1541 ROMs

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

1541 ROMs

William Levak

I have compared all the 1541 ROMs. The results are below. I have list the
-02 and -06 ROMs as revisons 1a and 2a because the are almost identical to
revisons 1 and 2, and because Commodore didn't think they were significant
enough to release world wide.


         ROMs          Revision
325302-01, 901229-01     r0
325302-01, 901229-02     r1a Japan
325302-01, 901229-03     r1
325302-01, 901229-05     r2
325302-01, 901229-06     r2a Europe
      251968-01           r3  1541B/C
      251968-03           r4  1541-II

      Number of differences between ROMs

901229-02 |  14
901229-03 |  16    3
901229-05 |  93   86   84
901229-06 | 102   95   93   10
251968-01 | 179  172  170   87   78
            ------------------------
901229-      01   02   03   05   06

901229-02 and previous versions use a header gap of 8 bytes. 901229-03 and
later versions use a header gap of 9 bytes. This is the only difference
between 901229-02 and 901229-03.

901229-04 is unknown. 901229-05 changes some of the serial bus routines,
and NOPs out the auto start routine. 901229-04 could potentially make only
some of these changes, if it exists.

901229-06 adds a patch that sets location $51 (current track) to $FF before
formatting a disk. This is apparently to recover from a prior failed format
attempt. This is the only difference between 901229-05 and 901229-06.

901229-07 is known only from the parts list of the 1541B/C. It is probably
the same as the upper half of 251968-01.

251968-03 is substantially different, even in the $C000-DFFF region.

[hidden email]
SDF Public Access UNIX System - http://sdf.lonestar.org

       Message was sent through the cbm-hackers mailing list
Reply | Threaded
Open this post in threaded view
|

Re: 1541 ROMs

Gerrit Heitsch
On 12/10/2014 06:56 AM, William Levak wrote:

>
> I have compared all the 1541 ROMs. The results are below. I have list
> the -02 and -06 ROMs as revisons 1a and 2a because the are almost
> identical to revisons 1 and 2, and because Commodore didn't think they
> were significant enough to release world wide.
>
>
>          ROMs          Revision
> 325302-01, 901229-01     r0
> 325302-01, 901229-02     r1a Japan
> 325302-01, 901229-03     r1
> 325302-01, 901229-05     r2
> 325302-01, 901229-06     r2a Europe
>       251968-01           r3  1541B/C
>       251968-03           r4  1541-II

You forgot 251968-02.

Can be found here:

http://www.zimmers.net/anonftp/pub/cbm/firmware/drives/new/1541/index.html

  Gerrit



       Message was sent through the cbm-hackers mailing list
Reply | Threaded
Open this post in threaded view
|

RE: 1541 ROMs

Martin Hoffmann-Vetter
In reply to this post by William Levak
Hello William,

> I have list the -02 and -06 ROMs as revisons 1a and 2a because the are
> almost identical to revisons 1 and 2, and because Commodore didn't
> think they were significant enough to release world wide.

>          ROMs          Revision
> 325302-01, 901229-01     r0
> 325302-01, 901229-02     r1a Japan
> 325302-01, 901229-03     r1
> 325302-01, 901229-05     r2
> 325302-01, 901229-06     r2a Europe

I think it's known that the revision part of the part number is chronology.
So the question is, why is p/n 901229-06 with revision 2a after revision 2
with p/n 901229-05 when p/n 901229-02 with revision 1a is before revision 1
with p/n 901229-03?

So the 901229-02 must be a r0a or the 901229-02 is r1 and 901229-03 must be
r1a!

> 901229-07 is known only from the parts list of the 1541B/C.

Ist this the only reference?

Martin



       Message was sent through the cbm-hackers mailing list
Reply | Threaded
Open this post in threaded view
|

RE: 1541 ROMs

William Levak
On Wed, 10 Dec 2014, Martin Hoffmann-Vetter wrote:

> Hello William,
>
>> I have list the -02 and -06 ROMs as revisons 1a and 2a because the are
>> almost identical to revisons 1 and 2, and because Commodore didn't
>> think they were significant enough to release world wide.
>
>>          ROMs          Revision
>> 325302-01, 901229-01     r0
>> 325302-01, 901229-02     r1a Japan
>> 325302-01, 901229-03     r1
>> 325302-01, 901229-05     r2
>> 325302-01, 901229-06     r2a Europe
>
> I think it's known that the revision part of the part number is chronology.
> So the question is, why is p/n 901229-06 with revision 2a after revision 2
> with p/n 901229-05 when p/n 901229-02 with revision 1a is before revision 1
> with p/n 901229-03?

