Join the SatelliteGuys Folding@Home Team!

[GeorgeLV]

There is a patch available to (hopefully) fix the Windows 6.22 SMP client.

Folding Forum • View topic - new 6.22beta2r2 client [fix terminate on EUE] [not ACK fix]

 

[jgantert]

I'm still waiting. My 5.91 is running like a champ. If I start running out of WUs then I'll upgrade.

 

[Scorpion.sting]

I think I am going to go with this PSU for my Gateway GT5428 so I can add another 8800GS to my farm.

Let me know what you think.



Newegg.com - CORSAIR CMPSU-550VX 550W ATX12V V2.2 Power Supply 90 - 264 V UL, CUL, CE, CB, FCC Class B, TUV, CCC, C-tick - Power Supplies

 

[GeorgeLV]

I think I am going to go with this PSU for my Gateway GT5428 so I can add another 8800GS to my farm.

Let me know what you think.



Newegg.com - CORSAIR CMPSU-550VX 550W ATX12V V2.2 Power Supply 90 - 264 V UL, CUL, CE, CB, FCC Class B, TUV, CCC, C-tick - Power Supplies

Kind of pricey for a PSU with only one PCI Express connector.

 

[Scorpion.sting]

Kind of pricey for a PSU with only one PCI Express connector.

Thats all I need is 1. I dont run SLI. Besides, I spent $100 for a PSU made by PC Power and Cooling. I dont mind paying more for a quality PSU. This one has a 5 year warranty.


Fold on :up

 

[Scorpion.sting]

Kind of pricey for a PSU with only one PCI Express connector.

George

Looked back at the specs and the PSU has 1 6 pin and 1 6+2 pin.

 

[Madtown HD Junkie]

George

Looked back at the specs and the PSU has 1 6 pin and 1 6+2 pin.

dude, you running a Cray super computer?

 

[Foxbat]

By the way, the campus at Stanford suffered a massive power failure today:

August 18, 2008
Here's the update from Stanford

Power mostly restored to campus
Updated 3 p.m pacific time. - Power has been restored to the majority of the Stanford campus. All normal power is expected to be restored by 5 p.m. or sooner. A major outage occurred today at 11:30 a.m., affecting P G & E customers in Stanford, Menlo Park, Atherton and Palo Alto. If you are still experiencing difficulties on the Stanford campus, call 723-2281. At Stanford Hospital, dial 286.


This one was a major disaster at Stanford and Palo Alto (and nearby cities), probably the biggest outage in a while. However, there seems to be some major Stanford power outage once a year, which is a major problem.

With this in mind, we have been distributing more of FAH to outside of Stanford (with servers at UCSF, Columbia, Cal State U Long Beach, and U. Pittsburgh). We hope to have a European site soon. Once those sites are a bit more established, we'll see about pushing an assignment server to a non-Stanford site and we should be much more safe to Stanford-related issues.

Also, it's good that we have servers in 4 different server rooms on campus. One stayed up the whole time, two came up fast, and the fourth (VSPGxx) is coming up now. Some servers will be slow to come up, so we expect this may take at least a few hours.

The stats update has been turned off until this gets sorted out. We hope to turn it back on later tonight, but it may have to wait until tomorrow morning.

I wondered what was happening. EOC was showing big holes in our Team's production, but if the servers aren't there, you can't upload results and you can't get new Work to Fold.

 

[SatinKzo]

Still fighting my IBM server, wow, I have never had a system need so much repairs out of the box.... Looks like I got a lemon as they are just sending me a new one today.

On another note, I got one of my SMP clients fixed yesterday that stopped working a few days ago and I never got a chance to fix.

I might of also scored a couple 8800gs cards, but that is not a given yet.

 

[Scorpion.sting]

dude, you running a Cray super computer?

The video card needs a 400 watt minimum PSU.

 

[Foxbat]

The video card needs a 400 watt minimum PSU.

Good thing I got a 425W then, isn't it?

Here are the first impressions of the Points Enhancement to my HP d325 2.33GHz Athlon XP+ minitower: Before the new ATI HD 3850 card: 5.5-11 pph. Some of the WUs took as little as 8 hours to complete, but the first one took nearly two days (!).

I've loaded the console version of the GPU2 (no pretty pictures (yet)) and it's crunching at 100% CPU and about 2m32s per percent of WU 4731 (193 points). InCrease reports a pph of 43.47 so far (it's creeping up slightly) which yields a little over 1000 ppd. Now I need to send in the $45 worth of rebates from NewEgg and I've given new Folding life to the old HP.

Now, granted, I could have ripped out the AGP Mobo and replaced it with something with multiple PCI-x16 slots, but I'd be better off forgetting all about the HP case and going with a whole new box. I should be spending close to $140 or so for this 4x increase in points.

Since I've been writing this, the GPU2 has finished another 7% (11% now) and the pph has gone up to 44.3. My 1.83 GHz C2D Mac Mini running the SMP client generates 54.7 pph and cost $499. I'd say I'm okay with that. Of course, the fan noise will drive me crazy, so i think it's going into the basement as soon as possible!

I'll probably end up downloading the Systray version of the GPU at some point because I'd like to see the GPU2 viewer application.

 

[Foxbat]

I'm sticking with the console version installed as a service. My poor 2.33 GHz Athlon XP+ does not have enough CPU power to run the Viewer and Fold at the same time. It will run for about five seconds, but the animation slows down until the protein stops "vibrating" and the progress indicator stops incrementing.

