OS source code - all revisions

[Kr0tki]

Update!

 

Atari_Ace noticed some typos, and also pointed me to a source listing of John Palevich's "Shoot" published in Compute!, which included the OS equates. This allowed me to a few lines that were truncated in the original source listings. As usual, remove the .txt extension from the filename, then unpack with 7-zip, and see Changelog and README for details.

 

[Kr0tki]

I have noticed that Atarimania, in its internal documents section, has scans of some preliminary internal documents that contain full source code to 400/800 OS rev. A - the only version with no truncated lines. This version is slightly different to the one from "Operating System Source Listing". I have added the new source listing to the archive, among some other changes.

 

As usual, remove the .txt extension from the filename, then unpack with 7-zip, and see Changelog and README for details.

a8-os-rom-2017-05-05.7z.txt

Edited May 5, 2017 by Kr0tki

[+kheller2]

2.2.6. BB000002 Rev. 3 (1984-02-22)

Called "Revision 3, Version 2" in the source comments; the date given there is, quite mysteriously, earlier than that of the original Rev. 3.[OSREV4]

This was a further enhancement for the 600XL/800XL/1450XLD computers, which also never reached the market. It was developed by Olivia Ying-Tzu Jang and Vincent H. Wu, and mainly included enhancements for the PBI routines, probably to support the 1450XLD’s built-in modem, voice synthesis and disk drive:

Dedicate PDVI ($D1FF) to external parallel device IRQ status

Dedicate IPDVI ($D1CF) to internal parallel device IRQ status

Using PDIMSK ($0249) for external parallel device IRQ selection mask

Using IPDIMK ($0254) for internal parallel device IRQ selection mask

After masking (PDVI, PDIMSK) & (IPDVI, IPDIMK), OR the result

together, piror to processing parallel device IRQ

 

On cold start, initialize PDVI = 0, to avoid potential

checksum error.[OSREV4]

No machine with this OS version has been found yet.

 

Hmmm...

 

I wonder if this is that version. Not sure I still have this board, will have to dig.. This ROM sticker says R3V2 4/16

 

[Kr0tki]

If you ever find it, please dump the ROM (and the V/T Handler ROM too, if possible). I'll do the disassembling.

[AtariGeezer]

This is what I have.

 

1450_OS_ROMS.atr

1450 Dis6502 WorkBooks.zip

 

There is 1 OS Rom for the 1400XL and 2 OS Roms for the 1450XL (3.0 and 3.X). and 2 versions of U6 V & T Handlers.

The Dis6502 Workbooks has the V & T Handlers and the 1450 OS 3.0 OS that I've been working on which should be a good place to start out

[Kr0tki]

On the 1450_OS_ROMS.atr above:

- 1450OS30.ROM and 1450OS3X.ROM are identical - they are BB000002 Rev. 3 (1984-06-21) - already known.

- 1400XL1.ROM is a bad dump - from $0000 to $384E it's an exact copy of 600XL OS rev. 1, and from $384F to the end it's a DOS binary file with part of 400/800 OS Rev. B.

Edited May 6, 2017 by Kr0tki

[AtariGeezer]

Okay, guess I copied the same OS twice. on the actual roms, 3.0 has a date code of 1984-03-23 BB000002 and 3.X has the date code you mentioned...

[Kr0tki]

Still, both ROMs are nothing new, sorry.

[AtariGeezer]

Understood, after looking at your previos post, I now know how to check the date code and version info

[+kheller2]

If you ever find it, please dump the ROM (and the V/T Handler ROM too, if possible). I'll do the disassembling.

 

Found it. Will dump this weekend.

[+kheller2]

Um.. please check this OS version.. I don't have a matching md5 and want to make sure I didn't dump or copy it incorrectly.

 

 

 

MD5 (OSR3V2-416.BIN) = 2dbc73da0d34994d1e2e62e22eb49224

MD5 (VTHANB42D.BIN) = d302d5fcd7abc1db933b9894ae8e5f8e

VTHANB42D.BIN This is marked "V.T HANDLER REV. C B42D"

OSR3V2-416.BIN This is marked "OS REV3 VER2 4/16"

here all all the known ones compiled from the distro:

MD5 (800antsc.rom) = a3c1585b5d19719f8acfa2b093bea75f

MD5 (800apal.rom) = eb1f32f5d9f382db1bbfb8d7f9cb343a

MD5 (800bntsc.rom) = 4177f386a3bac989a981d3fe3388cb6c

MD5 (800bpal.rom) = 89d5e5f4713267667ab713449944f8a9

MD5 (aa0r10.rom) = e3e8c74bfe1dcd6b56af50bd9a82dc15

MD5 (aa1r11.rom) = eacb8069c45e2ec4e0a19978bf2fc334

MD5 (bb0r1.rom) = 9aea45e724d2588fbbeda658c7dc53ee

MD5 (bb1r2.rom) = 06daac977823773a3eea3422fd26a703

MD5 (bb1r3.rom) = 54e704558a6aedfc45cebf8f8ac9c312

MD5 (bb1r4.rom) = b7a2a04677d34f069eeb643d5238bf86

MD5 (bb1r59.rom) = d467f55fb7643553b69b34bf7e805b7d

MD5 (bb1r59a.rom) = 69396860e53f58d798421d06d766c3ba

MD5 (bb2r3.rom) = 537cfa3038271e277d84d8f2c8a92006

MD5 (bb2r3v4.rom) = 9f5449c881475a5cca40849c743205f8

MD5 (cc1r4.rom) = 65020266380e33cce50ebf8b9d91122a

[AtariGeezer]

The VT Handler is Version 3, same as in my zip above.

