[RC5] [rc5] Performance

RogerKIWI at aol.com RogerKIWI at aol.com
Mon Jan 29 12:55:11 EST 2001


In a message dated 1/29/01 11:57:03 AM, basildaoust at home.com writes:

> Yet we are all reported equally, or should I say unequally. How can 
> I expect to compete against a sysadm of a university or even a 
> corporation lan? It JUST ain't fair.

Because there is no other way to report this, and it's not a competition. :-))

I'm running this on 4 Macs that I personally own that are in our house here, 
and am running at about 11 Megakeys/sec if they are all on and not doing 
other stuff like dealing with 70 MB FileMaker Pro files. My RC5 work this 
year has been 76,610 blocks submitted in 28 days - 2736 blocks per day.

That is a lot of horsepower - largely because of the suitability of the 
PowerPC processors to run this thing - two of my Macs have 333 MHz G3 chips 
and they run about 1.1 Megakeys/sec. My G4 with dual 450 MHz processors runs 
about 4 Megakeys/sec on each processor.

On the dual G4 - Here's the chip that the Mac OS can't ordinarily use - it's 
"flat out":

[Jan 29 17:48:13 UTC] Completed RC5 packet 0857299E:70000000 (4*2^28 keys)
                      0.00:04:24.47 - [4,059,888.62 keys/sec]

Here's the chip that is running the Mac OS and AOL and Quicken and Excel as I 
write this

[Jan 29 17:49:09 UTC] Completed RC5 packet 0857299B:50000000 (4*2^28 keys)
                      0.00:05:42.15 - [3,138,216.90 keys/sec]

Here's the summary of the last 2 1/2 hours since I restarted dnetc

[Jan 29 17:49:09 UTC] Summary: 70 RC5 packets (241*2^28 keys)
                      0.02:33:13.89 - [6.90 Mkeys/s]

I have no idea how many Pentium or Sparc or Xeon chips it would take to 
perform at this rate, or who owns them, but it seems there is no "easy" way 
to make any other comparisons in the stats than overall totals.

Cheers

Roger

--
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