home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!dtix!darwin.sura.net!zaphod.mps.ohio-state.edu!caen!batcomputer!theory.TC.Cornell.EDU!eacj
- From: eacj@theory.TC.Cornell.EDU (Julian Vrieslander)
- Newsgroups: comp.sys.mac.hardware
- Subject: Re: 68LC040 and SoftwareFPU
- Message-ID: <1993Jan24.063533.6155@tc.cornell.edu>
- Date: 24 Jan 93 06:35:33 GMT
- References: <1993Jan23.153524.16154@gn.ecn.purdue.edu>
- Sender: news@tc.cornell.edu
- Organization: Cornell Theory Center
- Lines: 19
- Nntp-Posting-Host: theory.tc.cornell.edu
-
- In article <1993Jan23.153524.16154@gn.ecn.purdue.edu> jonkatz@gn.ecn.purdue.edu (Jonathan W. Katz) writes:
- >Subject line pretty much says it all. Do you think it'll work?
-
- If the 68LC040 lacks an FPU, and Apple releases a Mac with that chip, they
- might very well provide a new System version with built-in FPU emulation.
- They are already doing something similar. The FPU in the 68040 does not
- do trig and transcendental functions, like the 68881/68882 does, so these
- functions are handled in software (as traps, I believe). This means that
- the 68040 Macs do not handle some math operations as fast as Macs with
- 68030/68882 chips.
-
- Does the 68LC040 have a coprocessor interface that can talk to the 68882?
- That might be a cheaper upgrade for those users who need more math speed,
- than the 68040 swap.
- --
- Julian Vrieslander
- Neurobiology & Behavior, Mudd Hall, Cornell University, Ithaca NY 14853
- INTERNET: eacj@theory.tc.cornell.edu BITNET: eacj@crnlthry
- UUCP: ..cornell!batcomputer!eacj
-