[comp.sys.amiga.emulations] Brideboard, Keyboard buffer & HD emulation

gsgi_ltd@uhura.cc.rochester.edu (Gregory Gibson) (03/29/91)

I have noticed two problems with my A2286. 1. When using Wordperfect,
during the auto-backup-saving of my document, characters are lost
from the keyboard buffer.  Does this have to do with #2?

2. The HD emulation is very slow?  Would an 8bit hd card for the IBM speed
things up.  Would I need 40ms or is 20ms really really much much better.
I am leaving the 16bit slot open for more memory - is this stupid? Should
I put a 16bit HD card there?

I appreciate the time it takes to reply.

-Greg Gibson
gsgi_ltd@troi.cc.rochester.edu


-- 

Gregory Gibson 
gsgi_ltd@uhura.cc.rochester.edu or  gsgi_ltd@uordbv.bitnet
UUCP: {decvax,harvard,ames,rutgers}!rochester!ur-cc!gsgi_ltd

drysdale@cbmvax.commodore.com (Scott Drysdale) (04/02/91)

In article <13047@ur-cc.UUCP> gsgi_ltd@uhura.cc.rochester.edu (Gregory Gibson) writes:
>I have noticed two problems with my A2286. 1. When using Wordperfect,
>during the auto-backup-saving of my document, characters are lost
>from the keyboard buffer.  Does this have to do with #2?
>
>2. The HD emulation is very slow?  Would an 8bit hd card for the IBM speed
>things up.  Would I need 40ms or is 20ms really really much much better.
>I am leaving the 16bit slot open for more memory - is this stupid? Should
>I put a 16bit HD card there?

problems 1 and 2 are indeed related.  there is a hardware fix (lift an IC
pin and solder a jumper from it to somplace else).  if your bridgeboard is
out of warranty, i'll give you the fix.  basically, due to incomplete
address decoding, when the pc sends the amiga an interrupt (for instance,
when it is requesting a disk block from the amiga) it also whacks the
keyboard controller (8042) upside the head.  this causes two things - 
if a key is being processed by the 8042, it is lost.  this is the "lose
keys while fake disk i/o is happening" problem.  the other effect is that
it causes the 8042 to generate a keyboard interrupt to the 80286, and
this unnessary interrupt drastically slows down fake disk i/o.

>I appreciate the time it takes to reply.
>
>-Greg Gibson
>gsgi_ltd@troi.cc.rochester.edu

  --Scotty
-- 
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Scott Drysdale           Software Engineer
Commodore Amiga Inc.     UUCP {allegra|burdvax|rutgers|ihnp4}!cbmvax!drysdale
		         PHONE - yes.
"Have you hugged your hog today?"
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

liberato@dri.com (Jimmy Liberato) (04/03/91)

drysdale@cbmvax.commodore.com (Scott Drysdale) writes:

>In article <13047@ur-cc.UUCP> gsgi_ltd@uhura.cc.rochester.edu (Gregory Gibson) writes:
>>I have noticed two problems with my A2286....

>problems 1 and 2 are indeed related.  there is a hardware fix (lift an IC
>pin and solder a jumper from it to somplace else).  if your bridgeboard is
>out of warranty, i'll give you the fix... 

I think most Bridgeboards are out of warranty.  Can you publish the fix here
since we are a technically sophisticated lot? 

Thanks!

--
Jimmy Liberato   liberato@dri.com
                 ...uunet!drivax!liberato