[RC5] Computational Power, Linux VS WinNT4 & P-II VS P-III

Ferry van Steen fvansteen at basix.org
Thu Jan 25 10:08:08 EST 2001


Hey there,

I got a question, which answer is pretty important to me. I'll attach logs
to clearify it.

We've got a local area network here on which I installed a keyproxy
(172.16.1.14) I'll ask you ppl to take a look at 5 computers. All other IP
addresses are not important since they do update slowly but it are only
P166's and they take a long long long time to crack OGR packets.

Anyways the 5 computers I'll be talking about have a fetch setup of
RC5=100,OGR=1 and the sequence is RC5,DES,CSC,OGR

4 of the computers (192.168.0.1-4) are P-II 400's running SuSE 6.4 with
256MB (which should matter much cuz dist client only uses about 1MB
right?) the 5th (172.16.1.104) will be the comparision machine and it has
a P-III 450 with 128 MB running Windows NT4 Workstation.

Ok to come to my question, the P-III (172.16.1.104) updates with the
keyserver 4 times more often than the P-II 400's and I'm wondering where
this huge speed difference is comming from. Does it come forth from the
P-III SIMD which can speed up several calculations by a lot, is the linux
just not as fast as the windows clients and if so how come, or could it be
the linux kernels on the P-II 400's since we're (still) using the standard
kernel that came with the distribution and thus isn't fully cpu/system
optimized?

Anyone know which of these (if any of these) causes the problem or which
combination of it?

Kind regards,

Ferry van Steen

--
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