The Folding@Home Pub

LOL, I passed you, MoosePower !! But looks like you'll pass me soon.

 

Hey - I just realised something "Disease" is spelt incorrectly in the forum descriptor -

"Please join the Folding@Home Research Program, join team Guru3D. Let's rank the charts high and fight desease !!"

Lol - only took me nearly 10 months to spot that

 

Folding@Home Items of Interest ...

Hello.


First off, I am cutting back on the ammount of machines I have folding. I was running about eight but will only be using two now. Six of them were 700 - 1100MHz Celerons that took almost a month to submit a WU anyway. I will still be folding with my Athlon 64 2.4GHz and Pentium 4 3.0GHz HT Two Instances.


I think I might have figured something out. For myself anyway. I hadn not done hardly anything but torrents and surf the web for about a month on both of my machines. I was getting a lot of Double Gromacs SSE2 Work Units during that time. Then, I started back playing F.E.A.R. Perseus Mandate and the original F.E.A.R. Multiplayer, and I am getting the regular Gromacs SSE Work Units now. Do Stanford's Servers decide what Work Unit to give your Folding@Home Client based on the ammount of CPU Cycles the client gets on average? Because, that is kind of what it seems to be.


I am looking forward to getting $975 for something I am selling so if I do I am buying a Q6600 ! And you can bet it will be folding 24/7.


Asterx - Congratulations on making it back up to number one for Team Guru3D !! Lets see how long you can hold it this time..

Good job to everybody! We are now at 185 and climbing daily.

Thanks!


- XeonPower

 

Sorry to hear that - guess they must be sucking up the juice tho' eh?

Can't say for sure but my 'legion' of P4s (about a dozen) just runs folding 24/7 and you can see them work though bouts of just having one type of work-unit or the other, so I think it's more directed by which project they (vijay & co.) want to push ahead on

Hey thanks! Pity D.C. / Bahamut had to drop-out like that... might try and PM him to see if he can get his servers back and folding...

 

Well, this is strange.

This is the order of WUs I have on my P4 and Athlon 64 ...

2451
2452
2453

ROFL, and I am not joking.

Unfortunately, all of them suck.

 

This is even more strange. I folded a bit on my 4200+, and im just starting to fold again on my E6550. I have two instances of the no nonsense version for 100% utilization on both cores. Using FahMon, one of the client's final deadline is march 3rd, downloaded today. My other client downloaded a gromacs core from december with a final deadline of January 23rd. Should i stop it and delete the files so it downloads a new one?

 

Weird - did it just newly get downloaded ? If not it's probably been AWOL for a while and I would remove it if I was you...

 

Tomtomz - I would delete that Folding@Home directory, create a new one, and start over. It's probably something in your quene.

Aircool - You're right behind Gaboou ! You might take him.

 

:flip2:

 

Ahaha....


Gaboou AirCool

 

Just a warning guys, my PPD is going to come crashing down over the next day or so because of the need to move to the new SMP client...

If you have anything extra to give now would be the time to do so...

DUH DUH DUHHH!

 

Heh - All in good time young padawan

Oh hey - Just thought I'd better let you guys know, I've decided to "spice" things up a bit, you know make things more interesting etc...

http://forums.hexus.net/news/129696-join-hexus-folding-home-team-2.html

 

Thanks but you unfortunately misplaced the names, hehe.

 

Problem - big ol' big assed problem.
I've started updating those clients I added during operation 'White Cycle' and I've discovered why despite having twice as many clients as I did over Christmas I'm getting just under half the points.

Basically only one in 8 WUs is running to completion. That's a phenomenal 87.5% FAILURE rate. Because of the negative impact this has on the project as a whole (i.e. at the moment I'm doing more harm than good) I've decided that these clients will NOT be invoked until such time as the cause has been identified and a workaround / fix put in place.

I have an idea about what may be causing the problem and will be implementing a work-around on a testbed system shortly.

In the mean time if any of you have experienced client-core communications error 7b (123) then let me know (especially if you have a fix for it )