> Hi,
>
> I was wondering if anyone had any real world data on powerpc systems
> and/or suns (Of course, running netbsd). I could look up how many
> instructions per second each cpu is capable of, but what I would really
> like to see is something like how many hits a second they can take, or 
> how
> fast they compile userland. This is data I just can't seem to find, and
> was hoping that someone here would have, and be willing to share.
>
> Thanks,
> Al

hi al,

there are several issues to take care of in this topic.

first, SPARC-based systems (and, furthermore, sgi machines, 
unfortunately NetBSD's support for these is not that well as it is for 
sparc/sparc64) are well known for the high throughput they provide 
(i.e. mostly in comparison to x86-based stuff).

that is, static websites are delivered very fast, even if you have a 
'relatively slow' machine (read: Ultra 1 or even SPARCstation, from 
20MP down to the SS4). if you make heavy use of PHP, Perl et al. you 
will discover that you need a somewhat faster machine. IIRC sparc64's 
MP support is in the works but not finished yet.

here PowerPC comes up (if you don't have very fast USII/USIIi machines 
handy) because you can i) use the newest systems (e.g. Xserve G5) and 
ii) it is SMP-ready.

on PowerPC throughput is also more than good (IMHO -- a 400MHz Power 
Mac G4 seems lots faster using *one* CPU than the dual PIII/500 
IntelliStation i compared it to)...

furthermore and as my last paragraph ;) i'd like to say that it's 
surely best for you to run your own benchmarks on the machine(s) you 
have available, compare them (i.e. the results) against your needs and 
go for the winner ;)

HTH

-- 
mit vorzueglichster Hochachtung/best regards,

Timo Schoeler
//macfinity -- finest IT services | Triftstrasse 39 | 13353 Berlin | 
Germany
Fon ++49 30 25 20 30 20 | Fax ++49 30 25 20 30 19
PGP data http://www.macfinity.net/~tis/contact/PGPPKB_timo.schoeler.txt