Re: Gemini 80-Bus System
 

I have it running now and the file is slowly building. For 'nnnn' in the example above I have simply used the formula:

Which with the Rodime 201 (320 cyls, 2 hds, 32 sec/trk, 256 bytes/sect) equates to: A000h

It's going to take a while....

Re: Gemini 80-Bus System
 

Yes it should take a while - your calculation looks correct.

At least you don't have to watch it - and is probably the easy way.

Re: Gemini 80-Bus System
 

Here it is, and it seems to work fine. As with the gem925MFB.dsk it won't boot without the BOOTMFB program you created.

Enjoy!

Re: Gemini 80-Bus System
 

Thanks for that.....Well done. I was surprised that it was only 5MByte - the
Rodimes I had seen were 20MByte but I checked the model number and
the size is correct - so you did well.

Re: Gemini 80-Bus System
 

Thanks, yes its a 5Mb disk. It appears as A:, obviously. B: is the First floppy.

Re: Gemini 80-Bus System
 

I am a little puzzled how the bios was created though. When using GENSYS I could not get it to build a bios with a single 5Mb partition.

e.g.

I would have loved just to use something like the following in the GENSYS .cfg to create an A: drive using the whole disk, however this doesn't work, it just raises an error when tryin to generate the bios.

There has to be a number specified e.g.

However, that gives a different error when using GENSYS. So far the only way to get anything to work is to use the following, which gives me 2 Winchester partitions, A: and B:.

Re: Gemini 80-Bus System
 

This maybe significant. Using disked I command drive A reports 340 cylinders
this is beyond the end of the disk. A websearch gives 320 cylinders for
rodime 201. The bios appears to have the wrong DPB.

Just maybe that was why the transistors that drive the stepper motor in that
the bios tried to seek beyond the end of the disk (I don't know).

Just check on your system how many cylinders the bios is set for using
disked I command.

Re: Gemini 80-Bus System
 

It reports 340 just as you had identified. I wonder why?

Is it worth me using disked to see if I can read tracks beyond 320 or is that a little dangerous.

Re: Gemini 80-Bus System
 

Check the spec - the version I found on the web stated 321 cylinders.

I would not recommend going higher, with an mfm drive I don't imagine
there will be anything other than the seek motor bashing on the endstop. In
those days they did not have intelegent controllers on board the inteligence
is provided by the xebec which needs to be setup by the configure drive
characteristics command.

I cannot remember the spec - such a long time ago - but I think that
the rodime had ramped seek i.e. the seek accelerated time was not fixed
but accelerated. You could check the set drive characteristics command
but that might also be incorrect.

The fact that the drive transistors were blown - do you know if these
were for the rotation or stepper motor drive?

If you do decide to do it go one track at a time > increments the track
and then d to display the sector and start a few cylinders below 320.

Personally I would not push a hard drive beyond its spec - but you might
have the spec in your documentation.

Re: Gemini 80-Bus System
 

The transistors were the main motor drivers not the stepper. The spec I have says 320 cylinders.

One thing I did wonder was whether I should park the disk somehow before moving house. I'll not be entrusting the thing to the removers, but it will still have to travel 220 miles south in the back of my car.

Re: Gemini 80-Bus System
 

Again check the documentation - it should tell you. If so you should find gemini
have a park program.

I think the 20MByte drives did not need park and I suspect the ramped seek
rodimes don't as they are voice coil controlled. Stepper motor drives generally
benefit from park.

You need to check with the documentation to be sure. I don't remember having
to write a park program for the 20MByte.

Re: Gemini 80-Bus System
 

I have done a bit of disassembly of the simon rom and found a problem with
the scsi emulation that caused simon to fail to detect the scsi hardware. This has
been rectified.

I have not completed the disassembly but broadly found my way round some areas. If
you are disassembling I can send you the partial disassembly which has some labels
identified.

I have also harmonised the way bus extensions work - to make it easier to add more. Because it appears that some versions of linux don't supply the required files to build
device drivers with their distributed kernel and booting a custom kernel is more difficult
than it used to be I have included the parallel port driver code into xbeaver. Would be
nice to get a usb version of the bus-extender going and would probably use user-space
usb to put the driver into xbeaver as another bus extender.