1050 ROMs

[+kheller2]

Tandon stepper: Novacom 1005-101 (US made, surprising.)

WST stepper: Sankyo Seiki MSHF200B71 (Japan)

Not much out there on the web, I doubt I have some old parts catalogs from back then.

 

They both use the same drive motor, of course Tandon has a Tandon part number on it

 

In case anyone wonders, yes the plastics are different. Pictures of steppers and mechanism follow:

Edited August 29, 2016 by kheller2

[ijor]

Of course then, wouldn't shorter seek times break some copy protection? Hmm maybe not.

 

I just checked one game with a protection that does depend on the seek and settle time. DRELBS by Synapse. But, at least in this case, the faster time might probably help.

 

The game doesn't seem to work on the (TANDON) 1050, only on the 810, precisely because the of the different seek times. I didn't check, but it is possible that it will work on the WST mech because its seek and settle times are closer to the 810 ones.

[+kheller2]

DRELBS is cool.

 

I guess I should try swapping mechs around just to confirm that they are or are not compatible.

[ebiguy]

 

hkeller2 said : Any thoughts as to why they didn't use the FDC for stepping?

 

The FDC can't control directly a 4 four phases stepper. It has a simplified DIR/STEP interface that requires a small logic at the mechanism, as PC drives do.

 

 

Hi Ijor, I think hkeller2's question is stil valid. I mean they could use a drive mech and a stepprt motor compatible with the FDC of that time.

I hardly imagine that Western Digital included a way to control stepper motor and, at the same time, mech manufacturers used some stepper motors that could not work with the controllers on the market.

Or did I miss something ?

[ijor]

 

Hi Ijor, I think hkeller2's question is stil valid. I mean they could use a drive mech and a stepprt motor compatible with the FDC of that time.

I hardly imagine that Western Digital included a way to control stepper motor and, at the same time, mech manufacturers used some stepper motors that could not work with the controllers on the market.

 

I don't know the historical or the actual market reasons. But I could imagine that WD was following market trends, and Atari implemented whatever was cheaper.

 

The very first WD's FDC, the FD1771 that is found on the 810, does have some kind of support for driving directly a multi phase stepper motor. It is 3-phases, not 4, as in the Atari. No idea how reliably, or how common it was. But that disappeared from later WD FDCs, obviously because the Shugart interface became too popular and it didn't make any sense to keep supporting something that was rarely used.

 

Atari really didn't need it. The drive already has its own dedicated processor, firmware, ports, etc (contrast this with other platform that control the drive directly). There is no much gain, if at all, in using the FDC built-in stepper interface. On the contrary, it requires some additional logic on the mechanism that would make it a little bit more expensive.

 

Of course, by the time of the XF-551, PC drives became so popular and so cheap, that building a custom mechanism would have been crazy.

[+Nezgar]

 

ROMS...

The World Storage Technology (Copal?) ROM identifies itself as K but it is very different binary wise.

 

I'm not sure if anyone has confirmed that only L supports the 2797. And I'm curious about the masked Tandon ROMs and what version of code they have. I'm too lazy to make a 2532 to 2732 adapter for my TL866. Actually, a universal ROM reader adapter for the TL866 would be sweet.

 

 

MD5 (1050-revJ.rom) = dd296b6e2aacfd89cc94cc7e6dcebfd3

MD5 (1050-revK.rom) = 5acf59fff75d36a079771b34d7c7d349

MD5 (1050-revL.rom) = 2575d0514c5fe2dd6f5bd3f0ec434eb7

MD5 (WSTR5.BIN) = 5754ab8bef5c93e0caf17425d96bd80e

 

 

It seems there's one more 1050 ROM I found on pigwa that's not in the list above:

"FLOPOS.ROM" MD5 = ad4b6ec7de5f3fe165df02d832e31be4

http://ftp.pigwa.net/stuff/collections/nir_dary_cds/ROMS/ROMS_HW/1050OS/FLOPOS.ROM

 

Curious if anyone can identify this? Maybe it's a Rev "E" or "F" ROM mentioned in the tech tip on page 40 of this 1050 service manual? They were to be replaced with Rev "J".

