Anyone have Percom firmware dumps ?

[cwilbar]

I have an AT88-SP1 (actually 3 of them).  I'd like to try an experiment.  The percom doubler hardware is reproduced....

I'd like to try that reproduction in an AT88-SP1, but I need a dump of the percom doubler firmware.

 

Anyone happen to have it, or have a pointer to somewhere it might be ?

 

I haven't had any luck finding it.

 

I don't know if firmware from RFD-4x will work.  If Percom used the same chips, possibly.

 

I read that serial may use a different chip on the drives with parallel ports ? 

 

Its a long shot, but I thought I'd check.

 

-- Curt

 

[+Nezgar]

Percom Doubler hardware reproduced? Most excellent news! I'm sure there are folks on these forums with single density drives that would love to get them DD upgraded.

 

I believe the firmware from the original RFD40-S2, AT-88, and the Astra "The One", "Big D" and 1001, 2001, 1620, etc all use "compatible" formware. The percom drives had revised versions throughout their life, which I'm trying to archive, and the Astra drives used firmware that has evidence of being derived from the original code as well.

 

I do not have a ROM dump from a single density percom drive yet, so i'd be interested if you are willing to do that.

 

I have a few firmwares here, and I also have an AT-88 myself which is DD capable, but I haven't dumped my own ROM yet!

 

I've been meaning to post more detail on my findings so far, but I'll attach what I have here with brief notes to get you started:

• PERCOM-2K 5A396459.EPR - this PERCOM ROM from an unknown source online, not known what type of drive it actually came from. I suspect from an RFD-40S1, since that is what Alitrra bases it's PERCOM support from.
• Percom RFD40-S2 Disk Drive EPROM (Ver 1.10)(11-8-82) from ballyalley E2D4A05C.Bin - Long descriptive filename - Thanks to @ballyalley for working with me to get this verified dump from his RFD40-S2. Not sure if it's older or newer than PERCOM-2K 5A396459.EPR
  
• Astra1001 2AB65122.rom - a verified dump from an Astra 1001, with an odd label also indicating "The One". @phaeron disassembled this ROM in a prior thread and found it had Percom RFD heritage, and it even runs with Altirra's PERCOM RFD-40S1 full drive emulation. Read more about it here: https://atariage.com/forums/topic/182786-astra-the-one-drive/?do=findComment&comment=4273812
• astra1620 2AB65122.bin - Potentially same firmware as the Astra 2001. Found from this page: http://blog.3b2.sk/igi/post/ATARI-floppy-disk-drive-ASTRA-1620.aspx

Look forward to hearing more about your project! And hopefully 2020 will be the year I tear apart my Percom AT-88S1 to dump it's ROM too.

 

EDIT: well look at that - I didn't notice this until now - the verified Astra 1001 dump, and the 1620 dump from blog.3b2.sk have the same CRC32!

 

PERCOM-2K 5A396459.EPR Percom RFD40-S2 Disk Drive EPROM (Ver 1.10)(11-8-82) from ballyalley E2D4A05C.Bin Astra1001 2AB65122.rom

astra1620 2AB65122.bin

[phaeron]

The Astra 1620 and Astra 1001 firmwares are exactly the same, as implied by their CRC32s.

 

The V1.10 ROM differs from the Astra 1001 ROM by exactly one byte -- the initial step rate. It is set to 1 in the Percom ROM and 2 in the Astra ROM. This basically confirms that the Astra ROM is based on the updated Percom firmware.

 

[+Nezgar]

25 minutes ago, phaeron said:

The Astra 1620 and Astra 1001 firmwares are exactly the same, as implied by their CRC32s.

Yeah, I edited my post after to indicate that I noticed the CRC32's were the same those two.

 

25 minutes ago, phaeron said:

The V1.10 ROM differs from the Astra 1001 ROM by exactly one byte -- the initial step rate. It is set to 1 in the Percom ROM and 2 in the Astra ROM. This basically confirms that the Astra ROM is based on the updated Percom firmware.

