Re: What's wrong with Macintosh ?
Date: 1995/10/18
Message-ID: <mikek-1810951943400001_at_franklin-tty16.jvnc.net>#1/1
In article <palardyn-1610951933040001_at_cad88.cadvision.com>, palardyn_at_cadvision.com (Norman Palardy) wrote:
> In article <mikek-1610951153360001_at_franklin-tty14.jvnc.net>, mikek_at_krt.com
> (Michael Kranzdorf) wrote:
>
> > In article <45jg31$k0k_at_news1.cle.ab.com>, "Katy Y. Wang" <wangk> wrote:
> >
> > floating point coprocessor not installed.
> >
> > This is basically equivalent to a GPF on Windows. It's probably not
> > related to a floating point coprocessor.
>
> Sorry but I'll disagree. The single biggest cause is programs that try to
> use an FPU without checking to see if one is present first.
>
> Inits can cause this error, but hardly ever do.
OK, perhaps I was a bit too inclusive, but I really have solved these errors by removing inits. I read somewhere recently (unfortunately I don't remember where) that this error is the "default" of a case statement, so a lot of errors flow there. (Maybe that only applied to System 6 though - it was documented in an Apple Tech Note that this error had nothing to do with FPUs.)
> I've got another of Oracle's tools that dies because of
> exactly this (Power Objects) but installing an FPU solves the problem.
Power Objects runs on our Quadra 605's (68LC040, no FPU) and PowerMac 7100's (PPC 601, no FPU). Do you have to do something in particular to get the error? I can create a blaze db, a form, and run it on both Macs. I don't develop on either one, so I haven't done much, just this test.
The basic question is whether Forms Runtime requires an FPU. It appears to run on the two machines above. But again, I've only done some very limited testing, so I might be wrong.
Katy - you can try SoftFPU (available at all the mac sites - let me know if you can't find it) without buying a real one. Please let us know the outcome (and the Mac model)!
Michael Kranzdorf mikek_at_krt.com Kranzco Realty Trust 610.941.9292Received on Wed Oct 18 1995 - 00:00:00 CET