Difficulty writing to cylnders >=64

[andshort]

Hi David,

I am wondering if you have any input on the following:

With two different drives and multiple DD and HD floppy drives, I get write errors from cylinder 64 and up. The symptoms are very consistent. When writing mac 800kb disks, when reaching cylinder 64 head 0, I get write errors. After many retries it might continue writing. This happens only on head 0, (head 1 is always fine - tried with --heads=1).

I understand this is probably not an issue with fluxengine, but in any case, do you know if there is something special with sector 64, making it and subsequent sectors difficult to write on? Could it be that the drives are trying to do something "smart"?

The issue might have been there when using previous versions of fluxengine, but are only now noticeable since you added the verification algorithm.

I would appreciate your thoughts,

Andrew

[davidgiven]

Track 64 is one of the places where Mac drives increase the motor speed (because the tracks there are towards the centre of the disk and move past the head more slowly). FluxEngine does this by increasing the clock speed. This usually works but it pushes some PC drives towards the edge of the envelope. I've got one drive that doesn't work with Mac disks.

There's a tool for testing this; look at http://cowlark.com/fluxengine/doc/driveresponse.html. Try running that and see what the graphs look like. I'd suggest increasing the interval to .5us to make the test run faster, and you might want to try cylinders 0 and 79 and see how they differ.

[andshort]

Thanks, that makes sense. Any reason why it is happening only on head 0 and not on 1?

I would guess that the pulse density is the same among sides.

I will try the driverresponse, however I know it won't look good...

[davidgiven]

If it were just one drive, I'd suggest it was a head problem where one head is reacting better to marginal situations. Two drives seems too coincidental, though. It'd be interesting to figure this out.

[hpingel]

Hi, Just stumbled over this issue and have to say that I have a similar issue. Trying to write single sided commodore1541 it stops at track 63 and after my latest recompile with latest master it stops at track 70. What I mean is that at track 70.0 I have Verify errors (Error: Write failed; uncorrectable errors during write).

[davidgiven]

Sorry for the delay; been working on other projects... can you try the driveresponse tool too? It's possible that something about the pulse density is making the drive unhappy. Also, are you using DD or HD disks, and are you setting the high_density flag?