Wow kewl, talk about finding a missing link! So for the benefit of others reading, I Phaeron's previous findings from the Astra 1001/The One ROM apply to the Ver 1.10 PERCOM ROM: https://atariage.com/forums/topic/182786-astra-the-one-drive/?do=findComment&comment=4280417

 

Cheers, N.

[+Nezgar]

Here's another one to check out @phaeron - @moonlight_mile has posted a ROM dump labelled "RFD 1.20" from an RFD-44 with CRC32 C6C73D23, which doesn't match any others I have posted so far in this thread.

 

Binary here:

https://atariage.com/forums/topic/297197-need-help-percom-rfd-board/?do=findComment&comment=4432520

 

Lots of bytes shifted comparing this to the other dumps, so potentially a later version as the label suggests. It works in Altirra full drive emulation, can read SD and DD disks.

[phaeron]

RFD V1.20 disassembly attached. There are two sets of changes relative to RFD V1.10:

• The format routine reverses the sector interleave for side 2 to take into account the reversed logical sector mapping. This is done by reversing the sector numbers (sector = 19-sector).
• Fix in the track/sector division routine for a funny bug where the last few sectors on side 2 were forced to 128 bytes in double density. This was fixed by moving the boot sector check before the side 2 flip code.

It looks like they still hadn't fixed the off-by-one bug that makes the last logical sector on double sided disks map to track 0, sector 0. Am I missing something or did they really ship three revisions of firmware that couldn't write all of the sectors on the disk?

 

percom-1.20.s

[cwilbar]

Is the binary for 1.20 available ?

 

IIRC my AT88-S1 have 1.10 firmware.... but I don't know if that is different from RFD40-S1 (as an example)'s 1.10.

 

I will pull one of my ROMs and read it with my programmer.  We'll see if it matches up to any of the ones posted here, or adds to the collection.

 

-- Curt

 

[+Nezgar]

45 minutes ago, cwilbar said:

Is the binary for 1.20 available ?

Yep, moonlight_mile's "RFD 1.20" binary is here: https://atariage.com/forums/topic/297197-need-help-percom-rfd-board/?do=findComment&comment=4432520

 

So cool to find and unravel a of all these.

 

I think I misread your first post, and you meant you are looking for a percom doubler board, because all 3 of your existing Percoms are single density only?

[cwilbar]

The Percom doubler board (intended use in TRS-80) is reproduced.  My intent is to try one in a Percom with the appropriate firmware.

 

It opens up other thoughts, like if the 810 uses the same single density controller, might be able to put one of these in place of the Atari data separator.... but then custom firmware will be required.

 

-- Curt

 

[+Nezgar]

1 hour ago, cwilbar said:

The Percom doubler board (intended use in TRS-80) is reproduced.

Where can I find this?

 

1 hour ago, cwilbar said:

It opens up other thoughts, like if the 810 uses the same single density controller, might be able to put one of these in place of the Atari data separator.... but then custom firmware will be required.

Yes this has been done for the 810, check out this topic: https://atariage.com/forums/topic/291937-neanderthal-computer-things-810-turbo/

 

And this podcast interview with its creator: https://ataripodcast.libsyn.com/antic-interview-377-james-hugard-neanderthal-computer-things

[cwilbar]

3 hours ago, Nezgar said:

Where can I find this?

 

Yes this has been done for the 810, check out this topic: https://atariage.com/forums/topic/291937-neanderthal-computer-things-810-turbo/

 

And this podcast interview with its creator: https://ataripodcast.libsyn.com/antic-interview-377-james-hugard-neanderthal-computer-things

I have previously read that.  While cool, I think the simplest approach would be to use one of the Percom doubler recreations and ?adapt? 1050 USD firmware.  This would keep the requirements down to a Percom Doubler recreation and additional RAM to handle 256 byte sectors.  A interleave scheme would be required since it would not incorporate an entire track buffer.  At least that would be an interesting starting point.

