[RC5] DES-Test, Day 2

John Campbell jcampbel at lynn.ci-n.com
Fri Jan 8 17:37:03 EST 1999


On Fri, 8 Jan 1999, Jim C. Nasby wrote:

> I disagree as to the need for smaller blocks. The 486/16 that I used to run RC5 
> on would do 1 31 bit RC5 block per day, which means it would do a 28 bit block
> in 3 hours. Assuming that DES is 50% faster (I can't get to my 486/100 to check
> right now), that means it would take only 2 hours on a 28bit DES block.
> 
	I've got a pair of 386-25s that, by the numbers from the current
test, do a single 2^28 DES block in 14-18 hours. While that's probably
sufficiently fast for the upcoming contest, if the next one has even shorter
deadlines, I'll have to sideline them and my other 386es (keep them cracking
RC5... no deadlines) for DES-II-4 (DES-IV?). While this doesn't seem like a
big loss - a modern machine can do in minutes what it takes them days to do
- those machines' contributions multiplied by the number of 386es in the
effort might add up to something significant (I admit, I'm not certain...
it'd be nice to see some numbers on the relative contributions of the
various x86 chip classes, but I don't think the data exists)... and we want
all the CPU power we can get, right?

---
John Campbell
jcampbel at lynn.ci-n.com

QotD:  Do you ever feel like a Quentin Tarantino character in a Walt Disney
world?

--
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