Unable to successfully read anything on new v4.0 board

[PXAbstraction]

Hey all. I'm at my wit's end here and I'm hoping someone might have an answer for me. The fine folks at the VCF forums have been trying to help me out and there's a lot of information in the thread over there. But, I'll try to give a synopsis here.

I have a v4.0 Greaseweazel board I purchased on eBay. I've got a new old stock 3.5" drive and a Teac FS-55GFR 5.25" drive. The GW has the most current firmware. I'm on Windows 11 Pro. I'm using the FluxMyFluffyFloppy GUI for the GW tools, but have tried the straight command-line as well.

Initially, I tried ripping Atari ST disks (the main reason I bought it), but have also tried ripping Commodore 64 5.25" disks and regular DOS 3.5" disks. Nothing has worked. No matter what I try to rip, I get variants of this:

Unknown mark 42
T78.0: Ignoring unexpected sector C:79 H:0 R:10 N:2
T78.0: Ignoring unexpected sector C:79 H:0 R:17 N:2
T78.0: Ignoring unexpected sector C:79 H:0 R:1 N:2
T78.0: Ignoring unexpected sector C:79 H:0 R:18 N:2
T78.0: Ignoring unexpected sector C:79 H:0 R:13 N:2
T78.0: Ignoring unexpected sector C:79 H:0 R:14 N:2
T78.0: Ignoring unexpected sector C:79 H:0 R:15 N:2
T78.0: Ignoring unexpected sector C:79 H:0 R:16 N:2
T78.0: Ignoring unexpected sector C:79 H:0 R:11 N:2
T78.0: Ignoring unexpected sector C:79 H:0 R:12 N:2
T78.0: IBM MFM (0/18 sectors) from Raw Flux (1624582 flux in 3996.94ms) (Retry
T78.0: Giving up: 18 sectors missing

So far, I've tried the following:

• New USB cable
• Brand new floppy cable
• Running the drives in sequence and each independently
• Atari ST, DOS and Commodore 64 formatted disks
• Multiple USB ports on my main machine, but I've also tried my laptop

I'm hoping I don't have defective hardware here, but I am honestly at a loss of what to do now. I know this isn't just bad disks because I get the same constant errors, regardless of the disks or the drive.

Does anyone have any steps I could pursue to try to figure this out? Thank you in advance.

[keirf]

Firstly if you can, ditch the GUIs for now and test gw on the command line. You're bypassing a layer of abstraction this way and gw command line isn't too unfriendly.

Maybe concentrate on one type of disk for now, say PC 1.44M 3.5" disks. And concentrate on reading a couple of cylinders during testing, rather than a whole disk:
gw read --tracks=c=0-1 --format=ibm.1440 name.img

You may need --drive=b if your cable doesn't have a twist. name.img is just a name for the generated IMG file.

Your errors look like you are off by one cylinder in your reads. Note unexpected sector reports with cylinder (C) reported as 79 when Greaseweazle is trying to read cylinder 78 (for example). That's the first thing to sort out. I would say that's a drive calibration error unless you see the same on multiple 3.5 inch drives?

[PXAbstraction]

Hi there and thanks for replying! Sorry for my delays, holidays and whatnot. :)

The GUIs just generate and pass command lines anyway, but I get your point about removing abstraction so I did try that, but sadly got the same results, at least when using the same commands.

I updated my GW to the latest firmware and then ran the command you got using two different 1.44MB PC floppy disks. These are the results I got:

