home *** CD-ROM | disk | FTP | other *** search
- Path: news.teclink.net!usenet
- From: rad@teclink.net (rad)
- Newsgroups: comp.sys.amiga.misc
- Subject: Re: Speed: 68040 vs. 68060
- Date: 2 Mar 1996 06:35:12 GMT
- Organization: TECLink, Inc.
- Message-ID: <1295.6634T1437T2109@teclink.net>
- References: <john.hendrikx.4hrx@grafix.xs4all.nl>
- NNTP-Posting-Host: tc2_11.teclink.net
- X-Newsreader: THOR 2.22 (Amiga;TCP/IP) *UNREGISTERED*
-
- John Hendrikx <john.hendrikx@grafix.xs4all.nl> wrote:
- >In a message of 26 Feb 96 Haavard Jakobsen wrote to All:
-
- >> In a message of 22 Feb 96 Lee Huggett wrote to All:
- >> >> SysInfo is crap, that 40MIPS is really wrong. My 040/40 does that.
- >> LH> I was mearly stating a fact Sysinfo does give 40 MIPS for an '060
- >> The 060 can't run *more* than 2 instructions at once.
-
- HJ>> This is not quite true. It's working on more than 2 instructions at the
- HJ>> same time in the pipeline doing datafetch and so on. (see
- HJ>> 68060Guide.txt on aminet (or was it 68060.txt??))
-
- >Okay, I should have said the throughput is never higher than 2
- >instructions/cycle :-)
-
- It sounds like someone should correct either 68060Guide.txt (or 68060.txt).
- Motorola's data sheet (M68060/D) states 2 integer & 1 branch or 1 integer, 1
- float & 1 branch per cycle. That of course is peak, expect about half that in
- normal programs.
-
- --------------------------------------------------------------------------
- - Richard Deken E-Mail: (personal) rad@teclink.net -
- - VLSI design engineer (AuE business) rad@aue.com -
- - Advanced Microelectronics PGP public key available -
- --------------------------------------------------------------------------
-
-
-
-
-
-
-