 

OS shows as BB000002 Rev. 3 (1984-03-27)

Only changes are 4 bytes difference from BB000002 Rev. 3 (1984-03-23) which is the checksum, day and a change at offset $04F7 from $FF to $CF...

 

In code change from:

LDA #$FF

STA PORTB

 

to

LDA #$CF

STA PORTB

Edited May 12, 2017 by AtariGeezer

[+kheller2]

That isn't a result of my dump is it? I did a verify a few times on the content when reading. odd that the checksum would be off, or are you saying this is a valid checksum just with a different date and the main code matches 1984-03-23 exactly otherwise?

[Kr0tki]

Thank you Karl!

 

Your dump is, in fact, "Revision 3" dated 1984-03-27, whose existence is known from the source comments of Rev. 5. The dump that I previously thought to be Rev. 3 (known as 1450R3V0.ROM) is in fact a preliminary version dated 1984-03-23. Compared to your dump, the preliminary dump lacks a single bugfix, as described by AtariGeezer above.

 

I incorporated the source listing of your dump in the package. New version released!

 

As usual, remove the .txt extension from the filename, then unpack with 7-zip, and see Changelog and README for details.

 

That isn't a result of my dump is it? I did a verify a few times on the content when reading. odd that the checksum would be off, or are you saying this is a valid checksum just with a different date and the main code matches 1984-03-23 exactly otherwise?

Your dump appears to be correct. The checksum is valid.

[Atari_Ace]

FYI, there's a significantly different assembly listing for 800-fpp.asm from The Atari BASIC Source Book, which likely predates the reconstructed one. There's an OCR of it at http://users.telenet.be/kim1-6502/6502/p3.html, which I've edited somewhat and placed at https://ksquiggle.neocities.org/absb/p3.html.

 

[+JAC!]

Today Kevin Savetz sent me the OS DUMP for his 2nd Arabic 65 XE (Thanks Kevin!). It is equal to a dump I have from Nir Dary.

From what I see, based on the number of differences between the dump, I'd say it is based on the SYSROM_BB01R3 / OS Rev 3 (1985) (Atari) (800XE-65XE-130XE).rom

1073 different bytes which mainly consist of charset changes, and changes in the editor at first glance.

 

MD: 69396860e53f58d798421d06d766c3ba
CRC32: 0xf0a236d3
Check Sums (Correct): $C000/$C001=$4b/$f2, $FFF8/$FFF9=$dd/$71

 

Arabic 65 XE 2 OS.zip

[Kr0tki]

Thanks Peter. This ROM is identical to the first Arabic dump that Kevin Savetz published back in 2003 - check its description in README.html, chapter "BB000001 Rev. 59 - enhanced (1987-07-21)".

 

Could you ask Kevin if this new ROM dump came from a different machine than the one from 2003? And what is the ROM chip's part number?

[Savetz]

I have two Arabic 65XEs. The "old" one I got in 1999 from Best Electronics. (Story here: https://savetz.com/vintagecomputers/arabic65xe/) I dumped the ROM (available at that link) around that time.

 

The "new" one I got in 2013 from a guy named Yazeed through an email exchange. The ROM dump from a couple of posts up is from this machine.

 

Here are photos of the innards of the new machine.

 

 

-Kevin

[luckybuck]

Thank you sooo much Kevin, the database is going more and more complete. :-))))

 

Thank you!

Edited September 4, 2018 by luckybuck

[+JAC!]

Thanks Peter. This ROM is identical to the first Arabic dump that Kevin Savetz published back in 2003 - check its description in README.html, chapter "BB000001 Rev. 59 - enhanced (1987-07-21)".

 

Could you ask Kevin if this new ROM dump came from a different machine than the one from 2003? And what is the ROM chip's part number?

 

I don't see that it's identical. The ROM I have posted above from Kevins 2nd machine it totally different from the one on his web site.

 

MD5 69396860e53f58d798421d06d766c3ba

CRC32 0xf0a236d3

Picture C101700-002C 1988 (this is given as the same!)

 

... or I'm mixing something up completely, but I doubt it.

[Kr0tki]

Thank you Kevin!

 

I have seen one other photo of this revision of the OS, and it also was an EPROM with a paper label, instead of a mask ROM as in the case of the earlier Arabic ROM.

 

One more question - do you know which Arabic country did it come from?

 

I don't see that it's identical. The ROM I have posted above from Kevins 2nd machine it totally different from the one on his web site.

Both downloadable ROMs on Kevin's page are corrupted. They were fine when I downloaded them some twelve years ago, and a non-corrupted copy also ended up in version 3.3 of Freddy Offenga's ROM collection. You should have read my README when I told you to :-)

[Savetz]

> One more question - do you know which Arabic country did it come from?

 

No.

> Both downloadable ROMs on Kevin's page are corrupted.

 

Sigh. Do I need to re-dump my "old" Arabic machine (and take a picture of its innards too)?

 

Kevin

[Kr0tki]

Nah, no need to redump anything. As I said, I had downloaded both ROMs back when they were still correct. The file ARABICROM would be identical to the one in post #66 above, and BASICROM was just a standard Rev. C BASIC.

[_The Doctor__]

perhaps, Mr. Kr0tki we all would like access to the roms again

[Kr0tki]

I'm not sure what you mean. How can I help you?