[RC5] New competition and csc challange

Chris and Carrie Harrison carchris at texas.net
Mon Nov 15 17:55:44 EST 1999


On Fri, 12 Nov 1999 11:30:59 -0500, Jeff Gilchrist wrote:

>----- Original Message -----
>From: Jim C. Nasby <jim at nasby.net>
>To: <rc5 at lists.distributed.net>
>Sent: Friday, November 12, 1999 1:00 AM
>Subject: Re: [RC5] New competition and csc challange
>
>
>> Well, I can't give you the 'official' opinion, because there
>> isn't one.  In general though, we're glad to see the
>> 'competition' that Dcypher brings; it helps motivate us a
>> bit more. :)
>
>Has anyone done any benchmarking to see if the Dcyper client is
>faster/slower than the Distributed.net client?
>
>All I know right now is that the CSC key checking is MUCH slower than DES
>and even slower than RC5.
>
>Jeff.


Here are some quick benchmarks i collected for dcypher.net CSC contest.
To get keys/sec divide Kbps by 64.
p2-266       45000Kbsp       703kkeys
p2-400       70000Kbsp      1092kkeys
k6-3/400   60500Kbsp        945kkeys
k7 600     120000Kbsp        1906kkeys
k6-200       14000Kbsp          218kkey(Integer code is faster than single issue mmx)

You will find that on dual issue MMX cpus, CSC is about the same speed as d.net RC5 for P6 family and about 2x faster than d.net CSC.
K6-2 users will be significanly faster than d.net rc5.
K6 users will still suffer =)

To sum it up, Dcypher.net has faster clients for dual issue mmx boxes.
Unfortunatly I dont have any non mmx machines around to bench..


Later
Chris



--
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