun 14, 2017 1:16 pm

Well done Jon, another Superbrain up and running.
=D> =D> =D> =D>

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Wed Jun 14, 2017 1:19 pm

Not quite.

This next problem is even weirder than the previous one. The Superbrain will not start up unless the CPU is mounted in the ICE (with the ICE plugged into the CPU socket on the main board. This is jolly odd indeed, and means I can't reassemble the case!

Naturally, I checked the socket for imperfections and cleaned the legs on the CPU but it's not helped. Funny thing is, it does work with uIDE's Z80 shim plugged in, so I guess the socket on the mainboard is toast!

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Wed Jun 14, 2017 1:44 pm

Whilst we are on the subject of uIDE...

IMG_0900.JPG
uIDE-8 in situ..


IMG_0901.JPG
Identifying the IDE device


IMG_0902.JPG
Formatting a DOM..


I know it says "Amstrad PCW" but these are just strings in the executables (which are the latest versions, which is why I'm using them here). So... I have the floppy drive working, I can use kermit to transfer files from the PC, uIDE works with the standard I/O address range (C8-CF).

What do you think my next project will be?

:wink:

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Wed Jun 14, 2017 2:50 pm

With the case refitted, it starts to look impressive (even without the missing Drive B)..

IMG_0903.JPG



IMG_0904.JPG

StephenM
Posts: 19
Joined: Mon Jun 27, 2016 7:54 pm

Re: Fixing a Superbrain up

Postby StephenM » Wed Jun 14, 2017 2:53 pm

Well done(again), you have been busy. I need to make up my UIDE when I have the time.

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Wed Jun 14, 2017 2:54 pm

Not sure how long the driver will take... but I start work soon!

StephenM
Posts: 19
Joined: Mon Jun 27, 2016 7:54 pm

Re: Fixing a Superbrain up

Postby StephenM » Wed Jun 14, 2017 2:55 pm

Have you had any thoughts about putting it on the expansion bus?

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Wed Jun 14, 2017 3:32 pm

Well, yes, but the Superbrain bus is missing a signal or two (principally, /M1); plus it would require me to design an adapter similar to the PCW expansion bus adapter (which will add to the cost). There is enough space under the cover for the Z80 shim and uIDE (as you can see, it is laid on top of the motherboard for now, and I have left it like that and refitted the top case).

This does not preclude me from designing an expansion port adapter in future, because the driver code will be the same for both, in theory; or will be usable on the expansion port with very little effort.

Incidentally.. those disks you kindly sent me that won't boot with my Tandon drive.. so far I have been able to read the files. I think one of our drives may be misaligned, at least as far as the boot track is concerned. It's not a big deal, though. I successfully laid a boot track down on one of them and I can boot from it. You may have noticed that I patched the Compustar M30 boot track... ;)

And speaking of patches, I'm keen to modify my Superbrain so that it switches the drive motor off after disk access. The constantly spinning drive is a little noisy and it is driving me only slightly mad.. :shock:

User avatar
1024MAK
Posts: 6684
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...

Re: Fixing a Superbrain up

Postby 1024MAK » Thu Jun 15, 2017 12:47 pm

@Jon - proper job. Well done =D> =D> =D>

Mark
For a "Complete BBC Games Archive" visit www.bbcmicro.co.uk NOW!
BeebWiki‬ - for answers to many questions...

User avatar
jonb
Posts: 2052
Joined: Sat May 21, 2011 12:42 pm
Location: South Coast of England

Re: Fixing a Superbrain up

Postby jonb » Thu Jun 15, 2017 1:42 pm

Thank you, Mark, and thank you also for your advice - Dave too.

I think between you both, you must have helped save a shedload of old computers from the scrapheap. Bravo!

Top geezers!
=D> =D> =D>


Return to “other vintage computer hardware, software and games”

Who is online

Users browsing this forum: Bing [Bot] and 1 guest