Yes, I did mention this before. This is a Beta test version of dtc with changes to make it play better with hard-sector diskettes. It is very possible they have extended the file format in some manner. Please contact the KF folks if you need input on that - they are very interested in getting the hard-sector capture / emulate flow working.
You did not respond to my question about logic analyzer trace. If it will help, I'll setup and capture it. Please let me know?
not sure that the logic analyser will help a lot.
The beta version of dtc interest me since this is the only way to check that the index event are correctly aligned with the data stream.
i also need to add the ns sector decoding support in the hxc software to check the data integrity of these images (and generate raw files latter). (and the encoding right after).
Btw is there any file format used to store the images ?
I dropped a note to Istvain at Kryoflux and asked him about updates made to the RAW format. Also requested that he contact you with any hints or advice.
I'm almost positive that the new dtc program is intended to preserve relationships between index and data, but we'll have to wait for Istvain's response to be sure.
Heath diskettes are generally imaged into the ".H8D" format. Northstar uses a format called ".NSI". I believe these are both sector data formats, but may have some additional metadata. I will see what I can find out about them and get you that information.
I don't know if this will help, but I found a section of assembly code written for the "SVD" emulator that supports Heath format for read and write. There are comments in the header that describe low-level organization. I'll forward it to you in e-mail.
Neither one of the emulations can be read, but I think we've moved a bit closer. The Northstar now reports a disk error at sector 20 (relative to start of disk, not track). The HFE file from yesterday threw errors at sector 0. Interestingly, the error code is listed in their documentation as "NO INDEX PULSE". That is what you see when trying to use a soft-sector diskette (single index hole). So, there still appears to be an issue with index timing (or, possibly, duration).
but i think that there is some index jitter problems. i have to check if this jitter is introduced by the loader or already present in the raw streams.
which drive have you used to dump the disk ? the SA 400 or another one ?
Still not working, but the error is now "Sync byte not found". The error appears at Sector 0, whereas yesterday it appeared to get past that point (if I'm understanding the diagnostic message properly).
I dumped the KF RAW file from an ALPS DFC222B05A 1/2 height mechanism. This is a direct drive unit with solenoid head-load and I believe it's in good working condition.
Would it help if I start with a fully demagnetized diskette and capture again after formatting that in the Northstar?
Still not working, but the error is now "Sync byte not found". The error appears at Sector 0, whereas yesterday it appeared to get past that point (if I'm understanding the diagnostic message properly).
I dumped the KF RAW file from an ALPS DFC222B05A 1/2 height mechanism. This is a direct drive unit with solenoid head-load and I believe it's in good working condition.
Would it help if I start with a fully demagnetized diskette and capture again after formatting that in the Northstar?
Can you instead measure the index jitter timing with an oscilloscope ? You should have 2 different inter-index timing values : ~10ms and ~20ms (falling edged synced). What interest me is the variation around these value. The index duration interest me too. You can measure it while dumping the floppy disk with the KF.
And if possible do the same thing with the SA 400.
The diagram appears to claim 5 msec. duration for inter-sector pulses and 1 msec. for the rotational pulse. However, since the holes on all hard-sector media I've seen are the same diameter I am skeptical of those numbers. Later today or tomorrow I will measure index pulses from a drive and settle the issue.