From: Marko Mäkelä (marko.makela_at_hut.fi)
Date: 2004-12-05 13:45:38
On Sun, Dec 05, 2004 at 02:11:19PM +0200, Marko Mäkelä wrote:
> Do you think that it could be the FDC ROM? I'll next adapt your
> "read6530.lst" to write the data to screen memory and dump it from there.
Done. The FDC ROM doesn't correspond to any of the ones archived at FUNET.
This is the closest match:
od -Ax -t x1 901869-01.bin | diff - fdc-dump.txt
7c7
< 000060 f7 a4 28 20 22 fd a5 29 78 9d eb 04 ea ea ea ea
---
> 000060 f7 a4 28 20 22 fd a5 29 78 9d eb 04 a9 08 95 23
18c18
< 000110 21 bd e4 ff 05 40 85 40 a9 5c 9d eb 04 4c 9f fe
---
> 000110 21 bd e4 ff 05 40 85 40 a9 5c 9d eb 04 4c ea ff
53c53
< 000340 8a 48 a9 05 85 8f 85 8f e6 00 a5 00 29 0f d0 02
---
> 000340 8a 48 a9 03 85 8f 85 8f e6 00 a5 00 29 0f d0 02
63,64c63,64
< 0003e0 f3 fc 08 02 0c 03 20 10 aa aa aa aa aa aa aa aa
< 0003f0 aa aa aa aa aa aa aa aa aa aa aa a3 04 fc 3f ff
---
> 0003e0 f3 fc 08 02 0c 03 20 10 aa aa 95 23 4c 9f fe aa
> 0003f0 aa aa aa aa aa aa aa aa aa aa 53 a3 04 fc 3f ff
As you can see, at $ffea there are some patch bytes (sta $23,y:jmp $fe9f,
though I'm not sure of the opcode $95). At $fd1d, jmp $fe9f has been
changed to jmp $ffea. So, this looks like a patched version. Could
you or someone else analyze these differences?
Interestingly, the FDC ROM is no way close to "fdc-2.7b.bin" from
a 8296D. The diffs of the upper 1024 bytes are 130 lines, while
the diffs above are only 18 lines. The back plate of this CBM 220
says:
C= commodore
8296-D
FTZ-Prüf-Nr.: C-041/81
220/240 V~ 50 Hz 85 W
SER.NO.WG 2802
MADE IN W.-GERMANY
Is the FDC ROM responsible for controlling the drive LEDs? Does the
main processor control the green/red status LED?
I'll next disassemble the drives one more time to figure out what is going
on with the write protect sensors. It's a nuisance to debug this, because
the drives are normally mounted to the ceiling of the base unit. Luckily,
the monitor can be operated next to the computer; the cable is long enough.
Marko
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail pre-2.1.8.