[rc5] Block size

Eric Schaefer es5 at irz301.inf.tu-dresden.de
Tue Oct 28 23:32:04 EST 1997


How about this:
If block are larger it takes long to calculate it -> higher chance of lost
blocks if machine crashes. Besides this, if you use 30 bit blocks, it would
take 4 times as long to do them, which would make stats more unaccurate.
BUT, why talk, like the blind about colors? what do the people say who
should know (bovine,nugget,...)

BTW. does anybody know if ultrix clients are to come???


*****************************************
* Eric Schaefer  es5 at inf.tu-dresden.de  *
*     http://www.inf.tu-dresden.de/~es5 *
*                                       *
* Got any spare cpu-cycles?             *
* Donate at http://rc5.distributed.net  *
*****************************************


> 
> On Tue, 28 Oct 1997, Colin L. Hildinger wrote:
> 
> > I'm curious as to why the organizers didn't opt for a slightly larger
> > block size, say 2^30 instead of 2^28.  Was it just because they didn't
> > want to make the necessary changes to the client/server code?  Would it
> > have been a big deal?
> 
> A bigger block takes longer to check.  Smaller blocks result in a greater
> perceived speed.  People like it better that way.
> 
> Of course this probably isn't why the block size didn't change.
> 
> Joseph
> 
> **************************************************************************
> ** C. Joseph Fisk (mdmbkr)                           mdmbkr at chillin.org **
> ** Visit http://www.distributed.net - the world's fastest supercomputer **
> **************************************************************************
> 
> ----
> To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.
> 

----
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.



More information about the rc5 mailing list