 

[cwilbar]

3 hours ago, Nezgar said:

Where can I find this?

 

Yes this has been done for the 810, check out this topic: https://atariage.com/forums/topic/291937-neanderthal-computer-things-810-turbo/

 

And this podcast interview with its creator: https://ataripodcast.libsyn.com/antic-interview-377-james-hugard-neanderthal-computer-things

 

You can find the Percom doubler recreation at:

 

https://www.ebay.com/itm/153764663296

and

http://members.iinet.net.au/~ianmav/trs80/newstuff.htm

 

-- Curt

 

 

[cwilbar]

I have an AT88-S1PD that I am fixing.  This one has a ROM with hand written text that says:

 

SPD

1.21

 

Percom data on the net seems to indicate all AT88 drives are single density, but I'm pretty sure this one is DD.  The FDC is right for DD, and the PCB in this is on the opposite side of the chassis like the RFD drives.

 

I will be dumping the ROM.

 

Right now voltages are a bit low (~4.X on the 5V rail, ~11.X on the 12V rail).  I think the axial capacitors on the board (3x33uF, 1x3300uF, and 1x4700uF) may be the cause of the power issues.  I don't have suitable caps around, so I may solder wires to the PCB and run this with a DC supply for testing purposes.

 

I'll post the ROM once it is dumped.  I do not believe I have seen this ROM ver mentioned here.

 

 

[+Nezgar]

We can verify your AT88-SPD v1.21 against the one posted here: https://atariage.com/forums/topic/304384-percom-at-88-doubler-rom/?do=findComment&comment=4503618

 

 

The "SPD" variants were later models compared to the original AT88's, and have very different controllers and ROM as well. The original AT88's were released with only single density support, but could be later upgraded by swapping out a PCB and adding a DD controller in parallel with the SD controller. Later drives came with this preinstalled.

 

We have 1 ROM dump from a single-density AT-88 and found the ROM is actually DD capable if the DD controller present, so I'm still not sure if there was an earlier ROM that did not include DD support code. This was originally assumed since the upgrade instructions included steps for swapping out the ROM. Maybe it was primarily for bugfixes.

[cwilbar]

I plan to dump the ROM sometime this weekend.  I'll post it here once dumped.

 

This SPD is much different from my other AT88s.

 

One think that I also see different with this, as it was a one drive model, to upgrade to 2 drives, you had to change (or cut) the front metal.  The screw spots are there in the bottom for a second drive, and I presume they at an AT88S2PD that would have 2 drives.  They must have used different metal for the front based on the configuration.

 

Putting in a combo 3.5/5.25 drives making a dual 720K setup would be interesting (720K 5.25 and 720K 3.5 ? ).

 

[cwilbar]

Here is a read of the rom, it is a 2732.

 

I have soldered a 4 pin drive power connector to the appropriate points on the backside of the AT88S1PD's board, and I'll be putting it back toogether and powering it via a external supply that I use with my USB to IDE/SATA adapter.  If it works that way I know the problem is all power.  If not, then I'll have to begin troubleshooting, checking the clocks, reset pin on the cpu, address and data lines for activity, etc. 

 

 

spd121.bin

[+Nezgar]

Thanks @cwilbar - your spd121.bin has CRC32 of FC6675F2, which matches JR>'s v1.21 dump.

 

Disassembly by Phaeron here:
https://atariage.com/forums/topic/304384-percom-at-88-doubler-rom/?do=findComment&comment=4502829

 

[_The Doctor__]

so long as nothing is dead on the board that should work, if something is dead on the board you could cook some stuff.

Edited June 29, 2020 by _The Doctor__

[cwilbar]

I ran it with the existing supply.... it was about a half a volt low on both rails.  I think if chips were going to pop, would have dragged the voltage lower and probably taken out the regulator(s).

 

What a miserable method of attaching those for heat sinking.... don't know what Percom was thinking.  One I worked on had two regulators with broken leads as the regulators take a lot of stress on SIO insertion/removal.  Probably the dumbest part of the design :-).

 

Wonder if anyone has ever figure out how to hard code the divider for faster SIO speed.

 

[cwilbar]

3 hours ago, Nezgar said:

Thanks @cwilbar - your spd121.bin has CRC32 of FC6675F2, which matches JR>'s v1.21 dump.

 

Disassembly by Phaeron here:
https://atariage.com/forums/topic/304384-percom-at-88-doubler-rom/?do=findComment&comment=4502829

 

Cool, well at least I know the EPROM should be good then ?

 

Wonder if that was the last version ?

 

[cwilbar]

On 6/28/2020 at 7:04 PM, cwilbar said:

Here is a read of the rom, it is a 2732.

 

I have soldered a 4 pin drive power connector to the appropriate points on the backside of the AT88S1PD's board, and I'll be putting it back toogether and powering it via a external supply that I use with my USB to IDE/SATA adapter.  If it works that way I know the problem is all power.  If not, then I'll have to begin troubleshooting, checking the clocks, reset pin on the cpu, address and data lines for activity, etc. 

 

 

spd121.bin 4 kB · 4 downloads

FYI, the Percom worked !  So issues with that unit are in the power supply section.  I've hacked in some incorrect capacitors (radials, one under voltage) to do testing/troubleshooting before ordering replacements (want to put together an order bigger than 6 caps ? ).  The 33uF caps seem to measure around 29ish uF, and all have consistent and seemingly OK ESR.  The ESR looks good on the two big caps too, but my old LCR meter can't measure caps that high in value, so don't know if they are in spec or not.  Problem might not be the caps.  Have to check the diodes, and then if they are OK, start looking at the regulators (which were all putting out about .5 volt too low, like something (I presumed leaking caps ?) was dragging down the unregulated side.

 

The drive mech (MPI half height) couldn't seem to get enough speed up, so I used an Epson mech and it booted up MyDOS perfectly.

 

I've since lubricated the bearings in the MPI mech and need to retest it.  Though I'll probably go with a 360K mech in the end with this drive since it does double density.

 

 

[cwilbar]

following up on my old thread.  I still have to fix that drive (other things got bumped to the front of the line). 

 

However, I have won an eBay listing for a Percom double density upgrade (original) for Atari.  The PCB has P/N 510-2030-001.  It has a ROM with it

with handwritten label that says:

 

0066

V1.12

 

This is still sealed, so I have not done any testing.  But I think I'm going to get all my Percom units out, and inventory them all... firmware, model info, etc, etc.

 

I do not know if the Percom Atari doubler is different from the Percom Tandy doubler which is what the eBay based Percom doubler is a recreation of.  I have not tried to compare them.

 

Just posting about this, and hopefully in the coming weeks, I'll figure out some more around this.

 

[moonlight_mile]

Ahh so you are the one that outbid me. Lol. Glad someone got it that can do something with it.  If ever it gets recreated I’d love one for my sole at-88. 

[cwilbar]

15 hours ago, moonlight_mile said:

Ahh so you are the one that outbid me. Lol. Glad someone got it that can do something with it.  If ever it gets recreated I’d love one for my sole at-88. 

While I haven't verified it yet, I believe the new recreation for the TRS-80 that is available on eBay (and probably online somewhere as well) is likely the same thing.

At a point, I hope to test that theory since I have an original (for Atari) one, and one of the recreations. 

 

I've never seen one of these available before.  So I had to scoop it up.  After I go through my Percom drives, if it becomes available, I'll let you know.

 

And if I determine the modern re-makes are identical or work in the Percoms I'll update this thread.

 

[Sikor]

Anyway: https://www.jsobola.atari8.info/dereatari/schematy.htm

Jerzy Sobola made many reverse-enginering schema and many ROM's dump.