Recent content by per

Nice to hear it could be of help! There's just so much diversity among the early micros that are lost because nobody kept stuff in use once things got fairly standardized with popular platforms. Did you have a look at the files from the already prepared 5.25" disk? Not sure they are sys-gen'ed...

Yeah, that will illustrate things pretty well. This thread has some examples of what you can typically expect to see when things are working: https://torlus.com/floppy/forum/viewtopic.php?t=4272 Now, if the write precompensation and/or reduced write current is off, the lines will be more...

I am fairly certain generating the TG43 signal is handled by the SCSI to 8" floppy interface and not the drive itself in the case. But in any way, you can check the quality of the timing if you use the drive directly on the Greaseweazle and read a written disk back with the --raw setting. Load...

Sorry, my bad. I see there's something connected to that pin in the drive now, but the NEC-PC 8881 manual doesn't list it.

There can be lots of smaller details.. From what I have seen: FD1165-A: No door-lock solenoid (can be retrofitted) 25x2-pin flatcable connector with "upside-down" Shugart pinout 7-pin Molex KK 254 power connector Jumpers DX, RD, DH, DR, DL, HD Type 1: No door-lever guard (can be retrofitted)...

Using directly on an old PC will probably be difficult, since this one can't be jumpered for "Disk Change". You get index-pulses all the time and PC-type disk controller won't be happy with that. Yeah, and it's an absolute must for writing double-density disks. Indeed so. You are right here...

Your FD1165A looks identical to my FD1165A from 10 years before, except for mine not having the front panel. I did briefely scribble down the schematics in the area before the data input of the VFO. Data output from the VFO goes to one side of the RD jumper, the undecoded RD goes to the other...

For reference, if anyone else needs this info, what I mean by "ordinary 5-bit per nibble GCR", is to the point the GCR scheme described in section 3 of the Shugart SA800 Double Density Design Guide. This also matches the encoding @Chuck(G) arrived at by reverse-engineering back in 2013. The...

...and data! Did add it as a modulation type to Eric Smith's flux decoder tool, which he wrote in Python. Not the fastest thing around, but it got quite a nice PLL implementation.

Tried to do some manual decoding from the timing-states, and yeah.. Seems like this is the ordinary 5-bit per nibble GCR. Seems like it's using ordinary "CCITT False" CRC-16 checksums too.

Ok, after a lot of fine-tuning the final version is now done. Due to keeping surfaces nice and the details of the print sufficiently strong, especially the mounting-clips and tabs.. If you have an extrusion printer, I will recommend printing at -100 degrees rotation on the X axis. 3D-print...

Yeah, I think that document might be onto something. The timing is a lot closer to 1T 2T 3T rather than the 1T 1.5T 2T of MFM.

When I briefly had a look at them after reading, it seemed to have three bands of timing states, indicating some kind of Double-Density modulation scheme. However, the bands were quite asymmetric compared to standard MFM so HxC was not able to make sense of it. I can examine it a bit closer...

These are soft-sectored, single-sided 8" disks. I was able to borrow some RDOS-II system disks to read using a raw flux-level format, but I don't have any particular hardware at hand. Disks are marked: Processor 8085, Version 2300-5802-02/01, Self Instruction (Drive 0), 2-26-81 Processor 8085...

Does anyone have a reference about the low-level floppy format of GenRad RDOS-II? It seems to be quite non-standard stuff, or just very non-standard timing.