C:\COMSTUFF\Greaseweazle>gw read --tracks=c=0-1 --format=ibm.1440 "c:\comstuff\no save\test.img"
Reading c=0-1:h=0-1 revs=2
Format ibm.1440
T0.0: IBM MFM (0/18 sectors) from Raw Flux (157455 flux in 400.08ms)
T0.0: IBM MFM (0/18 sectors) from Raw Flux (393678 flux in 999.96ms) (Retry #1.1)
T0.0: IBM MFM (0/18 sectors) from Raw Flux (629841 flux in 1599.79ms) (Retry #1.2)
T0.0: IBM MFM (0/18 sectors) from Raw Flux (866003 flux in 2199.62ms) (Retry #1.3)
T0.0: Giving up: 18 sectors missing
T0.1: IBM MFM (2/18 sectors) from Raw Flux (183322 flux in 400.05ms)
T0.1: IBM MFM (2/18 sectors) from Raw Flux (458170 flux in 999.88ms) (Retry #1.1)
T0.1: IBM MFM (2/18 sectors) from Raw Flux (733027 flux in 1599.71ms) (Retry #1.2)
T0.1: IBM MFM (2/18 sectors) from Raw Flux (1007874 flux in 2199.52ms) (Retry #1.3)
T0.1: Giving up: 16 sectors missing
T1.0: IBM MFM (1/18 sectors) from Raw Flux (156256 flux in 400.04ms)
T1.0: IBM MFM (1/18 sectors) from Raw Flux (390534 flux in 999.84ms) (Retry #1.1)
T1.0: IBM MFM (1/18 sectors) from Raw Flux (624786 flux in 1599.65ms) (Retry #1.2)
T1.0: IBM MFM (1/18 sectors) from Raw Flux (859071 flux in 2199.46ms) (Retry #1.3)
T1.0: Giving up: 17 sectors missing
T1.1: IBM MFM (2/18 sectors) from Raw Flux (158831 flux in 400.03ms)
T1.1: IBM MFM (2/18 sectors) from Raw Flux (396912 flux in 999.84ms) (Retry #1.1)
T1.1: IBM MFM (2/18 sectors) from Raw Flux (635027 flux in 1599.65ms) (Retry #1.2)
T1.1: IBM MFM (2/18 sectors) from Raw Flux (873079 flux in 2199.44ms) (Retry #1.3)
T1.1: Giving up: 16 sectors missing
Cyl-> 0
H. S: 01
0. 0: XX
0. 1: XX
0. 2: XX
0. 3: X.
0. 4: XX
0. 5: XX
0. 6: XX
0. 7: XX
0. 8: XX
0. 9: XX
0.10: XX
0.11: XX
0.12: XX
0.13: XX
0.14: XX
0.15: XX
0.16: XX
0.17: XX

1. 0: XX
2. 1: XX
3. 2: XX
4. 3: ..
5. 4: ..
6. 5: XX
7. 6: XX
8. 7: XX
9. 8: XX
10. 9: XX
    1.10: XX
    1.11: XX
    1.12: XX
    1.13: XX
    1.14: XX
    1.15: XX
    1.16: XX
    1.17: XX
    Found 5 sectors of 72 (6%)

