AW: [RC5] Athlon XP/MP even faster? K5 faster than K6!

Kagain kagain at
Thu Oct 18 23:26:53 EDT 2001

I've found something strange at .
Many people submitted their speeds there.

Did you know that the k5 is faster than the k6 (not in general but regarding
rc5 only)?

133MHz are faster than 180MHz:
AMD K5 133MHz 326,625 keys/sec
AMD K6 180MHz 324,639 keys/sec

Both at 200MHz:
AMD K5 200MHz 434,804 keys/sec: ~2174 keys/sec/MHz
AMD K6 200MHz 330,745 keys/sec: ~1653 keys/sec/MHz

?? Comparabel with Intel nowadays or was the K6 bad in general ??

Speaking about todays CPUs: To build an rc5-friendly system don't use an
Athlon but 2 Duron CPUs instead!
Shouldn't be much more expensive but it should be double as fast :-)

AMD K7 Athlon Thunderbird 1000 3,541,940 keys/sec
AMD K7 Duron 1000              3,524,049 keys/sec

With greetings from Germany

-----Ursprungliche Nachricht-----
Von: owner-rc5 at
[mailto:owner-rc5 at]Im Auftrag von Bjoern Martin
Gesendet: Donnerstag, 18. Oktober 2001 21:45
An: rc5 at
Betreff: RE: [RC5] Athlon XP/MP even faster?


>Its possible a faster core could be developed to take advantage
>of their new features but at the moment they run the same core
>[rc5 core #6 (RG/HB re-pair II)] that the Athlons and Tbirds use.
>Here is a quick comparison of keys / MHz on the Athlon systems I
>have access to.
>Original K7, 800 MHz: 2,852,953 keys/sec: ~3566 keys/sec/MHz
>Athlon MP 1.2 GHz:    4,230,781 keys/sec: ~3525 keys/sec/MHz
>so it looks like with the current core Athlon MPs only scale
>linearly with MHz compared to Original Athlons.

Yep, same for me. I never found a machine that didn't scale
linear within about 5% discrepancy.

To make things clear, I could also ask: does SSE improve
the performance of the client regarding RC5?



boris59 at -> PGP mails welcome, public key on request
Member of RC5-64 effort by - team #6941

To unsubscribe, send 'unsubscribe rc5' to majordomo at
rc5-digest subscribers replace rc5 with rc5-digest

More information about the rc5 mailing list