Because there are only three bytes different between r1a and r1.  That is,
r1a is closest to r1.  If you want it to be strictly chronological, then
it should be r0z and r1.

> So the 901229-02 must be a r0a or the 901229-02 is r1 and 901229-03 must be
> r1a!
>
>> 901229-07 is known only from the parts list of the 1541B/C.
>
> Ist this the only reference?

It is the only reference I can find.

> Martin
>
>
>
>       Message was sent through the cbm-hackers mailing list
>

[hidden email]
SDF Public Access UNIX System - http://sdf.lonestar.org

       Message was sent through the cbm-hackers mailing list
Reply | Threaded
Open this post in threaded view
|

Re: 1541 ROMs

William Levak
In reply to this post by Gerrit Heitsch
On Wed, 10 Dec 2014, Gerrit Heitsch wrote:

> On 12/10/2014 06:56 AM, William Levak wrote:
>>
>> I have compared all the 1541 ROMs. The results are below. I have list
>> the -02 and -06 ROMs as revisons 1a and 2a because the are almost
>> identical to revisons 1 and 2, and because Commodore didn't think they
>> were significant enough to release world wide.
>>
>>
>>          ROMs          Revision
>> 325302-01, 901229-01     r0
>> 325302-01, 901229-02     r1a Japan
>> 325302-01, 901229-03     r1
>> 325302-01, 901229-05     r2
>> 325302-01, 901229-06     r2a Europe
>>       251968-01          r3  1541B/C
          251968-02          r4  1541C
          251968-03          r5  1541-II
>
> You forgot 251968-02.

I didn't forget it.  I didn't know it existed.

It is substantially different than both -01 and -02. I have corrected the
table above.

> Can be found here:
>
> http://www.zimmers.net/anonftp/pub/cbm/firmware/drives/new/1541/index.html
>
> Gerrit
>
>
>
>      Message was sent through the cbm-hackers mailing list
>

[hidden email]
SDF Public Access UNIX System - http://sdf.lonestar.org

       Message was sent through the cbm-hackers mailing list
Reply | Threaded
Open this post in threaded view
|

Re: 1541 ROMs

Spiro Trikaliotis
In reply to this post by William Levak
Hello,

* On Wed, Dec 10, 2014 at 05:56:46AM +0000 William Levak wrote:
 
> 901229-04 is unknown. 901229-05 changes some of the serial bus routines,

Which changes do you mean? Having a quick glance into my sources
(http://cbmrom.trikaliotis.net/), I only find the patches at $FF10 and
$FF20.

Or have I overlooked something more important?

 
> 251968-03 is substantially different, even in the $C000-DFFF region.

It is not so much different as one might think if you only look at the
bytes. It moves around some of the patches, which makes it appear as if
there were more differences.

For example, it adds a copyright notice at $C000 (cf.
http://cbmrom.trikaliotis.net/listings/1541-ii-03.lst, search for
00C002), which moves the patch routines that start at $C002 with
previous ROMs to higher locations ($C04E).

Regards,
Spiro.

--
Spiro R. Trikaliotis
http://www.trikaliotis.net/

       Message was sent through the cbm-hackers mailing list
Reply | Threaded
Open this post in threaded view
|

Re: 1541 ROMs

William Levak
On Sat, 13 Dec 2014, Spiro Trikaliotis wrote:

> Hello,
>
> * On Wed, Dec 10, 2014 at 05:56:46AM +0000 William Levak wrote:
>
>> 901229-04 is unknown. 901229-05 changes some of the serial bus routines,
>
> Which changes do you mean? Having a quick glance into my sources
> (http://cbmrom.trikaliotis.net/), I only find the patches at $FF10 and
> $FF20.
>
> Or have I overlooked something more important

Below is my comparison of the ROMs.

Comparing files 901229-03 and 901229-05
E683: 20 EA    autostart NOP'ed out
E684: 4E EA
E68B: 20 EA
E68C: 4E EA
E780: AD 60    RTS
E781: 00 EA
E782: 18 EA
E783: AA EA
E784: 29 EA
E785: 04 EA
E786: F0 EA
E787: F7 EA
E788: 8A EA
E789: 29 EA
E78A: 01 EA
E78B: F0 EA
E78C: F2 EA
E78D: 58 EA
E78E: AD EA
E78F: 00 EA
E790: 18 EA
E791: 29 EA
E792: 05 EA
E793: D0 EA
E794: F9 EA
E795: EE EA
E796: 78 EA
E797: 02 EA
E798: EE EA
E799: 74 EA
E79A: 02 EA
E79B: A9 EA
E79C: 2A EA
E79D: 8D EA
E79E: 00 EA
E79F: 02 EA
E7A0: 4C EA
E7A1: A8 EA
E7A2: E7 60    RTS

E9DC: 8D 4C    JMP FF20
E9DD: 05 20
E9DE: 18 FF

EAA4: 8E 4C    JMP FF10
EAA5: 03 10
EAA6: 18 FF

EBDA: A9 A9                               6522 Serial I/O
EBDB: 1A 00    LDA 1A      LDA 00      changes register setup
EBDC: 8D 8D                            probably because of
EBDD: 02 00    STA 1802    STA 1800    elimination of autostart
EBDE: 18 18
EBDF: A9 A9
EBE0: 00 1A    LDA A9      LDA 00
EBE1: 8D 8D
EBE2: 00 02    STA 1800    STA 1802
EBE3: 18 18

FEE6: 0E 3E    checksum

FF10: AA 8E    patch for reset routine
FF11: AA 03
FF12: AA 18
FF13: AA A9
FF14: AA 02
FF15: AA 8D
FF16: AA 00
FF17: AA 18
FF18: AA A9
FF19: AA 1A
FF1A: AA 8D
FF1B: AA 02
FF1C: AA 18
FF1D: AA 4C
FF1E: AA A7
FF1F: AA EA

FF20: AA AD    patch for LISTEN to serial bus
FF21: AA 00
FF22: AA 18
FF23: AA 29
FF24: AA 01
FF25: AA D0
FF26: AA F9
FF27: AA A9
FF28: AA 01
FF29: AA 8D
FF2A: AA 05
FF2B: AA 18
FF2C: AA 4C
FF2D: AA DF
FF2E: AA E9

FFE2: 52 AA    'RSR' replaced with fill bytes
FFE3: 53 AA     Was this a signature?
FFE4: 52 AA

[hidden email]
SDF Public Access UNIX System - http://sdf.lonestar.org

       Message was sent through the cbm-hackers mailing list
Reply | Threaded
Open this post in threaded view
|

RE: 1541 ROMs

William Levak
In reply to this post by William Levak
On Thu, 11 Dec 2014, William Levak wrote:

> On Wed, 10 Dec 2014, Martin Hoffmann-Vetter wrote:
>
>> Hello William,
>>
>>> I have list the -02 and -06 ROMs as revisons 1a and 2a because the are
>>> almost identical to revisons 1 and 2, and because Commodore didn't
>>> think they were significant enough to release world wide.
>>
>>>          ROMs          Revision
>>> 325302-01, 901229-01     r0
>>> 325302-01, 901229-02     r1a Japan
>>> 325302-01, 901229-03     r1
>>> 325302-01, 901229-05     r2
>>> 325302-01, 901229-06     r2a Europe
>>
>> I think it's known that the revision part of the part number is chronology.
>> So the question is, why is p/n 901229-06 with revision 2a after revision 2
>> with p/n 901229-05 when p/n 901229-02 with revision 1a is before revision 1
>> with p/n 901229-03?
>
> Because there are only three bytes different between r1a and r1.  That is,
> r1a is closest to r1.  If you want it to be strictly chronological, then it
> should be r0z and r1.

If you don't like those suggestions, how about calling it r1 beta.



[hidden email]
SDF Public Access UNIX System - http://sdf.lonestar.org

       Message was sent through the cbm-hackers mailing list
Reply | Threaded
Open this post in threaded view
|

Re: 1541 ROMs

Nate Lawson
In reply to this post by William Levak

> On Dec 26, 2014, at 10:51 PM, William Levak <[hidden email]> wrote:
>
> FFE2: 52 AA    'RSR' replaced with fill bytes
> FFE3: 53 AA     Was this a signature?
> FFE4: 52 AA

I believe it’s for “Robert S. Russell.” The “RRBY” in the C64 was for Robert Russell and Bob Yannes. Russell has said he wanted to put “RSR” in the C64 ROM but he didn’t want to waste another byte.

-Nate


       Message was sent through the cbm-hackers mailing list