Discuss All Folders: Please Read! in the Dev Folding forum on Dev Hardware. All Folders: Please Read! Dev Folding forum for discussing Dev Hardware’s folding@home team. The Dev Folding team contributes spare processor cycles to Stanford's research team, helping to find cures for disease. Join us to help science, help medicine, and help our team.
Time spent in forums: 1 Month 2 Weeks 18 h 57 m 19 sec
Reputation Power: 13055
All Folders: Please Read!
Attention all folders. I would like to conduct a series of tests to see how much folding impacts system performance. It's a rather neglible number, I just want to prove it to those people who need numbers to back it up.
I would like as many of you as possible to participate in this so that we can have as complete and accurate results as possible. I'm coming up with the testing criteria as we speak. In the meantime, I'd like to know how many of you are willing to participate.
Here's an example of what I'm talking about. This test is outdated and I'd like to test it with the latest cores and projects.
Let me know if you're interested in participating. Thanks.
Location: Romania - Târgu Mureş - the one and only
Time spent in forums: 3 Years 5 Months 1 Week 5 Days 10 h 23 m 3 sec
Reputation Power: 17902
Of course I'm in, since I suggested the idea- but it's cool, hope it'll be a nice project.
Nevermind, I'll do that this weekend, because the way I want it, it's "too complex" (you'll see it) - and will take time.
I'd think about some basic benchmarking softwares like 3DMark01SE (don't laugh, it's better for overall system perf than 3DMark03/05/06, because it also depends a lot on processor power, so for us it's better), SuperPI 1MB (the 1MB test barely touches the RAM so it's okay, it'll just use L2 cache and pure CPU processing power) / 32MB (it uses lot's of RAM and technically everything), SiSoft Sandra Lite 2007 -freeware-(I know it sucks, and that it's synthetic, but if you use at your system both tests, it'd give us an idea with those charts), PCMARK2004 (it's okay for every system). Few additions like Winstone should be also okay or tests like Aquamark.
If you're running on Linux, no problems- your scores will be also more than okay for a higher varity, when comparing scores.
Come up with ideas, write down numbers, make screenies, make charts, whatever you've got time.
You can count on me, I'll do this- very detailated, hope a few of us will also join. Have fun.
“Greatness, combined with the hint of a 24-carat lifestyle, is within us striving to make it through, craving for attention,
and in the end, recognition, leaving a record worthy of biography on forging success stories.” —ME Check out MadHyeNa's Article Index. Visit PrimeRanks - We Rank You Into Pole-Position! I succeed, conquer, and achieve, therefore I am.™
Last edited by madhyena : August 10th, 2006 at 12:42 AM.
Reason: guess ?!
Time spent in forums: 2 Weeks 2 Days 22 h 15 m 41 sec
Reputation Power: 2739
I'm up for it.....
Why not, I'm always up for trying out something new.
Its probably a good idea to get as many different rigs on this project so that the results can be compared.
Time to put my Athlon XP 2800+ through some hell!!
CURRENT MAIN RIG:
[ Opteron 165 @ 2.5 GHz ][ 2GB OCZ Platinum EL ][ DFI NF4 SLI-DR Expert ][ BFG 9600GT OC ][ 1.14 TB of space ][ Gigabyte 3D Aurora 570 ][ Hiper Type R 580w ][ Alphacool VFD ][ LG 22" 22SL4R + LG 19" 1952S ] CURRENT SERVER RIG:
[ Opteron 165 ][ 1GB Corsair XMS ][ 300 GB of space ][ Matrix Orbital VFD ][ Gigabyte 3D Aurora ][ Hiper Type R 580w ] CURRENT LAPPY:
[ Core 2 Duo T5500 ][ 1GB DDR2 ][ 160GB HD ]
Time spent in forums: 2 Months 2 Weeks 4 Days 14 h 41 m 8 sec
Reputation Power: 17750
Ya I agree with paradox, the more different rigs and cpu's we can get to benchmark the better the results will be. We will also have a mix of oc'ed and stock cpu's which may or may not make a difference.
Even if this just convinces a few people to fold, it will help out immensely.