home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!sun4nl!tuegate.tue.nl!blade.stack.urc.tue.nl!stephan
- From: stephan@blade.stack.urc.tue.nl (Stephan Eggermont)
- Newsgroups: comp.sys.mac.misc
- Subject: Re: Soft_PC. How compatible is it?
- Message-ID: <6741@tuegate.tue.nl>
- Date: 21 Dec 92 14:27:26 GMT
- References: <1992Dec17.145445.968@slcs.slb.com>
- Sender: root@tuegate.tue.nl
- Organization: MCGV Stack, Eindhoven University of Technology, the Netherlands.
- Lines: 33
- X-Newsreader: TIN [version 1.1 PL6]
-
- Harvey Brydon (918)250-4312 (brydon@asl.slb.com) wrote:
- : In art. <29600@castle.ed.ac.uk>, tjc@castle.ed.ac.uk (A J Cunningham) writes:
-
- : We have encountered problems with software that does goofy copy-protection.
-
- Well of course. It won't run on some PC's too. Stupid copy-protection schemes
- depend on timing. Or else they have a dongle which has to be put on the
- parallel printer port.
-
- : We also have encountered a problem with software that uses the 80287. This
- : was blamed (by Insignia) on a timing problem, caused by defective code in the
- : PC application that makes a request of the 287, and grabs data from the result
- : register before looking to see if the computation was complete yet. This
- : timing problem does not occur if the pseudo-287 is disabled, but boy does it
- : take a long time to run. Insignia said that this could be remedied by one of
- : several things (depending on source language): (1) recompiling with a
- : different compiler option, if a high level language such as fortran, pascal or
- : c, (2) rewriting some of the code, if it is written in assembler. We got back
- : to our application vendor and they were not interested in doing any of the
- : above without big bucks since we were not running it on a 'real' PC where the
- : problem does not show up.
-
- Did they test that with all different combinations of 286/386/386SX/SL/SLC/
- 486SX/SL... + 287/387SX/487SX...? I'm not sure I'd want to use software from
- a company which doesn't act upon bug-reports.
-
- Or Insignia could make a version of SoftAT which always halts the 286 till all
- 287 instructions are finished.
-
- b.t.w. I've seen Turbo Pascal working.
-
- Stephan Eggermont
- stephan@blade.stack.urc.tue.nl
-