[RC5] New Clients - are they slow?

Greg Wooledge wooledge at kellnet.com
Mon Jan 19 20:55:05 EST 1998

Peter French (tsdpf at ibm.net) wrote:

>   Old Client V2.6401         New Client V2.7003       GUI Client
>   1427 blocks                 18 RC5 blocks            1 RC5 blocks
>   200:06:45.45                08:29:13.43              02:00:55.49
>   528743.79 keys/sec          547701.32 keys/sec       147990.20 keys/sec
>                                                        2 DES blocks
>                                                        02:00:55.49
>                                                        591960.80 keys/sec
>  +- 7.135 Blocks/hour          2.11 Blocks/hour        0.5 Blocks/hour

The definition of a "block" has changed; they're no longer a fixed size.
Measures of blocks/hour aren't relevant only keys/sec are relevant.
But other people have addressed that issue already....

The reason I bothered to respond to this is that I have a related
observation.  Over this weekend, one of my computers at work did 7 RC5
blocks (probably due to a malfunctioning HTTP proxy server, no big deal).
When I came in and saw the tail end of the logfile, I saw that the RC5
and DES block reports both showed the same elapsed time; thus, both the
reported RC5 and DES keyrates were lower than the actual cracking rate of
this computer; the DES rate was slightly lower than it should have been,
and the RC5 rate was *much* lower.

This could be confusing to people.  I propose that the amount of time
reported by the client for the RC5 and DES cores should not be the
total time spent by the client.  Rather, separate RC5-cracking-time and
DES-cracking-time figures should be reported.  So instead of numbers
like these fictional ones

[01/19/98 12:00:00 GMT] Tot: 7 RC5 blocks 96:00:00.00 [21700.00 keys/sec]
                        Tot: 93 DES blocks 96:00:00.00 [577900.00 

we'd see numbers like these even more fictional ones:

[01/19/98 12:00:00 GMT] Tot: 7 RC5 blocks 14:00:00.00 [149100.00 keys/sec]
                        Tot: 93 DES blocks 82:00:00.00 [676500.00 

This would make the reported cracking rates "accurate", and would reduce

But given a choice, I'd rather see RC5/DES clients for all the common
platforms before worrying about little features like this.

# Greg Wooledge                         # "Daddy, why do those people 
have to
# wooledge at kellnet.com                  #   use Microsoft Windows?"
# http://kellnet.com/wooledge/main.html # "Don't stare, son; it's not 
#  I feel the need for speed...  DES II - http://www.distributed.net/des/

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