Interesting find

DroolMAN!

Senior member
Mar 4, 2000
462
0
0
I experimented running more than one distributed computing clients on several of my Ultra Ninja Bastiches and came up with some pretty interesting results. The experiments involved SETI@Home CLI, d.net and ECCp-109 (DOS and GUI based) clients.

First off, d.net client (469) has the lowest priority followed by ECCp-109 DOS client then ECCp-109 GUI and finally S@H CLI.

I ran three clients, d.net, ECCp-109 DOS and S@H and found that S@H hogs all the idle cycles. But if you ran the ECCp-109 GUI-based instead, ECCp-109 will run at half speed but S@H still completes within the average time!!! Unfortunately, d.net doesn't even get to crunch anything.

In other words, if you have S@H running on your system, it is highly recommended that you also run the ECCp-109 GUI client since S@H is not affected while you are still producing distinguished points for ECCp-109 and help affect TA's overall results.

Just sharing my findings.

[EDIT]

PS. One word of warning though, neither the DOS or GUI ECCp-109 clients transfer the distinguished points found over the proxy.

What I did was stop the network clients and moved it to a pc with a modem and do a transfer then move it back and re-run the client. You might just fumble and accidently delete the client (like I did today with my longest running client), so be warned.

This problem is acknowledged by the creator but no fix has been done thus far.

[/EDIT]
 

PieDerro

Senior member
Apr 19, 2000
813
0
0
I've actually found (although I wasn't really looking) that when I run S@H and DNET RC5 on my system together (Athlon 1350), I get about 8 hours for SETI and 1.2MKeys for RC5, this is with me using the comp for about 4 hours a day (nothing heavy, just browsing etc).

BTW, the SETI that I used was 3.03 GUI (I haven't been running SETI for a while now, wating for RC5 to hurry up and finish so I can be there when it does!! :D