IMSAI CP/M BIOS research - CP/M

This is a discussion on IMSAI CP/M BIOS research - CP/M ; Hi. I'm trying to restore an IMSAI 8080 with a Z80 CPU that used to run CP/M. It hasn't been used in many years and there is no documentation on the last working system configuration. The system came with several ...

+ Reply to Thread
Results 1 to 20 of 20

Thread: IMSAI CP/M BIOS research

  1. IMSAI CP/M BIOS research

    Hi. I'm trying to restore an IMSAI 8080 with a Z80 CPU that used to
    run CP/M. It hasn't been used in many years and there is no
    documentation on the last working system configuration.

    The system came with several disk interface boards and serial boards.
    More interestingly, it came with a Bytesaver board (8K of PROM) with
    four 1K PROMs present. One is unmarked and contains the Bytemover
    software. The second is unmarked and I don't know what it contains.

    The third is located at EC00 and is marked "CP/M boot 181 48k", and
    contains 4 jump vectors (JP HHHH) to locations within the PROM. The
    fourth is installed at location F000 and is marked "Matrox". It
    contains 8 jump vectors to locations within the PROM.

    Apparently to go along with the BIOS, I have two boxes of CP/M 8"
    floppy disks. They are marked with "Munn CP/M V 2.2B 48k" and the
    names of various application programs.

    Since I don't know what perhipheral cards go with the BIOS, I had
    planned on disassembling the BIOS to figure out what cards to use and
    what ports to configure them at. Unfortunately, the BIOS PROMS don't
    seem to be structured the way I expected
    (http://www.seasip.demon.co.uk/Cpm/bios.html) so I'm a bit stuck. If
    anyone has any suggestions on how to interpret the vectors or PROM
    arrangements, I'd appreciate a point in the right direction!

    Thanks.


  2. Re: IMSAI CP/M BIOS research

    The BIOS proms are not necessarily part of CP/M once it boots. They may
    only be boot code.

    The key here, which you did not touch on at all, is the disk drive
    controller. The boot code has to be specific to that. Of secondary
    importance, irrelevant for booting but not for running, is the I/O
    device (keyboard & screen) and it's interface board(s).

    It's very possible that you may have to start from scratch. Got an
    EPROM ERASER?


    nonmaskable wrote:
    > Hi. I'm trying to restore an IMSAI 8080 with a Z80 CPU that used to
    > run CP/M. It hasn't been used in many years and there is no
    > documentation on the last working system configuration.
    >
    > The system came with several disk interface boards and serial boards.
    > More interestingly, it came with a Bytesaver board (8K of PROM) with
    > four 1K PROMs present. One is unmarked and contains the Bytemover
    > software. The second is unmarked and I don't know what it contains.
    >
    > The third is located at EC00 and is marked "CP/M boot 181 48k", and
    > contains 4 jump vectors (JP HHHH) to locations within the PROM. The
    > fourth is installed at location F000 and is marked "Matrox". It
    > contains 8 jump vectors to locations within the PROM.
    >
    > Apparently to go along with the BIOS, I have two boxes of CP/M 8"
    > floppy disks. They are marked with "Munn CP/M V 2.2B 48k" and the
    > names of various application programs.
    >
    > Since I don't know what perhipheral cards go with the BIOS, I had
    > planned on disassembling the BIOS to figure out what cards to use and
    > what ports to configure them at. Unfortunately, the BIOS PROMS don't
    > seem to be structured the way I expected
    > (http://www.seasip.demon.co.uk/Cpm/bios.html) so I'm a bit stuck. If
    > anyone has any suggestions on how to interpret the vectors or PROM
    > arrangements, I'd appreciate a point in the right direction!
    >
    > Thanks.
    >


  3. Re: IMSAI CP/M BIOS research

    Perhaps the PROMS are only boot code; would the boot routine normally
    be 2 or 3K? In a system like this would the full BIOS normally come in
    off of the floppy?

    For the serial board, I have a Cromemco TUART and a Jade 1p2s board.
    For the floppy controller, I have a board I cannot identify (marked
    "8800 Interface BD") and a Tarbell 1011 Revision D board. The floppy
    drives themselves are missing, but I have replacements to work with.

