[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: MCL 2.0.1 and Accelerators?
- To: abegel@media.mit.edu, info-mcl@digitool.com
- Subject: Re: MCL 2.0.1 and Accelerators?
- From: reti@cambridge.apple.com (Kalman Reti)
- Date: Fri, 9 Jun 1995 14:41:48 -0500
- Cc: abegel@media.mit.edu
- Sender: owner-info-mcl@digitool.com
At 13:29 6/9/95, abegel@media.mit.edu wrote:
>I've created at application in MCL 2.0.1 that runs on almost every 32 bit
>clean computer that apple makes, but from feedback from several users,
>it doesn't seem to work on LC's with 030 accelerators.
>Here's one user's testimony.
>
>> I isolated the problem to the accererator software & hardware combo. The
>> board is an Harris "Performer LC" accelerator with an 030 33Mhz processer and
>> Math chip. The Gestalt manager thinks the machine is an LCIII or other 030
>> machine when the accelerator is running.
>
>We're not running any code that targets 030 accelerators and blows them out
>of the water, so is there anything on the MCL end that is incompatible with
>030 accelerators? (Our code requires 32 bit memory addressing on and
>5 megs of RAM. But these requirements were met by the users with problems.)
MCL uses some low-level memory-management hardware instructions in
some cases, and sometimes accelerator manufacturers aren't 100%
compatible with Apple hardware in this area. A rough rule of thumb is
that if the accelerator can run Apple's VM, then MCL is likely to work on it.
The MCL FAQ lists accelerators which were known to work/not work at the
time it was written, unfortunately I don't think it is up-to-date.
>
>Andrew Begel
>MIT Media Lab
>Epistemology and Learning