http://www.atarimania.com/documents/atari-1050-field-service-manual.pdf#page=40

[+kheller2]

 

It seems there's one more 1050 ROM I found on pigwa that's not in the list above:

"FLOPOS.ROM" MD5 = ad4b6ec7de5f3fe165df02d832e31be4

http://ftp.pigwa.net/stuff/collections/nir_dary_cds/ROMS/ROMS_HW/1050OS/FLOPOS.ROM

 

Curious if anyone can identify this? Maybe it's a Rev "E" or "F" ROM mentioned in the tech tip on page 40 of this 1050 service manual? They were to be replaced with Rev "J".

http://www.atarimania.com/documents/atari-1050-field-service-manual.pdf#page=40

 

 

I think that is a compiled version of the 1050 ROM based on the source in the DISASM.ZIP file in the same directory.

 

Now the 65XEM ROMs... they interest me...

[1050]

Confirmed MD5 identical to assembled FLOPOS source

code project found in that same folder at pigwa.

[+Nezgar]

Thanks for confirming. I also didn't notice before there's a description in the "FLOPOS.TXT"

"flopos.rom Floppy 1050 OS, assembled from mac/65 source"

 

I did find the mention of the pre rev "J" ROMs interesting though.

[+Nezgar]

But the stepping rate is much faster! If I'm correct, it's about 50% faster in the WST rom. Never tried if you can't increase the step rate on TANDON mechanisms, assumed always you can't too much. That would mean that WST mechanisms are better, at least in that regard! ... but you are missing the faster step rate currently when using the USD rom. You might want to try patching the USD with the faster rate

If the skill is out there, I would love to see the faster stepping used in the WST ROM patched into a USD ROM! But that would only work with WST mechs.

 

The Speedy stepping speed is similar, and works with with both types of mechs.

Edited March 22, 2018 by Nezgar

[+kheller2]

Confirmed MD5 identical to assembled FLOPOS source

code project found in that same folder at pigwa.

 

 

What version of code does this compile close to? J, K, L?

[1050]

K closest with 9 differences

Comparing files FLOPOS.BIN and 1050-REVK.ROM
00000253: 90 91
0000029A: 20 4C
000002AB: 00 01
000006BF: A0 9D
00000BD8: B2 B1
00000BDA: B1 B2
00000C08: B2 B1
00000C0A: B1 B2
00000F1C: 7F F7

L has 16 so next in line, but the rest are thousands

of differences. But that could be unfair to go on

because file compare isn't looking for the best match

between them so a one byte offset counts as the entire

file being different or perhaps 'bad' if one's head isn't

in the right place when using such a crude tool as MS-DOS

file compare.

 

Double checking the differences and I see a good number

of corrections that might be employed in everyday tasks

and not just the function 23h exhaustive self test as I

once heard this code fixed mainly. Thus the readme inside

this collection of source code done up just last December

is possibly spouting a good deal of nonsense.

 

http://atariage.com/forums/topic/273221-dd-disc-track-structure-kryoflux-image/?do=findComment&comment=3917681

[+MrFish]

File compare programs are generally better for finding out if files are different,

more than for finding out how different they are, unless you're willing to insert

your own intelligence (read: time) into the equation; in which case you become

the file compare program.

Edited March 23, 2018 by MrFish

[+Nezgar]

If the skill is out there, I would love to see the faster stepping used in the WST ROM patched into a USD ROM! But that would only work with WST mechs.

 

The Speedy stepping speed is similar, and works with with both types of mechs.

 

It appears I've answered my own request! There already exists a US Doubler ROM that has enhanced seeking speed mechanisms:

http://atariage.com/forums/topic/275166-one-last-eprom/?p=3991800

[slaker]

 

It appears I've answered my own request! There already exists a US Doubler ROM that has enhanced seeking speed mechanisms:

http://atariage.com/forums/topic/275166-one-last-eprom/?p=3991800

 

Ugh, I wouldn't recommend that ROM (see that thread for the gruesome detail).

