Image Map Image Map
Page 1 of 4 1234 LastLast
Results 1 to 10 of 33

Thread: ROM binaries for a PET 2001 clone

  1. #1

    Default ROM binaries for a PET 2001 clone

    Hi guys, my 1st post here.

    My name is Glen and I am currently building a clone of an original PET 2001 entirely in current-production CMOS logic (CPU, VIA and PIAs from the WDC line). I am currently working on the PCB layout and will be making the design publicly available when finished. You and see the initial prototype construction in action on Youtube here:

    https://www.youtube.com/watch?v=b66FU2z5w5w


    Now a I have a question about the firmware ROMS. My prototype is currently loaded with both BASIC#1 and BASIC#2 (BASIC#2 being identical to BASIC#1 except for an upgraded $C000-$CFFF ROM).
    I can basically have multiple firmware versions in ROM and select the one to run with a switch. I found these binary ROMs here on Bo Zimmerman's site (scroll down to the bottom of the page):

    http://www.zimmers.net/anonftp/pub/c...computers/pet/

    Now the BASIC#1 and BASIC#2 sets specified for the 2001 work on my prototype perfectly OK. Both versions display **** COMMODORE BASIC **** upon boot up.

    However I cannot get the binary files for BASIC#3 to so much as successfully boot. Would I be right in assuming that the binaries here called BASIC#3 are the upgraded ones that alternatively display #### COMMODORE BASIC #### ?

    Note that my project is a hardware clone of the 2001x, not the 2001xN and I am not sure if the #### BASIC was even compatible with this model, so I don't know if that is the issue of if the #3 binary files on Bo's site are corrupt or mislabeled.

    If anyone knowledgeable could advise on the ROM version compatibility an/or suggest an alternative source of #### ROM binaries that are known to be good I'd very much appreciate it.

    Cheers.

  2. #2
    Join Date
    Dec 2005
    Location
    Toronto ON Canada
    Posts
    6,725

    Default

    Should work; I also had a 2001 PET that I modified to select both versions with a switch.

    Which specific images are you using (part numbers)? You are aware that there are both 2K and 4K images of BASIC 3 there, as well as different 'E' ROM versions?

    m

  3. #3

    Default

    Hmm, I can decipher two sets of BASIC 3. These are the seven 2kbyte ROM files. Organised into sets the files are:

    rom-3-c000.901439-13.bin
    rom-3-c800.901439-14.bin
    rom-3-d000.901439-15.bin
    rom-3-d800.901439-16.bin
    rom-3-e000.901439-17.bin
    rom-3-f000.901439-18.bin
    rom-3-f800.901439-19.bin


    rom-3-c000.901447-20.bin
    rom-3-c800.901447-21.bin
    rom-3-d000.901447-22.bin
    rom-3-d800.901447-23.bin
    rom-3-e000.901447-24.bin
    rom-3-f000.901447-25.bin
    rom-3-f800.901447-26.bin


    I do not know what the numbers 901439 and 901447 refer to. The serial number of the computer or the motherboard revision that the binary files were copies from? I'm reasonably sure that the last two digits are the IC number designations.

    I've done a file compare between the two sets however and they are 100% identical (ditto for the 901439 and 901447 file sets provided for BASIC 1 & 2).

    Which files are BASIC 3 4k images and the alternate E ROM?

  4. #4
    Join Date
    Dec 2005
    Location
    Toronto ON Canada
    Posts
    6,725

    Default

    The original PET came in four versions, using standard 2316 or 6540 2K ROMs and standard 2114 or 6550 1/2K SRAMs; 901439 and 901447 (and 901465) are the IC type.

    The 901439 series chips are 6540s while the 901447 series are 2316s (2716 compatible); the respective contents are identical (as you found).

    The later 2001N used 4K chips (2532 compatible): 901465-01, -02 and -03 for Cxxx, Dxxx and Fxxx respectively while the Exxx ROM was only 2K, the same 901447-24 as above; again, the contents are the same as the 2K versions.

    The suffix numbers do indeed identify the particular ROM (contents); note that unlike the rest the suffix numbers of the BASIC 1 901439-01 to -07 set are in sequence not by address but by location, in case that's relevant.

    m

  5. #5

    Default

    OK, thanks, I'm starting to sort this out now. I have (from Bo's site) schematics for all four; three 2001 variants + the 20001 xN. On the schematic diagrams for the three variants of the 2001 the 2kb ROM chips have identical part # designators regardless of type, which match the suffixes applied to the BASIC1/2 ROM image files on Bo's site and, as you note, are not in sequential order.

