Problem with PCMCIA when expansion boards added

[Hodgkinson]

Im wondering if anyone else has experienced problems with the PCMCIA port when a Blizzard 1230 IV board and 6860 board are added to a A1200 ? (Kickstart v40.68 , OS 3.0)

Without these boards coupled up, the PCMCIA port works fine with a 1MB SRAM card. However, when these boards are added, the card becomes unreadable, the card name becomes garbled, it doesn't detect the card instantly and the system reports errors when it reaches block 1024.

A alternative motherboard has been tried (With the ROMS from the original board) and the problem remains. Both boards were tested and the PCMCIA ports were found to be operating FINE before the expansion boards were added / after original board was removed and tested individually.

Also, booting from floppy disk with the original OS 3.0 diddn't work either.

Perhaps there is a memory allocation problem?

Can anyone shed light on the problem? Will the port still work correctly with other PCMCIA cards (More specificially a 3Com 3C589D perhaps?)

Thanks,
Hodgkinson.

[Wayne]

some A1200 turboboard are incompatible or cause incompatibilities issues with the A1200 pcmcia port
some works fine....for example all apollo turboboards or the blizzards 1240 and 1260

I think that your turboboard is on the black list

bye, Laser

[Animagic]

If it is a Blizzard 1230IV it works FINE. I had one among with Squirrel SCSI attached to the PCMCIA port and it worked just fine (with 3.0 roms)
Something else is wrong...

[Karlos]

I wonder if upadating to OS 3.1 might help here?

[Karlos]

Animagic wrote:
If it is a Blizzard 1230IV it works FINE. I had one among with Squirrel SCSI attached to the PCMCIA port and it worked just fine (with 3.0 roms)
Something else is wrong...

Ah, the Squirrel... A guy I used to know (alas I haven't seen him in ages - maybe he's a member here, hi Mike) had one of those and found it seriously slow on his 68060 card. It turns out the driver was using the MOVEP instruction, which was being emulated on 060, which being from   the Apollo stable didn't have the best handling code.
He actually patched the offending code with a reassembler, getting a serious performance kick :-)