 

I disassembled and compared the stepping code in the Tandon Rev. K and WSTR5 ROMs. Attached is a reassembled Usdblr_err.rom with the correct stepper motor phase encodings and shorter seek delays for the WST mech. The hokey-pokey "half-step-in, half-step-out" code on inward seeks is also skipped as per WSTR5. The changes amount to six bytes.

 

The usual disclaimers about trashing your data, incinerating your drive, etc. apply, but the ROM seems to work fine in my WST 1050, though I didn't do much beyond boot DOS, format a floppy, and start a few applications. Perhaps someone more ambitious than I will benchmark it.

Usdblr_err.wts.rom

[+Nezgar]

Very nice to have a WST optimized US Doubler image, thanks! I'll have to try this on a WST mech.

 

I've tested the WSTR5 code before with a Tandon mech and it's definitely incompatible... stepper does not move properly. It would be interesting to see the timings used on Speedy, turbo 1050, and the archiver/chip emulator as they all do a faster stepping rate that appears to work well with Tandon mechs.. (Although all most likely out of published specs) and use this or an average for a 'tandon optimized' us doubler.

[_The Doctor__]

err wts errrrr ummmmm err wst?

[+Nezgar]

err wts errrrr ummmmm err wst?

I think wts = wst = World Storage Technologies

Must be a typo in the filename

[slaker]

I think wts = wst = World Storage Technologies

Must be a typo in the filename

 

That's what I get for posting at 1AM. Here's the same file with the correct name for the sticklers out there.

 

Usdblr_err.wst.rom

[_The Doctor__]

so long as the re assembly code changes didn't happen at 1 am ... we're all good

[+kheller2]

This is probably the same code floating around the Internet, but I wanted to upload a pure copy just dumped today. This is the official EPROM I got from ICD back in the 80's.

 

ICD USD REV L EPROM:

MD5 (ICDUSDREVL.BIN) = d182078af7dc997ebcd9f30029d58b5b

ICDUSDREVL.BIN

[+Nezgar]

My US Doubler EPROM from B&C matches your MD5 checksum:

crc32  18690ca3
md5    D182078AF7DC997EBCD9F30029D58B5B
sha1   8a5a6082596b61d276a59207ad9f8507af328db5

Thanks for verifying.

Now to verify Geister's US Doubler mask ROM dated Nov '86. This would logically make sense to be the same 'last' version. With that "Rev L" hint, it implies there was a maybe a previous version of USD ROM that didn't support the 2797 WDC controller. Support was added for it in Atari's stock 1050 "Rev L" ROM.

[dhinson919]

File compare programs are generally better for finding out if files are different,

more than for finding out how different they are, unless you're willing to insert

your own intelligence (read: time) into the equation; in which case you become

the file compare program.

 

 

Here are some additional details regarding the differences between revK and revL:

1,2c1,2
< HAC/65 v0.3 6502 Inferencing Disassembler [run:Mon Oct  1 03:49:58 2018]
< hac65 -a1050 -od 1050-revK.rom[md5:5acf59fff75d36a079771b34d7c7d349]
---
> HAC/65 v0.3 6502 Inferencing Disassembler [run:Mon Oct  1 03:49:59 2018]
> hac65 -a1050 -od 1050-revL.rom[md5:2575d0514c5fe2dd6f5bd3f0ec434eb7]
10c10
< F004  57                  .BYTE $57
---
> F004  38                  .BYTE $38
669c669
< F552  A9 82               LDA #$82
---
> F552  A9 88               LDA #$88
748c748
< F602  A9 A2               LDA #$A2
---
> F602  A9 A8               LDA #$A8
798c798
< F670  A9 82               LDA #$82
---
> F670  A9 88               LDA #$88
1876c1876
< FF7F  A9 82               LDA #$82
---
> FF7F  A9 88               LDA #$88
1893c1893
< FFA5  A9 82               LDA #$82
---
> FFA5  A9 88               LDA #$88
1952c1952
< FFF9  4B                  .BYTE $4B
---
> FFF9  4C                  .BYTE $4C

And revK and FLOPOS:

1,2c1,2
< HAC/65 v0.3 6502 Inferencing Disassembler [run:Mon Oct  1 03:49:58 2018]
< hac65 -a1050 -od 1050-revK.rom[md5:5acf59fff75d36a079771b34d7c7d349]
---
> HAC/65 v0.3 6502 Inferencing Disassembler [run:Mon Oct  1 03:49:59 2018]
> hac65 -a1050 -od FLOPOS.ROM[md5:ad4b6ec7de5f3fe165df02d832e31be4]
296c296
< F252  85 91               STA ERROR
---
> F252  85 90               STA STATUS
329c329
< F29A  4C 91 F1            JMP $F191
---
> F29A  20 91 F1            JSR $F191
336c336
< F2AA  A9 01               LDA #1
---
> F2AA  A9 00               LDA #0
832c832
< F6BE  20 9D F2            JSR $F29D
---
> F6BE  20 A0 F2            JSR $F2A0
1433,1434c1433,1434
< FBD7  E6 B1               INC SBUF
< FBD9  C6 B2               DEC $B2
---
> FBD7  E6 B2               INC $B2
> FBD9  C6 B1               DEC SBUF
1457,1458c1457,1458
< FC07  E6 B1               INC SBUF
< FC09  C6 B2               DEC $B2
---
> FC07  E6 B2               INC $B2
> FC09  C6 B1               DEC SBUF
1834c1834
< FF1B  29 F7               AND #$F7
---
> FF1B  29 7F               AND #$7F

The latter one uses some EQUs extracted from the FLOPOS source files. FLOPOS clearly corrected a number of silly bugs in revK that are amusing and simply demonstrating that Atari engineers were human too!

[+kheller2]

It should be simple then to update flopos to support the 2797 as that was the major difference between k and L. And it will be nice to have an up to date and fixed semi documented version.

[dhinson919]

It should be simple then to update flopos to support the 2797 as that was the major difference between k and L. And it will be nice to have an up to date and fixed semi documented version.

 

Yes, the diffs between revL and FLOPOS are simple:

1,2c1,2
< HAC/65 v0.3 6502 Inferencing Disassembler [run:Mon Oct  1 13:07:10 2018]
< hac65 -a1050 -od 1050-revL.rom[md5:2575d0514c5fe2dd6f5bd3f0ec434eb7]
---
> HAC/65 v0.3 6502 Inferencing Disassembler [run:Mon Oct  1 03:49:59 2018]
> hac65 -a1050 -od FLOPOS.ROM[md5:ad4b6ec7de5f3fe165df02d832e31be4]
10c10
< F004  38                  .BYTE $38
---
> F004  57                  .BYTE $57
296c296
< F252  85 91               STA ERROR
---
> F252  85 90               STA STATUS
329c329
< F29A  4C 91 F1            JMP $F191
---
> F29A  20 91 F1            JSR $F191
336c336
< F2AA  A9 01               LDA #1
---
> F2AA  A9 00               LDA #0
669c669
< F552  A9 88               LDA #$88
---
> F552  A9 82               LDA #$82
748c748
< F602  A9 A8               LDA #$A8
---
> F602  A9 A2               LDA #$A2
798c798
< F670  A9 88               LDA #$88
---
> F670  A9 82               LDA #$82
832c832
< F6BE  20 9D F2            JSR $F29D
---
> F6BE  20 A0 F2            JSR $F2A0
1433,1434c1433,1434
< FBD7  E6 B1               INC SBUF
< FBD9  C6 B2               DEC $B2
---
> FBD7  E6 B2               INC $B2
> FBD9  C6 B1               DEC SBUF
1457,1458c1457,1458
< FC07  E6 B1               INC SBUF
< FC09  C6 B2               DEC $B2
---
> FC07  E6 B2               INC $B2
> FC09  C6 B1               DEC SBUF
1834c1834
< FF1B  29 F7               AND #$F7
---
> FF1B  29 7F               AND #$7F
1876c1876
< FF7F  A9 88               LDA #$88
---
> FF7F  A9 82               LDA #$82
1893c1893
< FFA5  A9 88               LDA #$88
---
> FFA5  A9 82               LDA #$82
1952c1952
< FFF9  4C                  .BYTE $4C
---
> FFF9  4B                  .BYTE $4B