22 Mar
2006
22 Mar
'06
11:53 a.m.
Hi Chuck,
The most likely candidate is that you're running
afoul of a timing issue.
Thats what I suspect as well.
A good candidate is the reset delay (140 usec
minimum). Since TeleDisk
I don't think it's reset - I wait 55ms (not us) with reset LOW, then 55ms with
reset high before proceeding - that should be plenty of time.
supports 1M/2.88MB data rates as well, it also issues
CONFIGURE commands to
the FDC when it changes rate--there might be something there. And a data
Is this a 82077 related command? The uPD765 documentation does not mention
a CONFIGURE command - only SPECIFY which just sets step/head-load/unload
rates.
rate change is always accompanied by a controller
reset--in theory it
doesn't have to be that way, but I found that some controllers gave very
erratic operation if this wasn't done.
I reset and configure the FDC at the beginning of any operation, however I
am currently NOT resetting it for each rate/mode attempt during the analysis
phase - are you saying that I should do this? This will cause all kinds of
disgusting things like the controller forgetting what track it's on... not to
mention slowing down analysis substantially as the controller will have to
be re-initialized and probably another whole revolution will slip by during
the chaos...
I'm inclined to think that this is not the problem either, because the user
reports that he cannot WRITE my Osborne SD images ... During WRITE, I
only setup the controller once (unless the mode/rate changes which it
doesn't in the SD disks) - and the controller is reset just before the initial
setup takes place.
If no other solution turns up, I'll try this as a last resort, however I'd make
it an option, so that the performance degradation and reseeking will not
have to occur if you have a better behaved FDC (this is the first one I've
seen with these problems - and we don't know that this is the cause yet).
While using the 765 data sheet might be good enough
for most things, you
should really be using the Intel 82077 as a reference--there are some
differences.
I guess it's too much to hope for that the manufacturers implement backward
compatibility to the 765 properly ... ?
Thanks for the input - I'll check out the 82077 data sheet (downloading it now)
and see if comparing it to the 765 gives me any ideas.
Dave
--
dave06a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.parse.com/~ddunfield/museum/index.html