Posted on 01/07/2007 11:28:30 AM PST by texas booster
The command prompt is due to a problem running as a service. It shuts down immediately. I'm hoping the next version due out soon will fix this. Current beta expires on Feb 1 if I'm not mistaken.
JosephW
Congrats on making it into the top 50 Freepers that Fold.
Haven't heard from you in awhile so I hope you're not suffering with too much snow,
Joseph
The PlayStation 3 FAH client has been delayed by Sony. Everyone is under NDA and won't say anything in a public forum.
Speculation by the unwashed masses is that Sony is planning a major firmware update to coincide with the European launch of the PS3. So that places the upgrade in late Feb or early March.
The PS3 betas should roll out in April or May according to rank uninformed speculation.
If anyone knows how to reach the new folder FormeDa, please have him ping or FReepmail us.
He is having some problem with EUEs and is losing a lot of points.
We can help. We can rebuild him better than before.
We can help him borg other systems.
For less than $6 million.
There is a problem being reported that ATI X19xx cards with 256 MB of RAM are having lock ups. It seems to be more of a PCIe issue (vs AGP), and seems to be related to chipset/driver issues.
I was reading a few forums and found a nice post that may help you fine tune the card:
Returned 158.42 if that helps any. I suspect it is a combination of ATI's 6.11 and the current 5.91 beta 2 that is keeping things from where they should be. Wouldn't be surprised if the next beta due out Feb 1st (unless I've read it wrong) from Stanford and ATI's 7.2 drivers don't take care of this perfectly.
Tried to restart the system so that the GPU version would run as a service and got this as part of the log file:
Entering M.D.
[03:11:19] Will resume from checkpoint file
[03:11:22] mdrun_gpu returned 99
[03:11:22] GPU initialization failed
[03:11:22]
[03:11:22] Folding@home Core Shutdown: INTERRUPTED
[03:11:25] CoreStatus = 66 (102)
[03:11:25] + Shutdown requested by user. Exiting.
Folding@Home Client Shutdown.
I then proceeded to start it in a command prompt without incident
The problem is that the GPU requires that the GUI be run as the owner of the process. If you start the GPU as a service it dies when the GPU context switches to your user desktop.
Found an entire thread dedicated to folding GPU as a service, with autostart capability.
http://forum.folding-community.org/ftopic17399.html
Since I don't own an X1900 (yet) I have to translate a lot from my right brain to my left brain. Hopefully this link will help you get past the problem.
There's some good stuff there and obvious things that I missed.
I'll set the service log on to be me, but go into the registry and set the start-up option to be after log-on (Start-up type 3 I believe) in the current control set.
However, I suspect the next client version may take care of most of the issues, and as long as mine doesn't lock up I'll probably leave it be (unless I hear it calling me...)
... The folds... the folds... you must pass Egon while you have the chance... the folds...
Nope don't hear anything yet :)
Hah! Won't be long now-- I'm down to one Mac, and that one is gone next week.
[05:51:37] Completed 47000 out of 50000 steps (94) [05:54:01] Quit 101 - Fatal error: Box exploding. [05:54:01] [05:54:01] Simulation instability has been encountered. The run has entered a [05:54:01] state from which no further progress can be made. [05:54:01] This may be the correct result of the simulation, however if you [05:54:01] often see other project units terminating early like this [05:54:01] too, you may wish to check the stability of your computer (issues [05:54:01] such as high temperature, overclocking, etc.). [05:54:01] Going to send back what have done. [05:54:01] logfile size: 9183 [05:54:01] - Writing 9736 bytes of core data to disk... [05:54:01] ... Done. [05:54:01] [05:54:01] Folding@home Core Shutdown: EARLY_UNIT_END [05:54:03] CoreStatus = 72 (114) [05:54:03] Sending work to server
I was getting the instability message this summer (like 5 or 6 times in a week) on one of my machines when it got too hot. When I limited the folding to 80% of the CPU time, the messages went away.
Given that this is a one time message and in winter, I doubt this is your problem, but you may want to look at the CPU temperature if there's a chance of overheating.
That's a good tip, regardless, but my office is nicely climate-controlled and the PC is in not enclosed, PLUS the thing has fan speed control built-in. If it was heating up, it'd sound like a 747 revving up for take-off.
My HOME unit, however, is an entirely different story. I've got one of those pieces of gee-whiz computer furniture that closes up to look like an over-under cabinet. The PC actually occupies a shelf built into the inside of the bottom, right-hand door. When the cabinet is closed, the whole under-desk volume is enclosed, except for the rectangular cable opening top, center at the back, and a ~1 inch gap all the way around at the floor. Left to convection alone, it doesn't even come close to the cooling requirements of my 2.26GHz P4 folding at 100% CPU utilization.
After several overheating incedents last July that stopped the CPU, I took the machine down in August for some corrective action. I disassembled, cleaned and remounted the heatsinks with fresh Antec "Formula5" thermal compund, switched the flow direction of one of the case fans, and installed a 6" 120Vac fan in the back of the cabinet to pull negative pressure on the enclosure, drawing cool air in from around the floor level.
After those modifications, I was able to run without any overheating incidents at all.
Sounds like you did the cooling thing the right way. I just took the cover off, blew out the dust and left the cover off. I do have a bigger fan now ready for next summer.
The p3302 protein has been EUEing for a couple of months now. Apparently the initial parameters give too wide a margin for the simulation to run into never-never land.
Many p3302 work seems to finish OK, but maybe half are abending. Looking at Egon's monitor (FReepmail him if you want the link) it seems that Chris_Primavera and HangThemHigh have at least 75 EUE's between them this month on this protein.
In most cases you get partial credit for work completed. Once Stanford has identified the bad parameters then the work units will be sent out again.
I just got through reading up on these EUE's over in the FAH forum. These are from my office box since Tuesday:
Project: 3302 (Run 5808, Clone 1, Gen 1)
Quit 101 at step 18500 out of 50000 (37%)
Project: 3302 (Run 3543, Clone 1, Gen 1)
Quit 101 at step 8500 out of 50000 (17%)
Project: 3302 (Run 4994, Clone 3, Gen 1)
Quit 101 at step 19500 out of 50000 (39%)
Project: 3302 (Run 2398, Clone 0, Gen 1)
Quit 101 at step 5500 out of 50000 (11%)
Project: 3302 (Run 39, Clone 3, Gen 1)
Quit 101 at step 47000 out of 50000 (94%)
Bump
Finally made #10 in FR Folding :)
Chris Primavera, I'm catching up to you. Only 14,300+ days until I overtake you (although yesterday was a bit of an anomoly and has probably has me at a rate higher than I am).
JosephW
I am 18th top producer of points over the last week.
Source : The "points week" column on the stats page.
Not bad for one computer!
You rock!
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.