Just an update should anyone still be interested in my Winchester updates to the Gemini CP/M bios, I am making good progress.
Targets achieved:
- I have learnt Z80 assembler, still honing my skills but all OK so far.
- I have been through the Gemini V1.4 bios in detail and understand pretty much most of it.
- I have read the SASI spec and swallowed the XEBEC 1410 manual.
- I have started coding the changes to the Gemini Bios...
I am reasonably happy with my progress and to be honest would be quite disappointed if someone turned up with a Winchester version of the bios as I am having quite a lot of fun with it all. My aim has to have a light a touch as possible to the original code and this seems to be fairly easy given one compromise. i.e.
The Gemini bios supports SD and DD disks and auto senses, the density for the 4 possible floppy disks is stored in the upper nibble of a memory location with the lower nibble indicating the drive. The bios uses one bit to identify the drive (lower nibble) and one bit to identify SD (upper nibble) this means that a maximum of 4 drives is supported. If I accept the compromise (which I have) of supporting only two floppy drives I can then add two further Winchester drives (possibly 4 when emulating a Xebec 1410) with very little change to the Gemini bios code whilst at the same time, using 90% of it. This I am very happy with.
So far the only real change to the existing Gemini routines has been to prevent the density check for the Winchester and adapt the buffer-dma and dma-buffer code to support different sector sizes for Winchester and floppy. This last change was important to me as I wanted the bios to be able to support real period hardware, it turned out to be a very simple change. Naturally there will be additional code to control the Xebec and that is what I am working on as I write.
As usual, full story and progress is here;
https://glasstty.com/wiki/index.php/...ni_80-Bus_Saga