C:\COMSTUFF\Greaseweazle>gw read --tracks=c=0-1 --format=ibm.1440 "c:\comstuff\no save\test1.img"
Reading c=0-1:h=0-1 revs=2
Format ibm.1440
T0.0: IBM MFM (0/18 sectors) from Raw Flux (161651 flux in 399.95ms)
T0.0: IBM MFM (0/18 sectors) from Raw Flux (403623 flux in 999.74ms) (Retry #1.1)
T0.0: IBM MFM (0/18 sectors) from Raw Flux (645515 flux in 1599.51ms) (Retry #1.2)
T0.0: IBM MFM (0/18 sectors) from Raw Flux (887195 flux in 2199.30ms) (Retry #1.3)
T0.0: Giving up: 18 sectors missing
Unknown mark ff
T0.1: IBM MFM (0/18 sectors) from Raw Flux (182837 flux in 400.05ms)
Unknown mark ff
Unknown mark ff
Unknown mark ff
T0.1: IBM MFM (0/18 sectors) from Raw Flux (457021 flux in 999.83ms) (Retry #1.1)
Unknown mark ff
Unknown mark ff
Unknown mark ff
T0.1: IBM MFM (0/18 sectors) from Raw Flux (731319 flux in 1599.61ms) (Retry #1.2)
Unknown mark ff
Unknown mark ff
T0.1: IBM MFM (0/18 sectors) from Raw Flux (1005474 flux in 2199.42ms) (Retry #1.3)
T0.1: Giving up: 18 sectors missing
T1.0: IBM MFM (0/18 sectors) from Raw Flux (158421 flux in 400.01ms)
T1.0: IBM MFM (0/18 sectors) from Raw Flux (396066 flux in 999.79ms) (Retry #1.1)
T1.0: IBM MFM (0/18 sectors) from Raw Flux (633496 flux in 1599.59ms) (Retry #1.2)
T1.0: IBM MFM (0/18 sectors) from Raw Flux (870853 flux in 2199.38ms) (Retry #1.3)
T1.0: Giving up: 18 sectors missing
T1.1: IBM MFM (0/18 sectors) from Raw Flux (155479 flux in 400.02ms)
T1.1: IBM MFM (0/18 sectors) from Raw Flux (388590 flux in 999.81ms) (Retry #1.1)
T1.1: IBM MFM (0/18 sectors) from Raw Flux (621717 flux in 1599.59ms) (Retry #1.2)
T1.1: IBM MFM (0/18 sectors) from Raw Flux (854825 flux in 2199.37ms) (Retry #1.3)
T1.1: Giving up: 18 sectors missing
Cyl-> 0
H. S: 01
0. 0: XX
0. 1: XX
0. 2: XX
0. 3: XX
0. 4: XX
0. 5: XX
0. 6: XX
0. 7: XX
0. 8: XX
0. 9: XX
0.10: XX
0.11: XX
0.12: XX
0.13: XX
0.14: XX
0.15: XX
0.16: XX
0.17: XX

1. 0: XX
2. 1: XX
3. 2: XX
4. 3: XX
5. 4: XX
6. 5: XX
7. 6: XX
8. 7: XX
9. 8: XX
10. 9: XX
    1.10: XX
    1.11: XX
    1.12: XX
    1.13: XX
    1.14: XX
    1.15: XX
    1.16: XX
    1.17: XX
    Found 0 sectors of 72 (0%)

These are retail disks from a boxed copy of X-Wing that I have. The first disk got 6% of sectors apparently and the second for 0%.

For kicks, I reformatted the command and tried a 1.2MB disk from my Teac drive and got this:

C:\COMSTUFF\Greaseweazle>gw read --tracks=c=0-1 --drive=b --format=ibm.1200 "c:\comstuff\no save\test1.img"
Reading c=0-1:h=0-1 revs=2
Format ibm.1200
T0.0: IBM MFM (0/15 sectors) from Raw Flux (63249 flux in 334.33ms)
T0.0: IBM MFM (0/15 sectors) from Raw Flux (158067 flux in 835.57ms) (Retry #1.1)
T0.0: IBM MFM (0/15 sectors) from Raw Flux (252885 flux in 1336.80ms) (Retry #1.2)
T0.0: IBM MFM (0/15 sectors) from Raw Flux (347701 flux in 1838.02ms) (Retry #1.3)
T0.0: Giving up: 15 sectors missing
T0.1: IBM MFM (0/15 sectors) from Raw Flux (128164 flux in 334.31ms)
T0.1: IBM MFM (0/15 sectors) from Raw Flux (320099 flux in 835.53ms) (Retry #1.1)
T0.1: IBM MFM (0/15 sectors) from Raw Flux (511977 flux in 1336.74ms) (Retry #1.2)
T0.1: IBM MFM (0/15 sectors) from Raw Flux (703884 flux in 1837.95ms) (Retry #1.3)
T0.1: Giving up: 15 sectors missing
T1.0: IBM MFM (0/15 sectors) from Raw Flux (68400 flux in 334.30ms)
T1.0: IBM MFM (0/15 sectors) from Raw Flux (171020 flux in 835.51ms) (Retry #1.1)
T1.0: IBM MFM (0/15 sectors) from Raw Flux (273524 flux in 1336.71ms) (Retry #1.2)
T1.0: IBM MFM (0/15 sectors) from Raw Flux (376165 flux in 1837.91ms) (Retry #1.3)
T1.0: Giving up: 15 sectors missing
T1.1: IBM MFM (0/15 sectors) from Raw Flux (118669 flux in 334.30ms)
T1.1: IBM MFM (0/15 sectors) from Raw Flux (296338 flux in 835.50ms) (Retry #1.1)
T1.1: IBM MFM (0/15 sectors) from Raw Flux (473796 flux in 1336.70ms) (Retry #1.2)
T1.1: IBM MFM (0/15 sectors) from Raw Flux (651355 flux in 1837.90ms) (Retry #1.3)
T1.1: Giving up: 15 sectors missing
Cyl-> 0
H. S: 01
0. 0: XX
0. 1: XX
0. 2: XX
0. 3: XX
0. 4: XX
0. 5: XX
0. 6: XX
0. 7: XX
0. 8: XX
0. 9: XX
0.10: XX
0.11: XX
0.12: XX
0.13: XX
0.14: XX

