Greaseweasel not reading 800k disks?

[eharmon]

I've been using my Greaseweazel to read 720k and 1.44mb IBM floppies and Amiga 880k floppies successfully, but 800k Mac floppies seem to be failing.

Obviously they're GCR which normally causes problems, but if I understand the docs correctly, the flux image is read at sufficient rate you can piece back together the data from the flux image on the computer side.

If I ask the Greaseweazel to verify the read I get the following:

$ gw read --format mac.800 /tmp/test.scp
Reading c=0-79:h=0-1 revs=1.3
Format mac.800
T0.0: Ignoring unexpected sector C:1 H:0 R:3 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:9 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:4 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:10 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:5 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:11 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:0 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:6 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:1 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:7 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:2 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:8 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:3 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:9 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:4 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:3 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:9 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:4 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:10 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:5 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:11 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:0 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:6 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:1 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:7 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:2 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:8 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:3 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:9 F:0x22
T0.0: Ignoring unexpected sector C:1 H:0 R:4 F:0x22
T0.0: Macintosh GCR (0/12 sectors) from Raw Flux (65377 flux in 259.10ms)
.....a bunch of retries....
T0.0: Giving up: 12 sectors missing

If I don't, it reads, but tools like HxC can't make any sense of it either.

Some posts indicate that PC drives sometimes damage the flux data stream for GCRs, but this feels like more of a rumor than something concrete. That said, I can't really find any information on this drive, an Alps DF354H914C, so...maybe?

Or am I just doing something wrong?

 

[eharmon]

Greaseweazel:

Understanding differences · keirf greaseweazle · Discussion #438

Hello all, I have an arcade system that is using a FDD to load the games. The original FDD is an old Mitsumi D357. When I am reading the disk with this drive I am able to read the whole disk (good ...

github.com

keirf on May 26, 2024
Maintainer
Yes it's common with GCR. Some PC drives are more tolerant than others.

HxC:

Understanding the differences - HxC Floppy Drive Emulator

by Jeff » Sun May 26, 2024 12:06 am

i will implement a compensation curve setting to deal with such cases. there is similar cases with people using pc drive to read macintosh gcr disks.

This drive is from maybe 2000-2003. I just tried a 1993/1994 drive and it reads properly.

Curiously, on the 2000-2003 drive I can consistently read the same 24 sectors, but that's it. So perhaps my drive is indeed intolerant, and the existing tools are unable to compensate. Unless anyone else has some tricks...if you stumble upon this thread, now you know what your problem is .

 

[olePigeon]

KyroFlux forums covers this as well. You'll want a specific drive of a specific era for best compatibility. The drives just weren't made for it, so some handle it better than others.

It's why I got an AppleSauce so I don't have to fiddle with that stuff. Only drawback is I can't write back copy protected disks without secret magic disk cracking knowledge that no one will ever help you with because everyone is super protective of their OMG-STEVE-JOBS-RARE software and assume you're a pirate looking to ruin the resale market.

 

[eharmon]

KyroFlux forums covers this as well. You'll want a specific drive of a specific era for best compatibility. The drives just weren't made for it, so some handle it better than others.

It's why I got an AppleSauce so I don't have to fiddle with that stuff. Only drawback is I can't write back copy protected disks without secret magic disk cracking knowledge that no one will ever help you with because everyone is super protective of their OMG-STEVE-JOBS-RARE software and assume you're a pirate looking to ruin the resale market.

It seems the problem goes back to 1989 even, see this documentation from Central Point for their Option Board: http://ftp.oldskool.org/pub/OptionB...kage Contents/Documentation/OB_Drive_Note.pdf

My working drive lives in my period correct Pentium system, and the rest of my drives are 2000s or Mac drives so...guess I'll have to go eBay hunting for a period TEAC or Sony.

 

[olePigeon]

I looked inside my Applied Engineering floppy drive once because I was curious how they got an off-the-shelf floppy drive to read and write variable rate Mac disks. It looked like the Borg invaded and attached 30 millions wires, so I closed the case up and didn't open it again.

 

[eharmon]

I found a Sony MPF920-1 with a June 2001 date in one of my spare PC cases. It seems popular for Greaseweazel and I can confirm that mine works wonderfully for Mac disks.

So it seems even fairly late MPF920s work nicely.

 

[Forrest]

FYI Adafruit sells upcycled, Sony MFP920 drives that have been tested for $19.95 https://www.adafruit.com/product/6037