[RC5] Key Rate Question

John Campbell jcampbel at lynn.ci-n.com
Thu May 6 15:28:10 EDT 1999


On Thu, 6 May 1999, Attila Csipa wrote:

> 
> On Tue, 4 May 1999 22:21:55 -0400
> "Aaron Kramer" <aaron.kramer at ins.com> wrote:
> 
> >    The Cyrix and AMD/Nextgen designers worked hard to increase the 
> >    efficiency of their integer math components since integer
> > But my experience has been the opposite. Anyone see similar results?
> > I have a AMD K62-450 running Solaris 7 that keys 750,000 keys/sec. I 
> > expected something over a million.
> 
> I found Cyrix to be faster (300.000 k/s on a P166+), and AMD to be
> somewhat slower (330.000 k/s on a K6-200), but still better then Intel.
> An Intel Pentium on 100MHz did about 130.000 k/s. BTW The picture
> changes drastically when you change from RC5 to DES, AMD is (was?) much
> better at that. 

	The difference between DES and RC5 is even more pronounced when you
start getting non-x86 architectures into the picture. I've got an otherwise
idle Sun Enterprise 5500 with dual Ultras running the client... doing RC5,
each processor manages around 650kk/s... my Celery 300 does better than
that. On DES, though, it benchmarks at 8000kk/s, per processor, which is the
highest rating I've personally seen on a single CPU.

> Operating systems may play a role in speed as well
> (scheduler/priorities).
> 
	Yup. The aforementioned Celeron dual-boots Linux and Win98. Under
Win98, it does around 750kk/s RC5. Under Linux, it's more like 850kk/s.

---
John Campbell
jcampbel at lynn.ci-n.com

QotD:  There may be a hundred stances and sword positions, but you win with
just one.

--
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