IPF creation and disk writing via Kryoflux

3rd Party Software, Tools & Add-ons for KryoFlux
dk_spb
Posts: 15
Joined: Wed Nov 14, 2012 3:14 pm

Re: IPF creation and disk writing via Kryoflux

Post by dk_spb » Sun Dec 16, 2012 2:40 pm

Keir, thank you very much for your software.
Did you have plan support FM (not MFM) formats?

User avatar
mr.vince
Posts: 1873
Joined: Tue Oct 05, 2010 5:48 pm

Re: IPF creation and disk writing via Kryoflux

Post by mr.vince » Sun Dec 16, 2012 7:16 pm

IFW can correct me as necessary... I don't think FM is supported by the IPF transport layer.

dk_spb
Posts: 15
Joined: Wed Nov 14, 2012 3:14 pm

Re: IPF creation and disk writing via Kryoflux

Post by dk_spb » Sun Dec 16, 2012 7:44 pm

Does it mean that FM disk writing is not supported by KF?
KF support only IPF and ADF image types for writing.
I need have possibility to read/write for few similar FM formats (like marker, track header, few sectors with checksum without any interval).
Disk Analyzer is very good because author provide the sources and I can add formats.

dk_spb
Posts: 15
Joined: Wed Nov 14, 2012 3:14 pm

Re: IPF creation and disk writing via Kryoflux

Post by dk_spb » Sun Dec 16, 2012 8:05 pm

I am a retrocomputers collector and possibility to write FM format is very important for me. Computer is dead without software. Sometimes I have received software (as disk images) from my friend from other cities and often they have not old computer for make disk copies. And I need have possibility to create disk from images. In new year holidays I will visit to other cities for get software for old soviet SM-1810 (base 8086, use Intel iRMX OS). Disk format is very strange: 1st track is FM with 128 bytes per sector and other track is MFM 512 bytes. Are you sure that KF can't write FM tracks?

User avatar
mr.vince
Posts: 1873
Joined: Tue Oct 05, 2010 5:48 pm

Re: IPF creation and disk writing via Kryoflux

Post by mr.vince » Mon Dec 17, 2012 8:10 am

I an talking about what's been implemented into the transport layer of the IPF container. It means IPF could support it once it's been put in.

KF hardware does not really understand about FM, MFM or even GCR. All it does is write flux changes. Therefore there is no limitation in KF hardware.

dk_spb
Posts: 15
Joined: Wed Nov 14, 2012 3:14 pm

Re: IPF creation and disk writing via Kryoflux

Post by dk_spb » Tue Dec 18, 2012 5:36 am

I have added FM decoding into libdisk.
Now I have core dump in disk_close() during memfree(di->track.dat);
but I still try fix it.

dk_spb
Posts: 15
Joined: Wed Nov 14, 2012 3:14 pm

Re: IPF creation and disk writing via Kryoflux

Post by dk_spb » Sun Dec 23, 2012 6:55 pm

Ok. I have fix it and now I can convert KF stream to raw sector dump for Nemiga MD format.

But I have next questions for Keir and IFW.
I have tried read Agat 840K disk. Agat it is soviet clone of Apple II, but they have very quick FDC based on two 8255 (each per disk side). It takes only 36 seconds for full disk read. This is MFM format with 21 sectors (256 bytes) per track.
Each sector start after sync fail (index can be ignored) and contain two parts: address and data.
Sync fail is a 0xA4, pause for 2 us (a half bit) and 0xFF.
Address part of sector start with 0x956A.
In MFM 0100010010010010 (0xA4) 0 (half bit pause) 0101010101010101 (0xFF) 0100100100010001 (0x95)0001010001000100 (0x6A)
Or in hex 44922AAA A4888A22 (and last zero bit in next word).

I have dumped disk with -i2 and -i0 (see files for first five cyl and .ct file for -i2 and .txt file for dtc output here=2Mb)
I can found signature in CT file at offset 23D0: 88 92 45 55 54 91 11 44 4A AA (after shift to left for 3 bits: 44922AAA A4888A22).

I have tried to find the signature in stream with libdisk by Keir. I have modified one of format files:

static void *xx_write_raw(struct disk *d, unsigned int tracknr, struct stream *s) {
unsigned int i;
// try to found 1st half of signture
while (stream_next_bit(s) != -1) {
if (s->word != 0x44922AAA)
continue;
printf("\nfound = %x ",s->word);

// try to find 2nd half of signature
i=0;
while (stream_next_bit(s) != -1) {
i++;
if (s->word != 0xA4888A22)
continue;
}
printf("position %d=%x ",i,s->word);
}
return NULL;
}
And I can't find 2nd part of signature (1st part is OK) :-(

Questions:
1) Does it mean that bit stream in .raw and in .ct is different? Or my code is wrong?
2) Could anybody provide description of CAPS (.ct) format?

Thank you in advance,
Merry Christmas!!

robbo007
Posts: 32
Joined: Wed May 23, 2012 2:56 pm

Re: IPF creation and disk writing via Kryoflux

Post by robbo007 » Mon Feb 18, 2013 5:49 pm

This is just what I needed :)

Will it work for Apple II disks? I need to convert a Kyroflux Apple II disk dump to a .dsk. That way I can use the images with the CFFFA3000 or an Apple II emulator.
Thanks,
Rob

Rixa
Posts: 52
Joined: Mon Dec 12, 2011 12:09 pm

Re: IPF creation and disk writing via Kryoflux

Post by Rixa » Wed Aug 07, 2013 4:10 pm

I tried this tool with my original-release Dungeon Master. This got me something that appeared to fail in WinUAE, but to work in my Amiga when written to a disk.

Well, that surprised me.
I guess I have no point. Carry on.
Last edited by Rixa on Thu Aug 08, 2013 4:56 pm, edited 1 time in total.

User avatar
mr.vince
Posts: 1873
Joined: Tue Oct 05, 2010 5:48 pm

Re: IPF creation and disk writing via Kryoflux

Post by mr.vince » Wed Aug 07, 2013 6:42 pm

I'd be surprised if an original (pre 3.x) version would work...

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest