[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian needs more buildds. It has offers. They aren't being accepted.

On Thu, Mar 18, 2004 at 11:07:59PM +0100, Thiemo Seufer wrote:

> BogoMips is 0.5 * MHz for R4xxx and 1.0 * MHz for R5000. The real
> difference is smaller than that. I'd guess a R4400-250 is a bit
> faster than a R5000-150 for integer and a bit slower for fp.
> I didn't do benchmarks, though.

Well, you know that there's a germany saying: "Wer viel misst, misst viel
Mist." So far to benchmarks. ;)

> [snip]
> > But when using a "real world benchmark" named dnetc ;)), the m68060 was as
> > speedy as a R10k at 195 MHz. 
> > Of course the dnetc benchmark is quite, uhm, well, wrong for some reasons.
> > Dnetc makes extensive use of commands that the mips CPU don't have (bit
> > shift left/right or so, IIRC).
> Rather rotate insns, shifts are there since MIPS I. Outside of crypto
> challenges the outcome will be very different. :-)
> > OTOH, the m68060 has no full FPU, which might
> > give significant drawbacks in build time for such packages like axiom.
> > Point is: 
> > There might always be some differences between CPUs, be it on the same arch,
> > be it cross arch. The 060 is f.e. 2-4x faster than the 040, although it's
> > just clocked 10 MHz fast.
> A R10k has roughly 3 times the integer speed of a R4xxx of same
> frequency, and even more for fp.

Yes. To contribute another UBC - useless benchmark comparison: 
An 4p r10k@250 Onyx2 IR has a dnetc keyrate lower than my old PII-350. But
for seti@home... fasten your seat belts! The Onyx2 runs circles around the
PII... ;))

Therefore I think that adding r4ks as buildds would make sense and help the
mips port to keep up. 

> [snip]
> > Uh, is the L2 Cache support on mips in the meanwhile?
> Was only an issue for r5k, fixed upstream. I hope we get a 2.4.25 soon.

Ah, just another reason to add more r4ks... ;)

Ciao...              // 
      Ingo         \X/

Reply to: