SAC.DYESGPF@E.ISI.EDU (07/06/88)
The two programs RD200.COM and RD100.COM for switching between z100 8" format and AT 1.2 meg format got me to really poking around in bios and now I have some questions/ideas that I would like feed-back on from this group. First - concerning the way these programs locate the BPB, while the search and compare methode in RDx00 programs work with original bios there can be situations where RD200 will not find the BPB (more details later). Does anyone know if the offset to the BPB is obtainable from the address refrences fol- lowing 0040:0061h. Also, since the media discriptor for the 1.2meg 5.25 is F9h, why not use the paramater block which starts 13 bytes later. If DOS 3.1 check the media identifier then the drive should read both the native format and the AT format by simply changing the second paramater table for the 8". The 80 tracks specified in the DOS disk parameter table does not seem to have any adverse effects.; I modified a copy of bios to hard wire 80 tracks, booted with it and had no problems with my 77 track disks (I used debug to confirm that the DPT was at 80 tracks after the boot-up). Any one know if this can be a problem (remember that DOS will not "write" to an unformatted sector). Now for the more aggressive ideas: 1. Although FORMAT.EXE cannot be used to get more than 77 tracks on the 8" controller, I have used ASMGEN to create a souce file from FORMAT.COM from DOS 1.25 and QFORMAT (for them other machines) and have figured out the steps needed to format to 80 tracks (and worked out the needed changes to bios for 80 tracks as a default - this causes RDx00 to need modification). PROBLEM - the CHRN (Cylinder, Head, Record, Number - 16 bits) does not seem to comply with the "text-books" and I still working on an algorithm to generate the CHRNs. I know that somewhere and sometime I read a complete discription of the required buffers between the sectors, the begining and end of each track and the code to generate the CHRN, but I can not locate the reference; any ideas please!!! This data on an AT disk would be very helpfull, but I don,t know anyone with access to a clone machine with the AT drive. 2. From what I do remember and a few calculations, it seems that the 8" format could be modified to 9 1k sectors per track with some 300 to 350 byte to spare, does anyone know if this has already been attempted. This would be 1.44 meg and not compatable with anybody (but then only machines with 8" controllers running HD drives can read/write 1.25 meg 5.25 floppies). The reason for doing this would be to store more data on a disk and with a bit of work, this format can be read by other machines (notice I didn't say it would be easy). 3. The final idea is to hook a 3.5" 1.44meg drive to the 8" controller and build a BPB for it, I believe the correct format is 18 512byte sectors, double sided, 80 tracks - does anyone know the media identifier code for 3.5 1.44meg? Any formatter program I come-up with from this information and/or bios mods will be posted to the PD1:<HZ100> directory at SIMTEL20 and generally made available to the INFO-HZ100 interest group. BTW, if anyone has access to an AT style machine and is willing to format a diskette at 1.2meg for me, please send me your USnail address and I will send you a disk with return postage, thanxs in advance. Al Holecek | ...probability factor of one to one...we have nor- <SAC.DYESGPF@E.ISI.EDU> | mality, I repeat we have normality. Anything you ----------------------- | still can't cope with is therefore your own problem.
GUBBINS@RADC-TOPS20.ARPA (Gern) (07/06/88)
I have also been interested in C: & D: 1.2Z and 1.2AT disk formats, although I do not have them. I also figured out you can cross the tracks and formats for 1.44M (now very unstandard, unless we all get together on it). Rumor has it that the 3.5 1.44M drives are 160 tracks per side and rotate at a completely different speed, however I lack proof. I would very much like to bypass 5.25" HD and use 1.2 to 1.44M 3.5". I would also recommend 512 byte sectors to stick to the 'standard' as disk utilities barf otherwise. The BIOS I/O, FORMAT-TRACK, and DISKCOPY.COM and FORMAT.COM would all have to be patched. A permanent patch would suit me fine, as I would really see no need to use the 1.2Z format. Cheers, Gern -------