home *** CD-ROM | disk | FTP | other *** search
- Path: fc.hp.com!news
- From: koren@hpsrk.fc.hp.com (Steve Koren)
- Newsgroups: comp.sys.amiga.datacomm,comp.sys.amiga.advocacy,comp.sys.amiga.misc,comp.sys.amiga.networking
- Subject: Re: New Press Release!
- Date: 14 Mar 1996 11:24:10 -0700
- Organization: HP Fort Collins Site
- Sender: koren@hpsrk.fc.hp.com
- Message-ID: <oj6rauvh6t1.fsf@hpsrk.fc.hp.com>
- References: <2937.6638T1404T1877@mozart.inet.co.th>
- <4hivul$nn8@server05.icaen.uiowa.edu>
- <4i440e$1b9@infa.central.susx.ac.uk> <4i5hlq$rn3@nyx.cs.du.edu>
- NNTP-Posting-Host: hpsrk.fc.hp.com
- In-reply-to: kathomas@nyx.cs.du.edu's message of 12 Mar 1996 21:07:22 -0700
- X-Newsreader: Gnus v5.0.9
-
-
- kathomas@nyx.cs.du.edu (Karl Thomas) wrote:
-
- > Sorry. Motorola's benchmarks showed the '060-66 being slower than a
- > Pentium-66 in integer calculations and a little faster in floating
-
- This seems roughly correct. Here are benchmarks for a scene rendered in
- Lightwave 4.0:
-
- Stock 4000/040/25 15m 19s
- Cyberstorm 060/50 4m 10s
- Pentium 100/NT 1m 52s
- P6/200/NT 0m 37s
-
- I've never seen any sort of real world test that shows the 060 to be
- comparable to any but the very slowest Pentiums. And note in the above
- that P100 is far from the fastest Pentium available, while the 060/50
- *is* the fastest 060 available. If you assume that a 66 Mhz Pentium is
- 66% the performance of a P100, then the P66 will complete the above test
- in 2m 50s, which is still a good bit faster than the 060/50 at 4m 10s.
- This is probably explained by bigger cache, faster memory subsystems,
- etc.
-
- Now, can we *please* stop posting every single day to c.s.a.* that the
- 060 is as fast as a P133 on steroids? It is not true, save for perhaps
- conditions found only in theory. It has never been true. It will never
- be true, no matter how much one wishes. Lets give it a break for a
- while, shall we? I see this claim posted just about every day. I see
- little value in fooling oneself about the performance of 68K CPUs. All
- the even remotely real-world benchmarks show the 060/50 to be comparable
- only to the P/66, and then on only floating-point intensive cache
- resident code.
-
- - steve
-