I played with the Catalyst Overclocking CP, but again, like George had warned, a single CPU is not going to be able to keep the GPU fed. No matter what settings I try, the GPU2 client is completing one percent every 2:32. With the default clock settings, the HD3850 is running at 58-62%, so I probably should have saved money with an HD 3650 card instead of the 3850. It has been a solid Folder, though, averaging about 45pph since I fired it up.

 

[SatinKzo]

Geesh, the SMP clients on the Xeon processors are a bitch.
I have 2 Xeon quad core 3.16ghz processors in my new server, Trying to run SMP clients for burnin testing and damn if I can't get 6.22 to run on both processors. I can run it on 1 at a time. Ended up running 5.92 on the other processor.

Anyways, system is doing a % about every 11 minutes on each processor.

 

[Foxbat]

SatinKzo, are you running with the new Denio interprocess communications package or the older MPI? Firewalls at work prevent me from playing with this effectively, so I don't have any experience to draw from.

I know the mpiexec subsystem that they use on *nix and the Mac uses the default NIC to ties the SMP cores together, so if your network flakes out, *poof*, there goes your WU. If I tried to run multiple SMP clients on my Mac, they would interfere with each other. I wonder if that is still an issue trying to run two SMP instances under Windows?

Still, the Team appreciates the additional points! Here in the Midwest, we've been treated to some cool, dry weather, so open the windows, and Fold Speed Ahead! Hopefully, some of our other Team members that shut down for the summer will be back soon.

 

[SatinKzo]

for 6.22, I'm using the Deino and for the 5.92 I'm using MPI. It was odd what was going on with the 2 quad cores and maybe it had to do with both trying to use Deino, but after I got the full install for 5.92 and got it going, both have been happily goings since 7am this morning. Not saying it's full proof, but hell, both were bombing out after the first % of the WU when I have both trying to do 6.22 with Deino.

yeah, I have been tempted to let the PS3 fold some more since it's been so cool at night and comfortable in the day. But I'll hold off until the fall and fire back up the PS3, wife's dual core and hopefully a couple 8800's (if I do infact get them). All of those are in the basement office and HT room so I can effectively shut all heat vents in the basement and it stays very comfy down there in the winter.

 

[Scorpion.sting]

What happen to George and all the folding power he was using? Hope the card didnt go out.

I have been folding with all my cards since it has been cool and I havent had to turn on the A/C. If I do at some point have to turn on the A/C, I will shut off one of the cards to offset the electricity costs. For now all is full steam ahead.

Foxbat - glad to hear you got your 3850 going. You really cant go wrong with the video card client, The points are amazing.

SatinKzo - That would be awesome if you can get the two 8800GS's going. That would be a good 50k a week added to your stats and the team.


Foldon guys:up

 

[GeorgeLV]

What happen to George and all the folding power he was using? Hope the card didnt go out.

Nope. No hardware problems, it was a planned outage. I'm back at full production.

I had to go out of town for a few days right when the Stanford server outage started. Since I didn't have any WUs going I decided just to shut everything down until I got back.

 

[Foxbat]

What happen to George and all the folding power he was using? Hope the card didnt go out.

Nope. No hardware problems, it was a planned outage. I'm back at full production.

Glad you're back with us, George! I get depressed looking at the Team's daily production being hovering around 50,000 ppd. Seems like a month ago, we were up around 70K ppd.

I want to know when Plazzman will be putting his Folding Farm back on-line. He stopped about two-three days short of 1,000,000 points due to rising temperatures.

Sorry guy's due to High Heat , Electrical Bill's , on and on. I will be stopping all my folding.

 

[Foxbat]

Folks over on the Folding Forum are reporting new WUs for the ATI boards. These WUs (P4742-4747) have more atoms in them and evidently do a better job at keeping the GPU's shaders fed, reducing the load on the CPU. My rig picked one up this morning and it looks like a winner! Each percent is being completed in 4m06s compared to the normal 2m32s I was seeing on the smaller simulation. So, instead of taking a little over 4 hours per WU, it's now closer to 7 hours, but the points value of the WU goes from 192 to 548. As a result, this old rig with the new ATI card is now cranking out over 1,900ppd!

 

[Foxbat]

Grrr!!!™ My AT&T DSL knocked out again at 4:35 this morning. Both my GPU2 and Mac Pro clients were twiddling their silicon for five hours or so until I re-established the connection. When I did get things connected again, my SMP client on the Mac downloaded one of the older SMP projects that uses the old core (about 60% CPU utilization vs. 99% with the new core). So, today might not be my best day, but at least I grabbed another 474x Project with the GPU2 client. Since the HD 3850 was at 99% GPU usage, I went ahead and slid the sliders for GPU Clock and memory up to max to see how it holds up over the WU. The GPU is still at 99%, so it appears that it's holding things up instead of the CPU (which is still at 100%, but half of that is Kernel mode). This has cut the time to complete one percent from 4m06s to 3m50s, and InCrease reports production at 85.197pph, yielding 2,045ppd! (well, call it "over 2000ppd" owing to upload and download delays).

I guess I need one of the OC folks to tell me if my ATI HD 3850 is going to fail at 82?C for extended periods of time. I'll probably let the CCC Advanced CP do the autotune for a sweet-spot. Or, get more fans for more airflow.