[RC5] Linux OGR code not optimized?

dan the person motion at es.co.nz
Wed Sep 20 01:57:58 EDT 2000


"Martin Schröder" wrote:

> Hi,
> I just noted this madness (machine is a PII/400)
> - Linux native¹ gives 1,886,314.89 nodes/sec
> - NT4² in VMWare under Linux gives 2,402,656.63 nodes/sec.
>
> Does VMWare just screw the clock or is the Linux code poorly
> optimized? Apparently it is, as
> http://n0cgi.distributed.net/speed/query.cgi?cputype=all&arch=0&contest=ogr
> shows 2,508,889 nodes/sec for a PII/400.

Yes, the RC5 core is mainly assembly, so it's the same on both platforms, the OGR code is mainly C code so performace depends on the compiler.  What compilers has dnet used on linux?  I understand the prereleases for gcc v3 have the new p6 backend that was sponsoured by intel, can produce much faster code i hear.

--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list