Bad ROM FBLite

[WebClaw]

Drive received, likely with a bad head.

When connected over USB, it presents in a BSY state. It does not click, but it makes the typical sound of heads failing to read. I suspect that if I performed a head map change in RAM, I could likely pull a partial image.

PCB was removed and the ROM was extracted using normal soldering temperatures.

The ROM dump is the problem. The contents look completely irregular, and PC-3000 cannot interpret it. I reread it using a different programmer and received the same CRC.

I then reattached the ROM to the original USB PCB. The drive is now dead.

I can attempt a head swap and try to rebuild the ROM from SA - 4F may be a pain, but I’m curious if anyone has seen this before or has an opinion on what may have happened.

The only oddity I noticed is that the ROM chip is not the usual brand I commonly see on these drives, such as Winbond. I do not recall the exact part number, but I can get a screenshot if useful.

ROM attached.

Attachments

rom.zip

(58.64 KiB) Downloaded 12 times

[fzabkar]

The second half is empty.

Code:

Analysing WO46657-2.bin ...

Searching for LDSCs and verifying PCMBlocks ...

LDSC   LDSC    Att   PCMBlock          RAM         size      PCMBlk CS
Start  ID CS        Start -  End     address     RAM / ROM    Exp/Act
-----  -- --   --   -----   -----   --------   ------ -----  ---------
    0  5A 63   02     100 -  2437      12800     2336  2336  6D0C 6D0C   OK
   20  01 4D   0C    2438 -  4CA4   88800000     286C  286C    CF   CF   OK
   40  02 65   01    4CA5 -  D051          0 c   ABB4  83AC    98   98   OK
   60  03 C6   01    D052 -  D6D6      117AC c    80C   684    87   87   OK
   80  04 22   03    D6D7 -  D8CB      80220 c    2A4   1F4    21   21   OK
   A0  05 E0   03    D8CC -  DABC   10157200 c    304   1F0    2C   2C   OK
   C0  06 AA   01    DABD - 24EB5   90347E00 c  1ED50 173F8    FF   4E   BAD
   E0  07 EF   01   24EB6 - 3DCD2   90006E30 c  23370 18E1C    FF   E4   BAD

LDSC   = PM Loader Config String (32 bytes)
ID     = ID byte of LDSC (byte #0)
CS     = Checksum byte or word
Att    = Attributes
PCMBlk = Program Code Memory Block
Exp    = Expected checksum for PCMBLock
Act    = Actual checksum for PCMBLock
c      = compressed PCMBlock
size   = size of decompressed (in RAM) and compressed (in ROM) PCMBlock in bytes

No ROYL directory modules (0x0B or 0x20B) found in ROM


[WebClaw]

Thank you; - how in the heck does that just "happen".
Sigh...

[pepe]

Hello,

i would suspect intentional abuse, but who knows... If it is a poor quality flash... However, it is odd to have not only bit errors but complete erasure of blocks.
I wrote some prog to help recovering 4f quite some yrs ago, so i can do that but need physical access to the drive.
pepe

[WebClaw]

Thanks Pepe; was able to rebuild 4F with a donor MHA - lucky donor... However drive was completely messed up.

02 contained (what I think was purposeful FF) in critical area's there there was jibberish in 109. I have no idea what this case was about. It's one of those cases that even if you knew the history you still would not care.

Your gut was right.

[Hardcore Games]

I figure this disk is far past the best before date

It does suggest some logic problems