    I was hoping to first figure out the serial board and bios routines and
    use it to transfer the proms to a computer for disassembly. Reading
    the front panel gets tiring after a while!

    I _could_ start from scratch but I'd like to first exhaust all means of
    getting the floppies I have to boot, which I think means figuring out
    what is on the proms.


  4. Re: IMSAI CP/M BIOS research

    First of all, a serial console connects to ONE serial card, period.
    Maybe the Tu-ART, maybe the 1p2s board, but one or the other.

    I have no idea what the 8800 Interface bd is, perhaps a card for the old
    Altair disk system (just a guess). But I know exactly what a Tarbell
    1011 board is, and that's the board that you want, it's a really good
    disk controller (single density only, unfortunately, but at least a good
    single density card). The Tarbell card has a built in boot system and
    boot code prom (on 32 bytes total, that's all it takes, and it goes away
    after booting). It uses a Western Digital 1771 chip, which is well
    known and well supported, with lots of documentation available. The
    manual for the Tarbell board is available in PDF format on the Harte
    site. All you need for a system is:

    -CPU card
    -Memory Card(s) (preferably static)
    -Tarbell disk controller
    -serial I/O Card

    Unless you know that there are software routines on the proms that you
    want to use, I would pull everything else. Use a PC as a serial terminal.

    But the problem is that you need a boot disk with a compatible BIOS.
    And that's a real bitch if you don't have one and don't have another
    [working] system on which to prepare one. It's easiest if you have a
    really, really dumb serial card that does not require any software
    initialization. The Tu-Art, unfortunately, takes quite a bit of
    software initialization. I am not personally familiar with the Jade card.

    nonmaskable wrote:

    > Perhaps the PROMS are only boot code; would the boot routine normally
    > be 2 or 3K? In a system like this would the full BIOS normally come in
    > off of the floppy?
    >
    > For the serial board, I have a Cromemco TUART and a Jade 1p2s board.
    > For the floppy controller, I have a board I cannot identify (marked
    > "8800 Interface BD") and a Tarbell 1011 Revision D board. The floppy
    > drives themselves are missing, but I have replacements to work with.
    >
    > I was hoping to first figure out the serial board and bios routines and
    > use it to transfer the proms to a computer for disassembly. Reading
    > the front panel gets tiring after a while!
    >
    > I _could_ start from scratch but I'd like to first exhaust all means of
    > getting the floppies I have to boot, which I think means figuring out
    > what is on the proms.
    >


  5. Re: IMSAI CP/M BIOS research

    The system has both serial boards, meaning I don't know which one goes
    with the BIOS and which one doesn't. I also don't know any of the
    ports since they were all randomly reset.

    The "8800 Interface Bd" is something designed to work with an oddball
    type drive (such as an ALTAIR or IMSAI) because there is no FDC chip on
    it. So, yes the Tarbell would be preferable.

    My reading of the Tarbell board BIOS is that it loads the first sector
    in at 0000 and then jumps to it. Could the floppies contain the BIOS
    for the system, and the PROMS be something entirely different? Based
    on the PROM labels and the jump table at the beginning of them, I have
    been assuming they are a CP/M BIOS chopped into two 1K sections somehow.


  6. Re: IMSAI CP/M BIOS research

    The Tarbell boot code does indeed read Sector 1 of Track zero into
    memory and then jumps to it. Normally the diskette itself contains the
    entire OS, including the bios, and there is no bios at all in the
    computer hardware itself (e.g. no ROM BIOS like you have in a PC). But
    some systems worked differently. What you have to understand is that in
    that world, every system was different from every other system. Even to
    nominally identical systems (same boards) might require different boot
    disks if the ports were configured differently.


    nonmaskable wrote:

    > The system has both serial boards, meaning I don't know which one goes
    > with the BIOS and which one doesn't. I also don't know any of the
    > ports since they were all randomly reset.
    >
    > The "8800 Interface Bd" is something designed to work with an oddball
    > type drive (such as an ALTAIR or IMSAI) because there is no FDC chip on
    > it. So, yes the Tarbell would be preferable.
    >
    > My reading of the Tarbell board BIOS is that it loads the first sector
    > in at 0000 and then jumps to it. Could the floppies contain the BIOS
    > for the system, and the PROMS be something entirely different? Based
    > on the PROM labels and the jump table at the beginning of them, I have
    > been assuming they are a CP/M BIOS chopped into two 1K sections somehow.
    >


  7. Re: IMSAI CP/M BIOS research

    What would normally fit in sector 1 of track 0? Would it just be a
    more complete bootloader than the Tarbell 32B loader, or would it be
    the CP/M bios and command processor?

