[RC5] Block times : an explanation

TwinTowers || rc5crack at kuala-lumpur.inet-one.com
Fri Jan 16 09:56:13 EST 1998


> 
> Kwoody wrote:
> > 
> > My dx2/66 did an rc5 block in about 1:10 mins averaging about 63kkeys/sec.
> > 
> > Now with DES stuff it does about 126kkeys/sec with a block size of 2^28
> > it takes well...lets just say I havnt gotten through a block yet and its
> > been well over 2 hours.
> > 
> > is this right? My cyrix 166+ does over 480kkeys/sec w/block of 2^28 and
> > it still crunches through a block about 3 mins slower than an rc5 one.
> > 
> > Whats up with that?
> 
> Ok, here's an explanation :
> 
> On DES, testing a key and its complement is faster than testing each key
> separately. Clients test all keys in a block *AND* their corresponding
> complementary keys. When you have completed a 2^28 sized block, you have
> checked 2^29 keys.
> 
> -- 
> R»mi		Don't waste your computer's time. Distribute it!

Something strange in the performance difference between Pentium and PPro! 
My PPro-200 does 919Kkeys/sec, whereas my oc'ed P-180 does 856Kkeys/sec 
(extrapolated to 951Kkeys/sec at 200). Is something wrong here?

Both clients use their respective optimised code, hard-coded in the 
config. The benchmark figures are closed to the accummulative rates 
reported in the log after running for more than 24 hours. Both are linux 
client v2.7002.383.


TwinTowers ||

> 			http://www.distributed.net/
> 	    RC5 cores source code : http://wwwperso.hol.fr/~guyom001/
> 
> 
> --
> To unsubcribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
> rc5-digest subscribers replace rc5 with rc5-digest
> 
> 


--
To unsubcribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list