1. 0: XX
2. 1: XX
3. 2: XX
4. 3: XX
5. 4: XX
6. 5: XX
7. 6: XX
8. 7: XX
9. 8: XX
10. 9: XX
    1.10: XX
    1.11: XX
    1.12: XX
    1.13: XX
    1.14: XX
    Found 0 sectors of 60 (0%)

Basically the same results. This 3.5" was brand new in box, but it's also a more "modern" drive (it advertised supporting Windows Vista on the box). I don't have another one, but I can see if I can track one down. I'm surprised that I'm getting the exact same result on a totally different drive though, that was sold to me as tested. Do you think that's still the root cause here or could it possibly be something else?

Thanks again. :)

[keirf]

The main issue here is that you should not specify c=0-79:h=0. The correct cylinder and head range will be selected for D64 image, which is 35 cylinders, single sided (ie. c=0-34:h=0). What you have specified is 80 cylinders and double stepping, hence Greaseweazle will attempt to get to cylinder 160, which of course doesn't exist! Many drives will silently reject seeks beyond their limit, but clearly not yours. What yours is doing is banging the drive head against the drive chassis. As you can imagine, this is not good for your drive's calibration!

You also don't need to specify --revs= in concert with a sector image format. Because Greaseweazle is doing track decode, it knows when it has a good or bad track, and will insert retries automatically. So --revs= is mainly useful for dumb raw reads, for example to raw flux files (such as SCP) with no --format specified.

Oh and you don't need --format= either, as it is implied by creating a D64 image file.

So your correct command line would be:

"C:\COMSTUFF\Greaseweazle\gw.exe" read --tracks=step=2 "C:\COMSTUFF\No Save\test.d64"

[PXAbstraction]

Hey again! With all your amazing advice and the help of the VCF forums, I think I have everything figured out. With the new (old) drive, I'm now able to successfully rip 3.5" disks and now that I know about the extra commands for double-stepping and the like, I think I'll be able to rip the 5.25" disks I want with the Teac drive as well.

I will say, the GW is incredibly powerful, but it definitely can be tricky for someone like me to start out who is knowledgeable, but not an expert in the nuances of how disk drives work. :) I run a small YouTube channel where I sometimes do retro content and I'm considering putting a couple of videos together about getting started with a GW using some of the more common retro formats. Maybe that would help give more lamen users a good place to begin.

Regardless, I thank you for your help and patience and I'm really excited to use this to get started on the archival projects I have planned. Cheers and thanks for making something so useful!

[wohali]

For a simpler process (esp. for beginners), you can also consider using something like the FluxEngine UI, which supports GW hardware.

[PXAbstraction]

I actually have tried that, but ran into many of the same problems I had here. Just needed more core knowledge in the end.

[keirf]

Yes Greaseweazle is painfully lacking in tutorial material. I recently had a new Getting Started wiki page contributed, which is a lot better than I had before. But more material would be amazing.