Folding@Home Weekly Team News - 25th January 2007

SiriusB said:
SMP is deffo the way to go I think! We were 13th overall in production not long ago and it has slipped due to the craptacular WUs in circulation. Since this is a computer enthusiast forum I reckon we have the edge over most other teams as people here are more likely to give Linux a bash [bwahaha - BTI will probably be the only one to get that :p].
Click to expand...

Thats terrible :p
 
Technically, my pitiful output this week is entirely from my laptop and should allow me entry to the single cruncher league but I'm so ****ed off, I'm not bothered.

I will get home tomorrow and should find out what's happened. All the rigs at home appear to have stopped at the same time, so it's either a catastrophic power failure or the net connection has gone down (or somebody unplugged the home hub).

:(

Stan :)
 
This board sucks for overclocking :mad:

I only managed to get another 200Mhz(or 800 on all 4 cores) Still it brought my WU down by about 1 Hr(3029 WU). Still no match for Cobs beast.
 
I had a pretty poor week to be honest, I've had all kinds of errors come up on my main rig saying something like this:

[14:49:39] Folding@home Core Shutdown: INTERRUPTED
[14:49:48] CoreStatus = 75 (117)
[14:49:48] Error opening or reading from a file.
[14:49:48] Deleting current work unit & continuing...
[14:49:52] - Preparing to get new work unit...
[14:49:52] + Attempting to get work packet
[14:49:52] - Connecting to assignment server
[14:49:53] - Successful: assigned to (171.64.122.128).
[14:49:53] + News From Folding@Home: Welcome to Folding@Home
[14:49:53] Loaded queue successfully.
[14:51:20] + Closed connections
Click to expand...

Then I found someone had turned off a couple of my other rigs.. GRR!

Still, should be back to full operating capacity now. :)
 
Bigstan said:
I will get home tomorrow and should find out what's happened. All the rigs at home appear to have stopped at the same time, so it's either a catastrophic power failure or the net connection has gone down (or somebody unplugged the home hub).

:(

Stan :)
Click to expand...
we did have some pretty nasty wind last week so that could have done it - depends what would have come back on after a power cut and if everything connection wise had come back on nicely :o

hope it's nothing serious anyway - and can definately understand why you're so hacked off with it :(
 
Bigstan said:
Technically, my pitiful output this week is entirely from my laptop and should allow me entry to the single cruncher league but I'm so ****ed off, I'm not bothered.

I will get home tomorrow and should find out what's happened. All the rigs at home appear to have stopped at the same time, so it's either a catastrophic power failure or the net connection has gone down (or somebody unplugged the home hub).

:(

Stan :)
Click to expand...

I feel sorry for you mate, after all the time, effort and money you have invested in your farm.
I do know how it feels, and it certainly makes you reconsider wether it is all worth it.
I just soldier on for it is a worthy cause.
Running the SMP client would certainly help your situation as when you are at home your points would almost certainly double.
A single C2D at the speed yours are left running whilst you are offshore would net you an average of 1800 to 2000 ppd.
I know linux isnt everyones cup-of-tea but you would certainly get plenty of help here in setting everything up

Trying to help, but I know just how seriously ****ed off you must feel.
Tom.
 
Cob. that is amazing output from a C2Duo. :eek:
It has to be an E6600 or higher, what speed are you running her at?

My E6600 is at 3.2GHz, I quickly set up the system and havent had time to try for any higher.
Im using an Artic Freezer Pro on it and the temps in Windows running two instances of F@H were around the 45c.
When I have more time Ill be changing it over to water cooling and would guess 3.4 to 3.6GHz would be no problem.
 
Thanks Cob,
just noticed its in your sig, doh!
Its been a long day. :p


Is it more stable using the 8 multi, or is it for better bandwidth?
 
Last edited:
Thanks again for the stats, pitiful output :(

Seem to have got some new types of wu's and my server is back home again to munch through some wu's :cool:

Still scared of linux so smp is out of the question for me :)
 
Just noticed on the summery page that the 24** Ribo WUs are down to 500 points. I wonder if they will complete faster.

I think Ive got that wrong they might have been 496 points before.
 
rich99million said:
You should give Ubuntu a try using VMware - even I could work it out so it can't be that scary :p
Click to expand...

Ive not tried the VMware / Ubuntu / SMP, mix.
Can you still use Windows whilst Ubuntu is running in VMware?
 
sculptor said:
Ive not tried the VMware / Ubuntu / SMP, mix.
Can you still use Windows whilst Ubuntu is running in VMware?
Click to expand...
yep it works fine - though with SMP running in Ubuntu under VMware and a GPU client running in windows (as VMware seems to be limited to use only 80-90% total CPU time) to take up the slack you can start to get a little short on RAM if, like me, you only have 1GB to play with.

I just minimise the VMware window and carry on using windows as usual - bit more laggy than usual but combined output makes it worthwhile :)

I'm getting about 11 mins a frame for a p3025 on my [email protected] - not sure what I'd be getting if run natively but the small amount of slack will run a GPU client at about half speed completing a GPU frame in about 12 mins.
Combined output of the SMP and GPU clients is about 1100PPD which is a pretty good improvement over what I've been getting previously. :cool:
 
An X2 3800 @ 2.6GHz is completing a frame in about 8 mins on the 3025 units
on native install.
Looks like you are matching it or getting slightly higher points wise by using the GPU aswell.
And you do have full access to Widows aswell.

It might be a good idea for me to use the VMware on my workstation instead of 2 X normal client.


Looks like Stanford has a SMP server down at the mo, Im getting the assigned to server 0.0.0.0 on one system.
 
For anyone interested in SMP, as rich mentioned VMware is an extremly easy to set up and use option. Normal windows running as per usual for me, just with my SMP crunching in the background, and GPU client on for any time that I'm not gaming.

Although for anyone using SMP take note of the "25th jan bug" thread, and read all of it before you start as it will save you much time and wasted effort.

Perhaps someone linux proficient could look to do an SMP setup guide and sticky it or add it to current folding sticky?
 
sculptor said:
Looks like Stanford has a SMP server down at the mo, Im getting the assigned to server 0.0.0.0 on one system.
Click to expand...

Typical! Got two SMP's waiting to return that have finished in the last half hour. :rolleyes:
 
Just checked the F@H forums again.

Uncle Fungus thinks the server may have collapsed under load.

Guess they didnt realise how big a genie theyd be making with the SMP client.
 
You must log in or register to reply here.
Back
Top Bottom