Discussion in 'World Community Grid (WCG)' started by Norton, Sep 4, 2013.
There's no GPU work ATM
I haven't gotten any errors yet...none of them have validated yet either lol
I'm running 6 wu of CEP2 now, CPU and VRMs seem to be running about 5-6c cooler than when I left this morning, Idk why but I'll take it lol. I'm at 4.5ghz with 1.33v and 8gb memory at 2400. I would like to try like 4.8, which I know I could do but it would need about 1.5v and the VRMs would be close to 60c which I dont like for running days on end.
Seems relatively stable, lol. Haven't stopped crunching to stress it since I set this memory clock. The new kingstons I got from the RMA seem to be working beautifully
Does anyone here have the BOINC screensaver?
Since I am running all CEP2's I get a screen saver that has a CEP2 logo of solar paneled leaves w/ wires. The question I have is in regards to the landscaped, dynamic, overlay.
What is it measuring? Is it related to the core working the WU displayed? If so what do the peaks and valleys mean?
I'm not often this puzzled and I hope someone here can tell me what it is. I'm mildly irritated that I can't figure it out.
I think we should post that on the WCG forums because I am curious too and am not familiar with the answer. BTW, I am not running the screen saver but I get the same stuff when choosing for "show graphics".
OK, posted on WCG, in the CEP2 forum.
Hopefully someone can get the answer, there, and I will post it here for you (and the rest of you that are similarly puzzled by this).
3 concurrent CEP2 WUs seems to be the magic number for me (FX8350@4Ghz, 4GB RAM, Linux). I'll up one of the crunchers to 4 and see how it flies.
@Arjai: Thanks mate. i also subscribed to the thread there so I willl keep an eye on it.
I just checked my WCG account. I had a bunch of "valid" and "pending verification" results, but I also had a few errors. I'll drop my OC 100Mhz and see if that helps. I guess it explains a little my low output.
It certainly is disappointing :shadedshu
Just checked my other FX-8350 rig- it's rocking CEP2's on all 8 cores atm
* memory usage is at 31% of the 8GB in the rig but it may go up as the wu's progress...
Another BSOD while doing nothing Dropped my OC a little, see if that changes things. Stupid CEP2 :shadedshu
BSOD's suck!!! :shadedshu
If they continue then consider going to All projects but increase the # of CEP2 jobs you can hold in your cache to 8 or 16 (recommend 8 first)
Ill have to see what happens. I got the BSOD when I wasn't even crunching; I had paused to do something else. Not sure why but I'll get to the bottom of this.
Screen Saver Solved!!
With a little help from SekeRob, did a Boolean Search on WCG, I have the goods on that overlay!!
Are you sure?
There it is. Never would have guessed THAT!!
im pretty proud of my rig, its been crunching at 4.8ghz for nearly 50 hours straight and kept around 50*c
and zero errors
That's awesome man. I have mine at 4.4. I don't want to push it any further until I get it under water and better mosfet cooling.
Also your just about fully spooled up on the 8350 rig.
Ive screwed an AMD fan onto the vrms, theyre pretty cool tbh!
what do you mean by spooled up?
With a new crunching rig/profile it takes anywhere from a couple days to a couple weeks for it to reach full capacity. You will see results right away but if you have noticed you totals for that rig has gotten bigger each day it has run. You basically have a few valids then a few go into pendings. Once a certain amount of days goes by the pending pile will level off with what is going into new pendings and what is getting validated out of pendings. That and the rig becomes "trusted" by wcg. I had one rig that ran for awhile and was posting great numbers then the boinc client went nuts. I lost all of the wu's that rig had. I had to reinstall boinc from scratch. The lost wu's error-ed out or was labeled detached. Wcg black balled that rig for almost 2 weeks. Every wu that rig sent in went to pending verification. They all had to be verified before the could be validated. Sucks because it took almost 3 weeks before that rig got back to putting out great numbers again. Oh the rig in question here is my i7 920 rig.
Thanks, im actually learning quite a lot about crunching. I have about 4 pages of results pending validation, I noticed it takes some time for them to get validated. Also are there different amounts of points for different projects? The clean energy take a lot longer than fightingaids@home.
I told my friend who is a nurse about all this and she was like WOW, and didn't know all this research was happening behind the scenes. She was pretty motivated about the cancer projects too. It was fairly easy to explain the concept to her, she even asked if any computer could do this.
Any computer can do it. As you have noticed the faster the rig the more work it can do. When I started wcg almost 2 yrs ago I was running it on a Athlon xp rig and 2x S939 rigs. My ppd was really low compared to what most are doing today. I might have hit 1 to 1.5k ppd. Also when the gpu unit came around full bore, we went through them like a hot knife through butter. Which in turned help to finish the Help Cure Cancer project way early. I'm talking years early. Team Tpu has a power house full of abled Gpu crunchers. And I know for a fact the some of us have stowed away or gpu equipment for the return of the gpu units. I know I have actually picked up some more stuff for my arsenal.
I replaced the 6670 in my HTPC with a 7870 for this reason. A couple of the office crunchers have 7770s installed too.
I just smelt something hot and melting, touched the psu on my 4100 rig (some 750w ezcool thing)
and well the metal case burnt my hand! im going to try rotating it and see if it helps, 2nd psu ive probably killed folding/crunching lol
I killed one at the beginning of the gpu wu's with a GTX280. 500w and 18a on 12v rail just wasn't enough to feed the hair dryer.
this thing is old and crap, the fan may have died tbh! the 4100 was throttling because of the mobo so PPD shouldn't be that much off.
im glad I was next to it and noticed, could been a lot worse if I wasn't here!
the last one spat a flame out the back and went BANG! it shit me up lol
So true man. When I fried mine I wasn't in the room but happen to walk by at the right time I guess. I got a wiff of the burning plastic smell and ran in to see the rig shut down. I unplugged the power cable from the back and let it sit. After an hour or so I fired the rig back up and it booted fine. I shut it back down and pulled the psu from my sli'd 8800 GTX rig and used it for a few days then bought an new psu for the cruncher.
Oh by the way mine was a pos Logisys 500 or 550 watt unit. Thing worked great for over a year in that cruncher until I jammed the 280 in there. That was all she wrote. Fat lady sung on that psu.
My Socket A, that I still have and will be updating, went up one night, a few weeks after having it built.
I was sleeping and heard a series of loud pops! Flip on the light to see a stream of smoke pouring out of the PSU hole on the back.
Well, I suppose most of you can guess what the pops were. It was a mess!
Fortunately, HDD was still good and the local shop replaced the MB and CPU and PSU, no charge.
It has run like a top since then. Replaced the Duron w/ a 2600+ OC'd it and crunched with it.
That's the only PSU I have had fail. I have repaired a bunch of computers, whose only problem was a bad PSU. Yet, none of mine have failed. Aside from that one that went boom in the night.
Separate names with a comma.