Image Map Image Map
Page 1 of 6 12345 ... LastLast
Results 1 to 10 of 57

Thread: Model 16A - HDC type 4 + DREM issue

  1. #1

    Default Model 16A - HDC type 4 + DREM issue

    Trying out a newly acquired type #4 hard disk controller in a Model 16A using a DREM as a MFM disk. The DREM has one of Pete's kindly hosted disk images as the boot image. I've tried 3 or 4 of them with the same result.

    Log is as below:


    Drive 0:
    CFG Name "Seagate ST4053 44.5M"
    _____________INTERFACE_____________
    HDD DS Bitmap 0xF ENABLED
    Mode ST506/ST412
    RWC Signal NO
    Ready Signal DRIVE SELECT
    Step Rate 64 us 7500/7500
    _______________DRIVE_______________
    Write Protected NO
    RPM/Bitrate 3600/5000 KBPS
    Encoding MFM
    RAW Track Size 10416
    _______________DREM________________
    L1/L2 Size 4/1024
    L1 Sync Event WRFIFO
    Snapshot DISABLED
    ______________FORMAT_______________
    Format WDC
    Encoding MFM
    TR# /Precomp. 1024/NO
    SD# /SEC# S:H/Size 5/17:0/512
    FirstID/Interleave 1/10
    Side/Track Skew 0/0
    GAP4a/SYNC/GAP4 0/14/432(AUTO)
    GAP1/GAP2/GAP3 37/0/32
    GAP Byte 0x4E

    ID CRC16/CCITT-FALSE
    order=16 polynom=0x1021
    init=0xFFFF xor=0x0
    direct=1 refin=0 refout=0

    DATA CRC16/CCITT-FALSE
    order=16 polynom=0x1021 direct=1
    init=0xFFFF xor=0x0
    direct=1 refin=0 refout=0

    0: MFM Read TR:0004 SD:00 L1
    0: MFM Read TR:0003 SD:00 L1
    0: MFM Read TR:0002 SD:00 L1
    0: MFM Read TR:0001 SD:00 L1-:0005 L2->L1
    0: MFM Read TR:0000 SD:00 L1-:0001 L2->L1


    At this point the M16A reports a boot error HC (CRC Error).

    The config file for this disk is shown below:


    [DSK]
    Name=Seagate ST4053 44.5M
    Encoding=MFM
    RPM=3600
    Tracks=1024
    Sides=5
    Sectors=17
    First Sector ID=1
    Sector Size=512
    Format=WDC
    Step Signal Hold=8
    Interleave=10
    Data ECC=NO


    Having had a number of email conversations with Pete, who unfortunately doesn't have this configuration, I'm at a bit of a dead end with knowledge, and direction.

    Does anyone have guidance as to why the drive would apparently successfully read and then consistently CRC at the same point in the boot process?

    Steve

  2. #2

    Default

    I believe the 16A needs an additional 16k or 64k (I can't remember) installed in order to boot from hard drive. The type 2 and 3 controller had this memory installed. The type 4 controller relied on the memory being installed on the model 16b/6000 motherboard.

    Don't quote me on this, but I think I'm pretty close.

  3. #3

    Default

    What do you get when running the HD diagnostic against the DREM? I haven't had much luck with a Type 4 or Type 2 and DREM myself. The HD diag (HDR15) always increments either ID NOT FOUND or CRC DATA FIELD errors. In my case there seems to be a pattern where if I run HDR15, do a continuous read + verify test and let it go for 200 passes the first 10-20 tracks generate a lot of CRC data field errors but eventually it slows down. Once this happens I can hit reset and boot from a TRSDOS-II DREM image, no luck with Xenix.

    I've tried the DREM with both a Type 2 HDA (+WD1010/WD2010 HDC) and Type 2. Output on the DREM console looks good (as does yours).

    Same results on both a 16A and 6000 but a normal MFM drive works fine using the exact same HDA/HDC.

    The DREM is such a nice solution I'd love to get things sorted out.




  4. #4
    Join Date
    Aug 2008
    Location
    Burlington, VT
    Posts
    180

    Default

    Quote Originally Posted by dlightman View Post
    The DREM is such a nice solution I'd love to get things sorted out.
    I suggest you contact Oleksandr at DREM. He's extremely responsive to issues with the DREM and rather tireless in pursuing solutions.

  5. #5

    Default

    Quote Originally Posted by shirsch View Post
    I suggest you contact Oleksandr at DREM. He's extremely responsive to issues with the DREM and rather tireless in pursuing solutions.
    See, that's the thing. There are others who have these same combinations working so I don't think it's the DREM.

  6. #6

    Default

    Quote Originally Posted by shirsch View Post
    I suggest you contact Oleksandr at DREM. He's extremely responsive to issues with the DREM and rather tireless in pursuing solutions.
    I'll update once I've run dlightman's suggestions in relation to testing the HDC, and i'll provide that and testing evidence to Oleksandr and report back.

    Thanks all for the advice and guidance.

  7. #7

    Default

    Quote Originally Posted by kb2syd View Post
    I believe the 16A needs an additional 16k or 64k (I can't remember) installed in order to boot from hard drive. The type 2 and 3 controller had this memory installed. The type 4 controller relied on the memory being installed on the model 16b/6000 motherboard.

    Don't quote me on this, but I think I'm pretty close.
    OK, I just checked some of my past exchanges with FDIV. I was wrong. To quote Frank:
    Code:
    f installing in a Model 12 (eg no 68000 cards present), change
    AP-AK to AG-AL, which places the RAM on this card at pages 8 & 9, where
    TRSDOS-II will use as the extra 16K bank used for OS overlays.  TRSDOS2.0
    and XENIX do not use Z80 RAM banks above 7 ever.

  8. #8

    Default

    Ah,

    So "TRSDOS2.0 and XENIX do not use Z80 RAM banks above 7 ever." so not needed?

  9. #9

    Default

    Quote Originally Posted by Tarkie101 View Post
    Ah,

    So "TRSDOS2.0 and XENIX do not use Z80 RAM banks above 7 ever." so not needed?
    Correct. It's not the "missing" 16k causing your problem. Xenix doesn't ever use it. It seems only TRSDOS-II does. (not 2, but II) Back to Frank
    In the TRSDOS-II family, you need 64+16K of RAM. The extra RAM could
    either be on the hard disk interface board, in the extra row of sockets
    on the Model 12/16B/6000 motherboard, or the 68000 memory card.
    (Or you could have a rare memory card that could support two rows
    of 64K of RAM which was meant for use in the Model II/16.)
    The additional 16K is used for OS overlays and buffers.

  10. #10

    Default

    Well, the Hard Disk Reliability Testing is going well.

    It's currently been doing the Random Read / Seek test for a while now with Zero errors.

    Very confused.

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
  •