[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <432C4C52.6020402@heidelbg.ibm.com>
Date: Sat, 17 Sep 2005 19:03:14 +0200
From: Christian Karpp <xn@...delbg.ibm.com>
To: john-users@...ts.openwall.com
Subject: 2 Qs on performance and benchmarking
Hi,
in a previous mail I read
> John's 1M c/s at traditional crypt(3) (on PPC G5 1.8 GHz or P4 3.6
> GHz) roughly translates to:
> 1M * 25 * 64 / 3 = 530 Mbps at 3DES
Q1: How does this estimation break down for other CPU types / clock
speeds? i.e. How will the "1M c/s" number vary?
Q2: Is there some further documentation on john's benchmarking
available? e.g. What's the difference between real c/s vs. virtual c/s?
How do the results map to Mbps, etc. And, some info or links for further
reading about the algorithms bechmarked would be nice as well.
AtDhVaAnNkCsE, Xn.
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
