[RC5] So long RC5, hello SETI

Greg Lee tatsu at tatsu.dynip.com
Thu May 20 10:53:54 EDT 1999

On Thu, 20 May 1999, Bernd Bernie Meyer wrote:

->> so they (nearly) don't slow down the computer.
->This is not a true statement in all cases.
->Under Windows 3.1, depending on the applications you're using, you might
->notice a *tremendous* slow-down with the RC5/DES client running.  I did.
->(With Linux, though, there is not a noticeable performance issue.)
-Actually, there is. Even running at its nicest, the rc5 process still
-gets slightly less than 10% of the CPU on a single processor x86 linux
-machine that has another process running full-out at default priority.
-That's due to the "fair scheduling" that linux uses by default.
-For most people, that isn't a problem. However, when you run iterative
-numerical simulations that take several hours to complete, those 10% can
-make a significant difference.
-Of course, if you do that, you probably don't have any trouble stopping
-and restarting the client appropriately.
-To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
-rc5-digest subscribers replace rc5 with rc5-digest

On my Dual PII 262 running Linux I have never had RC5 slow down anything
that wanted cpu time. The reason some things probably look like it is
getting slowed down is that the program tried to write to the disk and had
to wait for it to finish thus giving the rc5 program a chance to run in
the meantime. 

I have run a ray tracing that took several hours and the entire time it
got about 99.8% of the cpu time which is the most anything I have seen
ever gets.
   E-mail  : tatsu at tatsu.dynip.com              
   Web Page: http://tatsu.dynip.com/            
   PGP Key : http://tatsu.dynip.com/tatsu.asc   
   Uptime  : 1 days 3:14
   Load    : 2.01 2.04 2.00

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