    Or was there no "universal standard" for the CP/M boot sequence?

    mike.

    Barry Watzman wrote:
    > The Tarbell boot code does indeed read Sector 1 of Track zero into
    > memory and then jumps to it. Normally the diskette itself contains the
    > entire OS, including the bios, and there is no bios at all in the
    > computer hardware itself (e.g. no ROM BIOS like you have in a PC). But
    > some systems worked differently. What you have to understand is that in
    > that world, every system was different from every other system. Even to
    > nominally identical systems (same boards) might require different boot
    > disks if the ports were configured differently.



  8. Re: IMSAI CP/M BIOS research

    Normally, Trk 0 sector 1 contains the "rest" of the boot code. The prom
    on the disk controller (the Tarbell board) reads track 0 sector 1. The
    code in track 0 sector 1 then reads the rest of tracks 0 and 1 (51
    sectors ... track 0 sectors 2 thru 26, and all of track 1 (sectors
    1-26). Those sectors classically contain a binary image of the 3
    modules of CP/M itself (CCP, BDOS, BIOS), but in some more advanced
    systems they are instead a high-level intelligent loader that can read
    actual files off of the diskette, allowing for a bios that makes the
    entire OS too big to fit in that remaining space.




    nonmaskable wrote:

    > What would normally fit in sector 1 of track 0? Would it just be a
    > more complete bootloader than the Tarbell 32B loader, or would it be
    > the CP/M bios and command processor?
    >
    > Or was there no "universal standard" for the CP/M boot sequence?
    >
    > mike.
    >
    > Barry Watzman wrote:
    >
    >>The Tarbell boot code does indeed read Sector 1 of Track zero into
    >>memory and then jumps to it. Normally the diskette itself contains the
    >>entire OS, including the bios, and there is no bios at all in the
    >>computer hardware itself (e.g. no ROM BIOS like you have in a PC). But
    >>some systems worked differently. What you have to understand is that in
    >>that world, every system was different from every other system. Even to
    >>nominally identical systems (same boards) might require different boot
    >>disks if the ports were configured differently.

    >
    >


  9. Re: IMSAI CP/M BIOS research

    On 20 Apr 2006 05:50:50 -0700, "nonmaskable"
    wrote:

    >What would normally fit in sector 1 of track 0?


    One common approach is code that calls the base boot code to read in
    additional sectors. Another was the first boot load sector 1 track 0
    and that code is the disk boot loader that reads the required
    additional sectors. A few even loaded the whole mess from rom!
    Some of the ways to do it were often goverend by the hardware
    features or lack thereof.

    >Would it just be a
    >more complete bootloader than the Tarbell 32B loader, or would it be
    >the CP/M bios and command processor?


    More complete boot loader. The CCP/DBOS/BIOS/ is typically 6.5-8kb
    in size and a single density sector is typically less than 256bytes
    (yes, exceptions exist but most never exceeded 1K).

    >Or was there no "universal standard" for the CP/M boot sequence?


    There were common schemes that were similar but universal, no.
    It was largely left to the BIOS implementor and while there was a
    skeleton there was no "required" way to do it beyond getting
    everything into ram at the right places for execution.


    Allison

  10. Re: IMSAI CP/M BIOS research


    My advice, since I've just gone through the same thing, is to forget
    what you have now, be methodical, and start over. It's easier and you
    end up with a nice, clean system that you know the reason for every
    hardware setting and piece of software code.

    - Choose the set of boards you want to use: CPU, RAM, I/O, Floppy. I
    used a mix of Cromemco and Compupro.

    - Obtain manuals for each one of these boards.

    - Using a PC and DOS CPM emulation, write your own CBIOS (not too bad
    if you have the sources provided by the board vendors).

