Anyone ever seen this on an 810 Sideboard?

[joeventura]

 

It was underneath a Percom (Now destroyed) data separator board.

[Nezgar]

Interesting... it's a 2716 (2KB), so it's not an archiver/CHIP - which requires a 2732 (4KB).

 

The fact you say it had a percom data separator suggests this mod was done prior to the availability of the "grass valley" upgrades from Atari, so the owner at the time may have also opted for a ROM upgrade to get a faster sector interleave, either a copy of Atari's official "ROM C" or the user group generated patch that was documented prior to Atari's becoming available, and with an interleave that was a little faster that Atari's too. There were some commercial ads for a replacement "fast chip" ROM for the 810, which probably used the user group generated patches.

 

I'll dig up references for these later.

 

The jumper wire from the 2716 would be to provide inverted chip select compared to the original mask ROM - a common requirement when switching between mask ROM or EPROM.

 

A dump of this ROM would tell a lot to confirm my theories. If you have the means, it can be dumped as a 2732, which should repeat the same 2KB twice. If not, I'd very much like to offer to do it for you if you are willing to mail the chip, and I mail it back - possibly with another backup or slightly better one patched with the faster interleave if that one isn't.

[dzzy1982]

You should probably cover that eeprom with something to prevent it from getting damaged.

[Nezgar]

30 minutes ago, joeventura said:

Percom (Now destroyed) data separator board.

Another thought - were you able to salvage the 1771 controller chip from the damaged data separator board? If so you could plonk it into that empty socket on the sideboard for a "mostly" functional drive. Without the data separator reads may be less reliable, but maybe better than a completely dead drive.

[joeventura]

16 minutes ago, Nezgar said:

Interesting... it's a 2716 (2KB), so it's not an archiver/CHIP - which requires a 2732 (4KB).

 

The fact you say it had a percom data separator suggests this mod was done prior to the availability of the "grass valley" upgrades from Atari, so the owner at the time may have also opted for a ROM upgrade to get a faster sector interleave, either a copy of Atari's official "ROM C" or the user group generated patch that was documented prior to Atari's becoming available, and with an interleave that was a little faster that Atari's too. There were some commercial ads for a replacement "fast chip" ROM for the 810, which probably used the user group generated patches.

 

I'll dig up references for these later.

 

The jumper wire from the 2716 would be to provide inverted chip select compared to the original mask ROM - a common requirement when switching between mask ROM or EPROM.

 

A dump of this ROM would tell a lot to confirm my theories. If you have the means, it can be dumped as a 2732, which should repeat the same 2KB twice. If not, I'd very much like to offer to do it for you if you are willing to mail the chip, and I mail it back - possibly with another backup or slightly better one patched with the faster interleave if that one isn't.

2716rom here it is dumped as a 2716

 

[joeventura]

20 minutes ago, Nezgar said:

Interesting... it's a 2716 (2KB), so it's not an archiver/CHIP - which requires a 2732 (4KB).

 

The fact you say it had a percom data separator suggests this mod was done prior to the availability of the "grass valley" upgrades fro2732romm Atari, so the owner at the time may have also opted for a ROM upgrade to get a faster sector interleave, either a copy of Atari's official "ROM C" or the user group generated patch that was documented prior to Atari's becoming available, and with an interleave that was a little faster that Atari's too. There were some commercial ads for a replacement "fast chip" ROM for the 810, which probably used the user group generated patches.

 

I'll dig up references for these later.

 

The jumper wire from the 2716 would be to provide inverted chip select compared to the original mask ROM - a common requirement when switching between mask ROM or EPROM.

 

A dump of this ROM would tell a lot to confirm my theories. If you have the means, it can be dumped as a 2732, which should repeat the same 2KB twice. If not, I'd very much like to offer to do it for you if you are willing to mail the chip, and I mail it back - possibly with another backup or slightly better one patched with the faster interleave if that one isn't.

and as a 2732

Edited March 15, 2022 by joeventura

[Nezgar]

1 hour ago, joeventura said:

2716rom here it is dumped as a 2716

 

Nailed it.... This dump has a CRC32 checksum of 80724CDE, which matches the early user-group documented "Chicago layout" patch that I re-created!

