Re: R4400 vs. R10000 thoughts or benchmarks

New Message Reply Date view Thread view Subject view Author view

ken sartor (sartor++at++bur.visidyne.com)
Fri, 31 Jan 1997 13:54:00 -0600

At 10:53 AM 1/31/97 -0800, Curt Bryan wrote:
>Hello,
>
>I guess this is really more of a general SGI question but who would know
>better than all of you. Has anyone seen any information that offers a
>direct comparison between the R4400 and R10000 CPUs. Someone might also
>have first-hand experience with them both? The problem I'm having is that
>the benchmarks for the R4400 are SPEC92 and SPEC95 for the R10000. I
>haven't figured a way to normalize the benchmarks to get an
>apples-to-apples comparison. Any help would be appreciated.

On applications of interest to me (floating point intensive), i have
found a R10000 running at 175 MHz to be 3-4 times faster than
a R4400 running at 200 MHz. I had to compile the codes with the n32
flag to get these benefits, and they were *not* graphics applications...

Hope this helps.

ken

=======================================================================
List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/
            Submissions: info-performer++at++sgi.com
        Admin. requests: info-performer-request++at++sgi.com

New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:54:31 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.