Image Map Image Map
Results 1 to 10 of 10

Thread: Toshiba T5200/100 with CP30174E

  1. #1
    Join Date
    Feb 2012
    Location
    New Zealand
    Posts
    2,572
    Blog Entries
    2

    Default Toshiba T5200/100 with CP30174E

    As part of a recent haul I got a T5200/100 with a seized hard drive. I got it freed up and it runs great except the drive fails to write 90% of the time.
    I looked around and didn't have a spare CP3104E laying about and didn't want to downgrade to my spare CP344, which turned out to have quite a few bad sectors anyway.

    I did have a spare CP30174E though, formatted it as 930/8/46 and checked it's operation on a spare 486. Rather than going the XTIDE or SCSI routes, I thought I'd try the old ROM table method which as far as I'm aware has been successful for others.

    My ROM is the exact same verison as what's on minuszerogdegrees for the T5200C - a Phoenix based BIOS. I did a byte by byte comparison and it's identical.
    I changed entry #6 To: 00 87 03 08 00 00 FF FF 00 00 00 00 00 87 03 2E

    Which is 903 cyl, 8 heads, 46 SPT. I ran Toshiba's HDDID tool as a sanity check, and it gave me the same figures 0387 0008 and 2E.
    I updated the checksum byte, and wrote the ROM to a 27C1024. Installed in the machine, and it booted fine, and my Type 6 was now 162MB! 903 / 8 / 46 - wee!

    But with the CP30174E installed and the machine set to it's new Type 6, it complains about Fixed Disk Failure and leaves the HDD light illuminated. It automatically changes the CMOS entry to "No Drive" in this process.
    I then install a CP344, and the CMOS complains about a hard drive error [despite being set to no drive] - turns out it's automatically changed it to "Type 6" - where the old CP344 setting once was...
    The original CP3104E - no worries, auto detects as type 9 and boots :/

    What kind of voodoo is this?
    Those of you who successfully did this modification - which drive type did you override? Which BIOS version was used?

    I'm wondering if overriding Type 9 or using the later T5200/200 BIOS [which I can't find] may solve it.
    Last edited by SpidersWeb; March 17th, 2016 at 04:20 AM.
    WTB: IBM RT Parts and Accessories, and AOS.
    Twitter / Discord / YouTube / Twitch

  2. #2

    Default

    Not that much of a BIOS fumbler here, but is it possible that you only changed the bytes of the values "displayed" and not the values "used"?

    I am just guessing here, didn't look at the BIOS, but that was the first thing that came to my mind. Especially since your CP344 works with the "hacked type 6" makes me think it's something like that. Like you changed the values of some string that the BIOS displays to the user, but for configuring the actual drive it uses other values.

  3. #3
    Join Date
    Feb 2012
    Location
    New Zealand
    Posts
    2,572
    Blog Entries
    2

    Default

    I had a similar thought and did a search for hex 03 25 and 25 03 (hex 325 being 805 cyl - which is one of the standard types)- but couldn't find a second hard drive table.
    Also the CP-344 is generating an error after the mod. It detects as Type 6, but causes a Fixed Disk Failure until I manually force it to None. Previously it detected as 6 and booted.

    What the table looks like with my mod on type 6:

    T5200_drive_table.jpg

    eck, it resized it, but the table I'm editing starts at 1E400.
    WTB: IBM RT Parts and Accessories, and AOS.
    Twitter / Discord / YouTube / Twitch

  4. #4
    Join Date
    Feb 2012
    Location
    New Zealand
    Posts
    2,572
    Blog Entries
    2

    Default

    For anyone who wants to take a look, and maybe even take a crack at some more suggestions, I put together a ZIP with the original, my modifications, and a text file to make sure nobody who finds it in the future gets full of false hope.

    Download ZIP


    Searching groups.google shows a few people who have done this - but small details are lacking, such as the exact ROM used and the exact drive models used.
    I've also found a few laptop companies were offering it as a service in the early 90's. I haven't found a post where someone has said "editing the drive paramter table doesn't work" but I haven't found any tricks or hints.

    I also can't find other ROM revisions, such as the one used in the T5200/40 or T5200/200 - which may have different detection code.