    Trying to get an old system in an unknown state working, without
    docs, possibly without all the cards or correct disks, is nearly if not
    impossible unless you just know what you're looking at and doing, and
    already have running systems to test the boards and work with.

    ~ J

    nonmaskable wrote:
    > Hi. I'm trying to restore an IMSAI 8080 with a Z80 CPU that used to
    > run CP/M. It hasn't been used in many years and there is no
    > documentation on the last working system configuration.



  11. Re: IMSAI CP/M BIOS research

    nonmaskable wrote:
    > Perhaps the PROMS are only boot code; would the boot routine normally
    > be 2 or 3K? In a system like this would the full BIOS normally come in
    > off of the floppy?
    >
    > For the serial board, I have a Cromemco TUART and a Jade 1p2s board.
    > For the floppy controller, I have a board I cannot identify (marked
    > "8800 Interface BD") and a Tarbell 1011 Revision D board. The floppy
    > drives themselves are missing, but I have replacements to work with.
    >
    > I was hoping to first figure out the serial board and bios routines and
    > use it to transfer the proms to a computer for disassembly. Reading
    > the front panel gets tiring after a while!
    >
    > I _could_ start from scratch but I'd like to first exhaust all means of
    > getting the floppies I have to boot, which I think means figuring out
    > what is on the proms.


    All you likely need to know about the Tarbell board and any disks it
    boots CP/M from, is in the manual.

    On one of my Web pages which lists miscelaneous S-100 documents I
    offer:

    http://www.retrotechnology.com/herbs_stuff/d_misc.html

    I list the manual for your Tarbell board as;

    Tarbell Floppy disk Interface (M1011 8-inch single density), 1978, 92
    pgs

    I used it a number of times myself. It has a tiny PAL which has enough
    code to boot up the first sector of track 0. That sector has sufficient
    code to read the rest of track 0 and track 1. Those contain CP/M and
    the BIOS for a particular system. The board supports 8-inch (or 5.25
    inch) single density. You can look at your disks to see if they are
    single or double density. Again, details and listings for the Tarbell
    are in the manual.

    Your proms probably contain something other than CP/M support code, but
    who knows? If you want to read them, and assuming your IMSAI is working
    but without an operating system and is as described, here's an option.
    Write a simple program to operate your UART, get it working to just
    echo characters from the serial port. Then add some code to start
    reading bytes from some location, convert to hex, and dump them out the
    serial port. (look on the Web for "rom monitor" programs for sample
    code.) TOGGLE IN these programs to test and operate. In the old days,
    programs would read the front panel switches to control their features
    Of course, your alternative is to pull the PROMS and read them in a
    ERPOM reader.

    Reading manuals is a good idea. I have many S-100 manuals available in
    my copy service; contact me via my Web site for details. (Don't use the
    gMail address in this msg to reply.). There are also manuals available
    online for download, and other vendors offer copies too. A Web search
    will find them.

    Good luck with your "new" computer!

    Herb Johnson


    Herbert R. Johnson, phone 609-771-1503, New Jersey USA
    web site
    domain mirror
    my email address: hjohnson AAT retrotechnology DOTT com
    if no reply, try in a few days: herbjohnson ATT comcast DOTT net
    "Herb's Stuff": old Mac, SGI, 8-inch floppy drives
    S-100 IMSAI Altair computers, docs, by "Dr. S-100"


  12. Re: IMSAI CP/M BIOS research

    Hi
    As others have mentioned, many setups only have
    enough code in ROM to read the first track. Some
    even take this a step farther in that the code on the
    first track scans the disk for a specific file name
    and then reads that into memory as the BIOS routines.
    This two stage setup has the advantage that one can
    modifiy the BIOS as a file rather than having to do
    a complete sysgen to create a new boot image.
    As for the serial, many CP/M's will use I/O address 0
    as the serial's data port and address 1 as the status
    port. This is typical for the hardwired com chips used
    on early serial boards. You might write some simple
    code to check this out and have it send the read serial
    data to the output display on the front of the IMSAI ( I/O
    address 0FFH ). Do note that "LED on" means "0" written
    to this bit. You can experiment with Baud rates as
    well until the values displayed make sense.
    You can then feed some serial data into the various
    serial ports and see which one creates values on the
    I/O display.
    For the first pass, you can ignore the status address
    and just read the serial port. Once you find the wires
    that feed this port, you can determine the status bits
    in a similar method.
    Dwight


  13. Re: IMSAI CP/M BIOS research

    Allisonnospam@nouce.bellatlantic.net writes:


    >>What would normally fit in sector 1 of track 0?


    >One common approach is code that calls the base boot code to read in
    >additional sectors. Another was the first boot load sector 1 track 0
    >and that code is the disk boot loader that reads the required
    >additional sectors.


    The most difficult Board was the "Jade DD". It had an extra-Z80 on it.
    And the neccessary Files (EPROM on the Board, BoootSektor on Track 0,
    CP/M-Bios) were in 3 different variants of Z80-Mnemonics IIRC.

    I don't have that one anymore, so I can't look it up.

    Yours, Holger

  14. Re: IMSAI CP/M BIOS research

    On Thu, 20 Apr 2006 19:50:55 +0000 (UTC), Holger Petersen
    wrote:

    >Allisonnospam@nouce.bellatlantic.net writes:
    >
    >
    >>>What would normally fit in sector 1 of track 0?

    >
    >>One common approach is code that calls the base boot code to read in
    >>additional sectors. Another was the first boot load sector 1 track 0
    >>and that code is the disk boot loader that reads the required
    >>additional sectors.

    >
    >The most difficult Board was the "Jade DD". It had an extra-Z80 on it.
    >And the neccessary Files (EPROM on the Board, BoootSektor on Track 0,
    >CP/M-Bios) were in 3 different variants of Z80-Mnemonics IIRC.
    >
    >I don't have that one anymore, so I can't look it up.
    >
    >Yours, Holger


    The jade DD (I have one) was difficult because the board was dumb as a
    stump at boot. The Z80 had to be softloaded with boot code before it
    could do anything as there was no local Eprom/rom.

    Allison




  15. Re: IMSAI CP/M BIOS research

    First, I'd like to thank everybody for all the helpful information!
    When I got involved with computers back in the 70s, I jumped from the
    earliest systems (paper tape) to TRS-80s and then to PCDOS without ever
    learning CP/M.

    Now that I understand the boot process better, I think I'll ignore the
    PROMs for now and focus on getting track 0 sector 0 and the subsequent
    boot code loaded off the some of the floppies and disassembled to the
    point that I can figure out where to configure a serial board.

    Once I get the real BIOS disassembled, it should be fairly easy to
    figure out which board to use - I have the manuals for both. The TUART
    has (as someone pointed out) a complex configuration & status process.
    The Jade 1p2s is very simple by comparison.

    With respect to the Tarbell disk controller, I have a manual for it
    that covers up to revision "C". My board is a revision "D" so I'm
    missing some information. This board also needs to be configured for
    the exact drive that it is to be used with; I have Qume 842s which
    don't appear in the Tarbell manual and may not be compatible. I may
    need to do some more hardware hunting.


  16. Re: IMSAI CP/M BIOS research

    On 21 Apr 2006 04:02:32 -0700, "nonmaskable"
    wrote:

    >First, I'd like to thank everybody for all the helpful information!
    >When I got involved with computers back in the 70s, I jumped from the
    >earliest systems (paper tape) to TRS-80s and then to PCDOS without ever
    >learning CP/M.


    IMSAI predates trs80 by atleast two years.

    >Now that I understand the boot process better, I think I'll ignore the
    >PROMs for now and focus on getting track 0 sector 0 and the subsequent
    >boot code loaded off the some of the floppies and disassembled to the
    >point that I can figure out where to configure a serial board.


    The convention for most (there are exceptions) floppies is the first
    sector of a track is numbers as "1" so a 26 sector track is 1-26 not
    0-25.

    >Once I get the real BIOS disassembled, it should be fairly easy to
    >figure out which board to use - I have the manuals for both. The TUART
    >has (as someone pointed out) a complex configuration & status process.
    >The Jade 1p2s is very simple by comparison.