 

References:

• https://atariage.com/forums/topic/279363-the-atari-810-revision-b-rom-has-finally-been-found-dumped/?do=findComment&comment=4113840 - My previous post about references to the "Chicago Format" in old ACE newsletters from 1980-81 that describe how to patch Atari's 810 Rev B ROM with the faster sector interleave.
• https://atariage.com/forums/topic/279363-the-atari-810-revision-b-rom-has-finally-been-found-dumped/?do=findComment&comment=4114492 - This post contains my hand patched Rev B ROM code according to the above ACE newsletter articles, which also resulted in the same CRC32 checksum of 80724CDE as joeventura's.
• https://www.atariarchives.org/creativeatari/Atari_DOS.php This page is a transcription of an article from "Creative Atari" that describes the origin of 810 "Fast Format Chips" - "A group of users in Chicago modified the B layout to what is called the Chicago layout. This layout is 30% quicker than the B format and indeed is 10% quicker than Atari's own C layout."
• https://archive.org/details/AtariComputerEnthusiastsNewslettersVolume1/page/n111/mode/2up?q=810 - Page 112 of 216 in this archive.org PDF contains the original article from October 1981 by Larry Hitz of the Chicago Atari User's Group "CONVERT YOUR 810 DISK DRIVE FOR FAST FORMATTING" describing the specific patches to the Rev B ROM. Since this was a from a Eugene Oregon ACE newsletter, the original article would have been published in Chicago even earlier. Installation instructions I'll expand on below.
• The above article also describes the small wiring modification needed in order to use a 2716 EPROM, as seen in @joeventura's drive.
• Atari didn't start offering ROM revision C or data separator boards in new drives or as upgrades until October 1981: https://www.atarimagazines.com/v1n4/upgradesavailable.html

As per the articles installation instructions:

• On the foil side of the socket of A102, cut the foil trace connected to pins 18, 20, 21 as close to the pin as possible, then make the following connections:
• Connect pin 18 to ground
• Connect pin 21 to +5 VDC
• Connect pin 20 to pin 2 of Z103 (the D4069 inverter chip)

@joeventura - can you take a picture of the back of your side board to confirm the traces are cut as per above?
The last step is different than what we see pictured in your drive, I'm guessing because that particular pin on the MC14011B is connected via a trace to pin 2 of Z103, which would be functionally the same.

 

Here is another article by the same author Larry Hits from Oct 81 (Page 13) that recommends contacting PERCOM for a data separator board for $29.95. It also discusses the super sensitive RPM control pot on the original 810 power/analog board. (This was improved in the later replacement separate power+analog boards). On page 14, describes a 1-byte correction to the original ROM patch article, which I also incorporated into my resulting binary that matches joeventura's dump:

https://archive.org/details/AtariComputerEnthusiastsNewslettersVolume1/page/n127/mode/2up?q=810

[Nezgar]

  Heh... another gem in the September 1981 Eugene Oregon ACE newsletter on Page 11: https://archive.org/details/AtariComputerEnthusiastsNewslettersVolume1/page/n97/mode/2up?q=format&view=theater

 

" The new disk drive is now available. This drive differs, because it contains a new data separator board and comes with the Master Diskette II Package. The Master Dikette II package contains a new Dos 2.0S manual that is very good. Contrary to popular belief, the new drive does not contain a Fast Format ROM (C Chip). These chips may eventually be obtained from ATARI at no charge; according to sources.

...

Sources at ATARI are quoted as saying, "Anyone who wishes to duplicate the C ROM will not be harassed by ATARI." Up to two weeks ago this was not the case, and violators were threatened by ATARI!"

 

? Some interesting times there.... "Don't you dare update the crappy firmware in our drives without paying us..."

[cwilbar]

@joeventura Does this have the old power board, or the updated one ? 

What happened to the percom data separator ?

Is there an analog board on the top of the drive mechanism ?

 

Sounds like you have an early drive there.  I have one as well (it doesn't work ? ). 

 

 

[joeventura]

4 hours ago, Nezgar said:

Nailed it.... This dump has a CRC32 checksum of 80724CDE, which matches the early user-group documented "Chicago layout" patch that I re-created!

 

References:

• https://atariage.com/forums/topic/279363-the-atari-810-revision-b-rom-has-finally-been-found-dumped/?do=findComment&comment=4113840 - My previous post about references to the "Chicago Format" in old ACE newsletters from 1980-81 that describe how to patch Atari's 810 Rev B ROM with the faster sector interleave.
• https://atariage.com/forums/topic/279363-the-atari-810-revision-b-rom-has-finally-been-found-dumped/?do=findComment&comment=4114492 - This post contains my hand patched Rev B ROM code according to the above ACE newsletter articles, which also resulted in the same CRC32 checksum of 80724CDE as joeventura's.
• https://www.atariarchives.org/creativeatari/Atari_DOS.php This page is a transcription of an article from "Creative Atari" that describes the origin of 810 "Fast Format Chips" - "A group of users in Chicago modified the B layout to what is called the Chicago layout. This layout is 30% quicker than the B format and indeed is 10% quicker than Atari's own C layout."
• https://archive.org/details/AtariComputerEnthusiastsNewslettersVolume1/page/n111/mode/2up?q=810 - Page 112 of 216 in this archive.org PDF contains the original article from October 1981 by Larry Hitz of the Chicago Atari User's Group "CONVERT YOUR 810 DISK DRIVE FOR FAST FORMATTING" describing the specific patches to the Rev B ROM. Since this was a from a Eugene Oregon ACE newsletter, the original article would have been published in Chicago even earlier. Installation instructions I'll expand on below.
• The above article also describes the small wiring modification needed in order to use a 2716 EPROM, as seen in @joeventura's drive.
• Atari didn't start offering ROM revision C or data separator boards in new drives or as upgrades until October 1981: https://www.atarimagazines.com/v1n4/upgradesavailable.html

As per the articles installation instructions:

• On the foil side of the socket of A102, cut the foil trace connected to pins 18, 20, 21 as close to the pin as possible, then make the following connections:
• Connect pin 18 to ground
• Connect pin 21 to +5 VDC
• Connect pin 20 to pin 2 of Z103 (the D4069 inverter chip)

@joeventura - can you take a picture of the back of your side board to confirm the traces are cut as per above?
The last step is different than what we see pictured in your drive, I'm guessing because that particular pin on the MC14011B is connected via a trace to pin 2 of Z103, which would be functionally the same.

 

Here is another article by the same author Larry Hits from Oct 81 (Page 13) that recommends contacting PERCOM for a data separator board for $29.95. It also discusses the super sensitive RPM control pot on the original 810 power/analog board. (This was improved in the later replacement separate power+analog boards). On page 14, describes a 1-byte correction to the original ROM patch article, which I also incorporated into my resulting binary that matches joeventura's dump:

https://archive.org/details/AtariComputerEnthusiastsNewslettersVolume1/page/n127/mode/2up?q=810

There were no traces cut but the pins you mentioned were all pulled out of the socket there may have been a jumper on one of them that isnt there anymore.

The jumper from pin 20 to pin 2 of Z103 is still there

[joeventura]

3 hours ago, cwilbar said:

@joeventura Does this have the old power board, or the updated one ? 

What happened to the percom data separator ?

Is there an analog board on the top of the drive mechanism ?

 

Sounds like you have an early drive there.  I have one as well (it doesn't work ? ). 

 

 

Old power board,

Percom card was destroyed (all pins broken off when it came lose and bounced around for 4 or 5 house moves) then tried to repair it and trashed it

[joeventura]

4 hours ago, Nezgar said:

Nailed it.... This dump has a CRC32 checksum of 80724CDE, which matches the early user-group documented "Chicago layout" patch that I re-created!

 

 

Glad I could get you an original dump

[Nezgar]

5 minutes ago, joeventura said:

Percom card was destroyed (all pins broken off when it came lose and bounced around for 4 or 5 house moves) then tried to repair it and trashed it

Could you post a picture of it? I wonder "how bad" ... I wonder if there's any remote chance of repair ie with replacement header pins / jumper wires etc... also as asked earlier was the 1771 controller chip on it salvageable?

[_The Doctor__]

removing chip swipes was a nice way to make mods reversible without damaging pcb's looks nice too, what's not to love?

[joeventura]

18 hours ago, Nezgar said:

Could you post a picture of it? I wonder "how bad" ... I wonder if there's any remote chance of repair ie with replacement header pins / jumper wires etc... also as asked earlier was the 1771 controller chip on it salvageable?

I dont even know if I still have it