    Tonight I'm going to replace all 3 usuable drive types, and write the ROM again, to see if maybe anything changes [probably not].
    WTB: IBM RT Parts and Accessories, and AOS.
    Twitter / Discord / YouTube / Twitch

  5. #5
    Join Date
    May 2006
    Location
    Melbourne, Australia
    Posts
    5,936

    Default

    Have you tried the '2 - HARD DISK FORMAT' option of the TESTCE52.EXE (diagnostics and test) utility ?
    Maybe it puts a marker on the drive, a marker that the BIOS expects to see. Just a guess.

  6. #6
    Join Date
    Feb 2012
    Location
    New Zealand
    Posts
    2,572
    Blog Entries
    2

    Default

    Damn, I just put the ROM in the eraser!
    I will try that next, didn't know it existed. Thanks!

    I did notice something else just before I came back in, and that's when the drive type is set to 6 or 7 it just errors, but when set to Type 9 I can hear the drive seek a little first. No idea if it's relevant, just something I noticed.
    I'll make sure to use Type 9 this time, and try the format routine.
    WTB: IBM RT Parts and Accessories, and AOS.
    Twitter / Discord / YouTube / Twitch

  7. #7
    Join Date
    Feb 2012
    Location
    New Zealand
    Posts
    2,572
    Blog Entries
    2

    Default

    Unfortunately that's a no go.

    Changing to type 9 didn't make a difference.
    System removes drive type on boot, and leaves the HDD light on to show it's displeasure. Then when TESTCE5'2 format program runs it sees a blank entry, it then tries anyway and the hard drive returns BAD COMMAND ERROR.

    So I can't do the TESTCE52 format.
    If I leave it to No Drive and boot form floppy - the system is fine though - just no hard drive access. So at least an XTIDE ROM is viable, just wish I knew what was really going on with the native ROM.
    WTB: IBM RT Parts and Accessories, and AOS.
    Twitter / Discord / YouTube / Twitch

  8. #8
    Join Date
    Feb 2012
    Location
    New Zealand
    Posts
    2,572
    Blog Entries
    2

    Default

    One more thing I tried was using minus one cylinder, 902, no dice there either.
    Machine is reassembled with an XTIDE 1.15 ROM, but it still has the CP30174E in it, and to get to the BIOS is only 6 screws and four clips, and I have two of these.
    So if anyone feels creative in the future, please feel free to ask me to try something, it bothers me that I couldn't get this to work [but I didn't want *another* machine waiting to be finished on the bench]
    WTB: IBM RT Parts and Accessories, and AOS.
    Twitter / Discord / YouTube / Twitch

  9. #9

    Default

    Quote Originally Posted by SpidersWeb View Post
    I'm wondering if overriding Type 9 or using the later T5200/200 BIOS [which I can't find] may solve it.
    I've attached a copy of the 1989 Award bios for the T5200-100 together with a patched version with support for a 528MB drive which is the maximum possible on this BIOS.
    If you put any drive larger than 528MB into the machine it will usually work as a 528MB drive and if you compare these two versions, you should be able to work out how to customise it yourself.

    I have seen comments that some BIOS versions only work with Conner Peripherals drives, but this BIOS will work with any make of drive.
    However, note that the IDC header on the drive end of the 40 way IDE cable may be fitted backwards (it was on all 3 of my T5200s) so you can't plug it into another drive with a shroud around the 40 Way connector unless you file off the keyway bump on the connector or use a different cable. This didn't matter with the Conner drives as their IDE connector was unshrouded.

    BTW I found a simple way of securely mounting a compact flash card to IDE adapter so that the card could be swapped by just removing the keyboard instead of fully dismantling the machine. Let me know if you want further details on how to do this.
    Attached Files Attached Files

  10. #10
    Join Date
    Feb 2012
    Location
    New Zealand
    Posts
    2,572
    Blog Entries
    2

    Default

    Brilliant! Thanks for that, and hopefully it'll help others in the future too.
    I'd seen similar things for the T3200SX, but not the T5200.
    WTB: IBM RT Parts and Accessories, and AOS.
    Twitter / Discord / YouTube / Twitch

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
  •