    >With respect to the Tarbell disk controller, I have a manual for it
    >that covers up to revision "C". My board is a revision "D" so I'm
    >missing some information. This board also needs to be configured for
    >the exact drive that it is to be used with; I have Qume 842s which
    >don't appear in the Tarbell manual and may not be compatible. I may
    >need to do some more hardware hunting.


    Tarbell was a good board. Likely the rev level is minor changes.
    The disks, Qume may well be compatable but wasn't available at the
    time the manual was written . That was common as early on the number
    of drives were few and new vendors were just entering the market.
    Also the tarbell may have been supplied with other vendors drives as
    part of a packaged system and only documented for that configuration.
    As memory serves they came with Shugarts (as packaged systems.).

    Allison



  17. Re: IMSAI CP/M BIOS research

    In article <1145617352.201588.303670@j33g2000cwa.googlegroups. com>,
    "nonmaskable" writes:
    > First, I'd like to thank everybody for all the helpful information!
    > When I got involved with computers back in the 70s, I jumped from the
    > earliest systems (paper tape) to TRS-80s and then to PCDOS without ever
    > learning CP/M.
    >
    > Now that I understand the boot process better, I think I'll ignore the
    > PROMs for now and focus on getting track 0 sector 0 and the subsequent
    > boot code loaded off the some of the floppies and disassembled to the
    > point that I can figure out where to configure a serial board.
    >
    > Once I get the real BIOS disassembled, it should be fairly easy to
    > figure out which board to use - I have the manuals for both. The TUART
    > has (as someone pointed out) a complex configuration & status process.
    > The Jade 1p2s is very simple by comparison.
    >
    > With respect to the Tarbell disk controller, I have a manual for it
    > that covers up to revision "C". My board is a revision "D" so I'm
    > missing some information. This board also needs to be configured for
    > the exact drive that it is to be used with; I have Qume 842s which
    > don't appear in the Tarbell manual and may not be compatible. I may
    > need to do some more hardware hunting.


    Qume 842
    1.6 meg - double sided - quad density - used by Tandy in the Model 16/6000.
    Not likely to have been included in the BIOS of a machine of IMSAI vintage.
    I would imagine the controller may not be able to deal with it. Probably
    easier to look for an old Shugart to at least get things started.

    bill


    --
    Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
    bill@cs.scranton.edu | and a sheep voting on what's for dinner.
    University of Scranton |
    Scranton, Pennsylvania | #include

  18. Re: IMSAI CP/M BIOS research

    The differences in the Tarbell board from rev. C to D are irrelevant as
    to the software. The circuit may be slightly different, but the
    software will be the same.

    The Tarbell boards were not much more than wire-wrap boards in some
    regards, because drives were not well standardized. But most of them
    are configured for a shugart drive (800, 801, 850, 851) (the 850/851 are
    double sided). Your qume drives are shugart compatible, I believe, so
    they will probably work with the boards as they are probably configured.
    But try to verify that the configuration is shugart compatible. The
    qume drives you have are MUCH newer than the board and you may not find
    jumpering info for them specifically.

    [Tarbell also made a later double density disk controller board.]


    nonmaskable wrote:
    > First, I'd like to thank everybody for all the helpful information!
    > When I got involved with computers back in the 70s, I jumped from the
    > earliest systems (paper tape) to TRS-80s and then to PCDOS without ever
    > learning CP/M.
    >
    > Now that I understand the boot process better, I think I'll ignore the
    > PROMs for now and focus on getting track 0 sector 0 and the subsequent
    > boot code loaded off the some of the floppies and disassembled to the
    > point that I can figure out where to configure a serial board.
    >
    > Once I get the real BIOS disassembled, it should be fairly easy to
    > figure out which board to use - I have the manuals for both. The TUART
    > has (as someone pointed out) a complex configuration & status process.
    > The Jade 1p2s is very simple by comparison.
    >
    > With respect to the Tarbell disk controller, I have a manual for it
    > that covers up to revision "C". My board is a revision "D" so I'm
    > missing some information. This board also needs to be configured for
    > the exact drive that it is to be used with; I have Qume 842s which
    > don't appear in the Tarbell manual and may not be compatible. I may
    > need to do some more hardware hunting.
    >


  19. Re: IMSAI CP/M BIOS research

    The qume drive is almost certainly compatible with a Shugart 851. It
    may support buffered "fast seek", but even if it does, it won't require
    it and will work with Shugart 851 jumpering.