    However the suffixes of the BASIC 3 ROM images provided do not match the schematic diagrams and are just 13 through 19 sequentially for the 901439 followed, again sequentially, by 20 through 26 for the 901447.

    Unless there were additional PCB+schematic revisions of the 2001 predating the 2001-N (assuming that the 2001-N only ever used 4kb ROMs) which used the designators 13 through 19 for 901439 ROMs and 20 through 26 for 901447 ROMs, then perhaps I can assume that then the fourteen BASIC 3 ROM files were collated they were just arbitrarily numbered 13 through 26?

    If that is the case the maybe human error crept in at some stage and the ROM files were mislabeled and put in the wrong order. I'm a bit stuffed right now from a long day at work, but if my basic math is right that gives me 37 different possible combinations of ordering of the BASIC 3 ROM files to burn to EEPROM to try.

  6. #6

    Default

    Quote Originally Posted by GK2001 View Post
    Chapeau! I only wonder how your keyboard works.
    With kind regards / met vriendelijke groet, Ruud Baltissen
    www.Baltissen.org

  7. #7
    Join Date
    Feb 2009
    Location
    Southern California, USA
    Posts
    2,321

    Default

    Nice!
    Yes, keyboard design will be interesting.
    -Dave

  8. #8
    Join Date
    Dec 2005
    Location
    Toronto ON Canada
    Posts
    6,725

    Default

    No, all four versions of the 2001 are on Bo's site:
    320008 - 6540/6550
    320081 - 2316/6550
    320132 - 6540/2114
    320137 - 2316/2114

    and the 2001N:
    320349 - 6332

    The original BASIC 1 ROMs were numbered according to location on the board; note that because part of the address decoding is done in the ROM, low and high 2K segments are interchangeable on the board.

    But the BASIC3 ROMs are numbered according to memory address as you list them above, so there shouldn't really be any confusion. I created my 256KB 'combo' ROM from the files on Bo's site and don't recall any discrepancies in the numbers.

    There are PS/2 to PET/C64 keyboard adapters out there but I suppose you'd rather make your own.

    Very impressive project BTW!

    m
    Last edited by MikeS; July 18th, 2017 at 04:28 PM.

  9. #9

    Default

    Too late now, but you could have saved a whole lot of trouble by using just the RUN key instead of all those other keypresses.

    Why does the text appear to display in the wrong order? Do all PETs do this and I haven't noticed?

    Excellent work though.
    Be polite and I may let you live.

  10. #10

    Default

    Quote Originally Posted by MikeS View Post
    No, all four versions of the 2001 are on Bo's site:
    320008 - 6540/6550
    320081 - 2316/6550
    320132 - 6540/2114
    320137 - 2316/2114

    and the 2001N:
    320349 - 6332

    The original BASIC 1 ROMs were numbered according to location on the board; note that because part of the address decoding is done in the ROM, low and high 2K segments are interchangeable on the board.

    But the BASIC3 ROMs are numbered according to memory address as you list them above, so there shouldn't really be any confusion. I created my 256KB 'combo' ROM from the files on Bo's site and don't recall any discrepancies in the numbers.

    There are PS/2 to PET/C64 keyboard adapters out there but I suppose you'd rather make your own.

    Very impressive project BTW!

    m

    OK, I missed one of the four. Bo's site is an excellent resource but the organisation leaves a bit to be desired. But anyway the schematic I missed still uses the exact same ROM chip designators as the other three. Each BASIC1/2 ROM file is labelled with the board/schematic chip designator AND the address location (specifically the start address):

    $C000.xxxxxx-(H)1
    $C800.xxxxxx-(H)5
    $D000.xxxxxx-(H)2
    $D800.xxxxxx-(H)6
    $E000.xxxxxx-(H)3
    $F000.xxxxxx-(H)4
    $F800.xxxxxx-(H)7

    So this still doesn't explain why the BASIC 3 file sets were given the suffixes 13 through 19 and 20 through 26 rather than 1, 5, 2, 6, 3, 4, 7.

    Thanks for confirming that the BASIC 3 ROM files worked for you. That perhaps means that I have some sort of hardware incompatibility to track down.

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •