• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

F@^^!$^! GPU Servers!!!

M

MrBooMY

Gawd
2FA
Joined
Sep 27, 2004
Messages
991
They took a crap again, I've got 3 idling GPUs as I type this. FIX YOUR SHIZ STANFORD!!!!
 
I understand it's frustrating, but just be patient dude... sit back, breath deep and have a nice warm tamale. Wash it down with a couple of beers and things will be better before you know it...

Folding is something we all voluntarily choose to do, and getting all pissed off over some server glitches isn't going to solve anything. The huge number of GPU clients has got to be a bit overwhelming for the guys at Stanford, and they'll either catch up with the demand or they won't... it's that simple.

Just remember this is supposed to be fun, at least I think it is... so just be patient...

Fold on man, and beers and tamales to you!

 
Anytime man, anytime.

Due to my recent absence, I think I owe everyone here at least a beer or two, not to mention all the tamales.... that's going to kill me...

 
[05:28:07] + Attempting to get work packet
[05:28:07] - Connecting to assignment server
[05:28:08] - Successful: assigned to (unimportant).
[05:28:08] + News From Folding@Home: GPU folding beta
[05:28:08] Loaded queue successfully.
[05:28:08] + Could not connect to Work Server
[05:28:08] - Attempt #11 to get work failed, Please get a life, go outside and get a tan.
Waiting before retry.
 
I'm still trying to figure out why they only have 3 GPU servers. They need at least 4-5.
 
hahahah well, im happy one of my GPU work units got thru.. and i made my 100k tonight.. i think ill have one of those beers digital exhaust ;) (dont worry im old enough to drink)


 
You'll have to excuse me for my ignorance here but this is a serious question.

How many servers do they have for the standard client? How about the SMP client?

Who pays for them? Who pays for the support, the maintenance and the constant attention they need?

I don't know the answer to any of those questions (never bothered to try and find out)... So obviously, I don't know why they don't have more GPU servers.... all I know is that this is a voluntary project, nobody is making anyone fold. We should all be doing this because it's something we want to do for the right reasons, and those reasons should be finding a cure and helping the world be a better place. If we get a little notoriety along the way, great, but that really isn't the point. If your not getting any work, the odds are pretty good nobody else is either, and we all have to wait together.... right?

Now understand, I don''t like having my boxen sitting around doing nothing, but I can't really do anything about it, and neither can you or anyone else..... I just hate to see people getting so wound up over something that is completely out of their control... let it go, the stats page is back up, and I'm sure that your boxen will get work sooner than you think....

Remember why you do this... that's what's important....

 
Still, you'd think they would care enough to keep the work flowing as good as possible. It means their projects take longer when the units quit flowing. Why don't they take a couple of the classic client servers and make them GPU servers, there is NO shortage of classic servers.
 
Looks like the servers tanked with the Nikkei...

 
i just finished another unit, let see...

yup, no go here either.

Edit: At about 2:04am cdt my card started folding again.

 
digital_exhaust said:
You'll have to excuse me for my ignorance here but this is a serious question.

How many servers do they have for the standard client? How about the SMP client?

Who pays for them? Who pays for the support, the maintenance and the constant attention they need?

I don't know the answer to any of those questions (never bothered to try and find out)... So obviously, I don't know why they don't have more GPU servers.... all I know is that this is a voluntary project, nobody is making anyone fold. We should all be doing this because it's something we want to do for the right reasons, and those reasons should be finding a cure and helping the world be a better place. If we get a little notoriety along the way, great, but that really isn't the point. If your not getting any work, the odds are pretty good nobody else is either, and we all have to wait together.... right?

Now understand, I don''t like having my boxen sitting around doing nothing, but I can't really do anything about it, and neither can you or anyone else..... I just hate to see people getting so wound up over something that is completely out of their control... let it go, the stats page is back up, and I'm sure that your boxen will get work sooner than you think....

Remember why you do this... that's what's important....

Click to expand...

While I agree with you that servers hiccups happen, you are unfortunately not around long enough in the past weeks to notice it became a weekly occurence. It got worse since 2 weeks since it is now becoming almost daily.

To add insult to injury, I just learned that the new SMP 4433 unit is benchmarked much lower than A2 and even A1 units and the PG said it's benchmarked properly. Their own logic : Since they aim to keep the reference machine at 1760 PPD, the new unit took 12 hours so it gave 893 points. However, since it's much more efficient with the new core, it's normal the PPD will decrease :eek:

So the new trend is obviously to see lower and lower PPD while pushing the machine harder and harder (consuming more power by the way). What's the point to keep up big farms if they obviously don't care about those who invested a lot in it ?

 
I still feel like the Nvidia GPU units could get a little bigger personally. I don't know if some cards are having a hard time making deadlines as it is, but if the GPU work units were more like 1-2 a day for more of the cards, I think it could help the servers. But, that is just my uneducated opinion.


 
Wheresatom said:
I still feel like the Nvidia GPU units could get a little bigger personally. I don't know if some cards are having a hard time making deadlines as it is, but if the GPU work units were more like 1-2 a day for more of the cards, I think it could help the servers. But, that is just my uneducated opinion.


Click to expand...

I'd hate to think how big the upload would be. :eek:

I agree that they could be 2-3x bigger.

 
I sure wouldn't mind. This is getting old how the servers keep breaking down.
 
MrBooMY said:
I sure wouldn't mind. This is getting old how the servers keep breaking down.
Click to expand...

compy386.png


 
MrBooMY said:
Still, you'd think they would care enough to keep the work flowing as good as possible. It means their projects take longer when the units quit flowing. Why don't they take a couple of the classic client servers and make them GPU servers, there is NO shortage of classic servers.
Click to expand...

Im sure they care. Its in Stanfords best interest to keep WUs being processed. ;) Man a server goes down for a couple hours and everyone flips out.
 
I'm sure they are doing the best they can to get things running better. As long as the severs are only down for short periods of time I won't give them any hell....now if they go down for a day or more .....

I'll take a beer and some of those tamales digital.
 
Xilikon said:
While I agree with you that servers hiccups happen, you are unfortunately not around long enough in the past weeks to notice it became a weekly occurence. It got worse since 2 weeks since it is now becoming almost daily.

To add insult to injury, I just learned that the new SMP 4433 unit is benchmarked much lower than A2 and even A1 units and the PG said it's benchmarked properly. Their own logic : Since they aim to keep the reference machine at 1760 PPD, the new unit took 12 hours so it gave 893 points. However, since it's much more efficient with the new core, it's normal the PPD will decrease :eek:

So the new trend is obviously to see lower and lower PPD while pushing the machine harder and harder (consuming more power by the way). What's the point to keep up big farms if they obviously don't care about those who invested a lot in it ?

Click to expand...

All very good points and I agree 100%... I know on a financial level my contributions have been minor compared to what many of you have invested in this project. But at the same time, remember that I was one of those that rushed out and dumped a bunch of money on a couple of X19XX cards for the GPU1 client, and I am well aware of what it feels like to wind up getting "screwed over" (so to speak;)) by Stanford. Sure it pisses me off.... and I would expect anyone that has spent good money on dedicated hardware to be upset when those investments no longer produce the way they should..... And yes, the scaling is wacky, I agree. Just read through my thread about my recent production... Either someone around here is covertly folding under my name (and frankly, I have no idea why anyone would do that right now) or the answer is that my standard clients are getting fed a steady diet of those new wu's that are worth ~1300ppd or so.... it doesn't any make sense that the standard client wu's should be getting that kind of credit, but they are.... go figure...

However, the only point I was trying to make is that getting really upset over server glitches, rater they are rare or something that happens daily, is a waste of energy in my opinion. We obviously can't do anything about it.... And I know that it's been a bigger problem recently than it has been in the past, but again..... what can we do about it? Believe me, if there was something I could do, I would, just as everyone else here would. But like I said before, if the servers are down, all we can do is wait together and hope that they somehow manage to get their shit together and fix the problems that they have. That's all I was trying to say:)

Fold On guys... and beers and tamales all around!!!

 
digital_exhaust said:
..... And yes, the scaling is wacky, I agree. Just read through my thread about my recent production... Either someone around here is covertly folding under my name (and frankly, I have no idea why anyone would do that right now) or the answer is that my standard clients are getting fed a steady diet of those new wu's that are worth ~1300ppd or so.... it doesn't any make sense that the standard client wu's should be getting that kind of credit, but they are.... go figure...
Click to expand...
I haven't received any standard units worth that much. Which WUs are those?? :confused:
 
First, I screwed up... the wu's are apparently worth around 1300 points (not 1300ppd) but they're taking 3 to 5 days to complete. I don't know the unit numbers, and honestly, I don't even know if I'm getting any. I was talking with Evil and he speculated that those units may explain my odd production.

Sorry for being vague. I can't monitor my standard clients as they are all borged boxen.... so I can't help with the wu numbers....

 
You must log in or register to reply here.
Back
Top