    Bill Gunshannon wrote:

    > In article <1145617352.201588.303670@j33g2000cwa.googlegroups. com>,
    > "nonmaskable" writes:
    >
    >>First, I'd like to thank everybody for all the helpful information!
    >>When I got involved with computers back in the 70s, I jumped from the
    >>earliest systems (paper tape) to TRS-80s and then to PCDOS without ever
    >>learning CP/M.
    >>
    >>Now that I understand the boot process better, I think I'll ignore the
    >>PROMs for now and focus on getting track 0 sector 0 and the subsequent
    >>boot code loaded off the some of the floppies and disassembled to the
    >>point that I can figure out where to configure a serial board.
    >>
    >>Once I get the real BIOS disassembled, it should be fairly easy to
    >>figure out which board to use - I have the manuals for both. The TUART
    >>has (as someone pointed out) a complex configuration & status process.
    >>The Jade 1p2s is very simple by comparison.
    >>
    >>With respect to the Tarbell disk controller, I have a manual for it
    >>that covers up to revision "C". My board is a revision "D" so I'm
    >>missing some information. This board also needs to be configured for
    >>the exact drive that it is to be used with; I have Qume 842s which
    >>don't appear in the Tarbell manual and may not be compatible. I may
    >>need to do some more hardware hunting.

    >
    >
    > Qume 842
    > 1.6 meg - double sided - quad density - used by Tandy in the Model 16/6000.
    > Not likely to have been included in the BIOS of a machine of IMSAI vintage.
    > I would imagine the controller may not be able to deal with it. Probably
    > easier to look for an old Shugart to at least get things started.
    >
    > bill
    >
    >


  20. Re: IMSAI CP/M BIOS research

    nonmaskable wrote:

    > Now that I understand the boot process better, I think I'll ignore the
    > PROMs for now and focus on getting track 0 sector 0 and the subsequent
    > boot code loaded off the some of the floppies and disassembled to the
    > point that I can figure out where to configure a serial board.


    > With respect to the Tarbell disk controller, I have a manual for it
    > that covers up to revision "C". My board is a revision "D" so I'm
    > missing some information. This board also needs to be configured for
    > the exact drive that it is to be used with; I have Qume 842s which
    > don't appear in the Tarbell manual and may not be compatible. I may
    > need to do some more hardware hunting.


    I checked my Tarbell manuals. I have a no-revision manual from 1978,
    and a revision C or D from Oct-Nov 1979. I doubt your card changed much
    from C to D; there is not much change from no-rev to C. Check the
    schematic vs your board to see if there are new chips or not.

    Your plan to look at tracks 0 and 1 to check the BIOS is reasonable.
    You'll be able to see the start of CP/M BIOS by looking for a series
    of JMP instructions, corresponding to the BIOS entry points, at the
    beginning of some sector. It's possible they will be CALL instructions
    but they should be JMP's.

    Assuming you have some kind of computer which can read your 8-inch
    single density disks, of course you can look at any files to see if
    there is BIOS source code. (If your disks are double density, the
    Tarbell SINGLE density controller could not have written or read them.)

    The Qume 842 is an 8-inch double sided drive. It can operate in single
    density mode (which is all the Tarbell can do) or double density; and
    of course it can run either single or double sided. Again, the Tarbell
    manual describes the board; it was designed for a user to set up any
    drive for which you have information, not just the drives listed in the
    manual. The Qume manual informs you about the drive. You have
    sufficient information to connect the two for single density operation.

    Herb Johnson


    Herbert R. Johnson, voice 609-771-1503, New Jersey USA
    web site
    domain mirror
    my email address: hjohnson AAT retrotechnology DOTT com
    if no reply, wait & try: herbjohnson ATT comcast DOTT net
    "Herb's Stuff": old Mac, SGI, 8-inch floppy drives
    S-100 IMSAI Altair computers, docs, by "Dr. S-100"


+ Reply to Thread