home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!menudo.uh.edu!menudo.uh.edu!wagner
- From: wagner@grace.math.uh.edu (David Wagner)
- Newsgroups: comp.sys.mac.hardware
- Subject: Re: Motorola's 68060 for the Mac
- Followup-To: comp.sys.mac.hardware
- Date: 21 Jan 93 09:14:30
- Organization: UH Dept of Math
- Lines: 19
- Message-ID: <WAGNER.93Jan21091430@grace.math.uh.edu>
- References: <1993Jan17.065150.25462@wixer.cactus.org> <1993Jan17.192745.13494@netcom.com>
- <1993Jan18.024452.20791@bilby.cs.uwa.edu.au> <mhk=Av+@engin.umich.edu>
- <noah-200193075654@noah.apple.com>
- NNTP-Posting-Host: grace.math.uh.edu
- In-reply-to: noah@apple.com's message of 20 Jan 93 16:14:40 GMT
-
- It's nice to have all of these facts about size of registers,
- data bus, clock speed, etc. But nobody had said anything
- about numbers of registers. For example, the 68881 has
- 8 extended-precision floating-point registers. By way of
- comparison, the RS/6000 has 32 (double precision?) floating-
- point registers. I believe the comparisons for integer and
- address registers are similar, but I don't actually know.
-
- Clearly a processor can run faster if it can keep data in registers
- rather than writing values out to memory and reading them back in
- again when needed. Cache helps in this regard, but not quite as much
- as registers. However binary compatibility for software demands that
- processors have the same number of each type of register, or else only
- the minimum number of registers will be used. Hence I would expect
- that the 68060 will have the same register architecture as the
- 68030/68882.
-
- David H. Wagner
- You didn't hear it from me!
-