hellmel

My thoughts are, I don't mind running test WU's. If they improve the performance of the WU, then it will finish them faster or help create one for a project. Someone has to test them. Just my thoughts.

 
+1
better to have them optimized and tested, rather than release them FUBARed
 

leftygof

hellmel

My thoughts are, I don't mind running test WU's. If they improve the performance of the WU, then it will finish them faster or help create one for a project. Someone has to test them. Just my thoughts.

+1
better to have them optimized and tested, rather than release them FUBARed

I said something similar above. However the science is more important to me. So had this been the case I would simply reduce to one 680 and one 670 for testing and put everything else on science... Both need done and I am happy to help...

http://fah-web.stanford.e...IPswillbebanned?p=8057
 
no project info
all other projects have information on what science they do
 
unless they need this to be some great secret 
then I can only see it as a test protein or test model
ie ...a known WU they can test on both systems
 
it's also why people have posted at FF that the WU's in question are repeated (some people have done the same work 3 times)
 
http://foldingforum.org/v...8&start=60#p228156
 

you are correct sir...Im in err
 
 

8018's all night and just pulled an 8043 on the 670...

Punchy

Neither of these is proof that no science is being done.  Remember, this is a BETA project.  Often there is no description in the project info for beta projects.  Beta projects also have a tendency to repeat WUs to verify their calculations.  Please don't spread inferences as if they were facts.

Agreed. If you need a project description, view that of p8054. As for duplication, think for a minute about the sheer number of units being processed, returned, overclocked failures, connection timeouts, server resets, et cetera, and well, you get the picture. Unit duplication can be viewed as a waste of resources, but it is better to have a unit processed twice than not processed at all.

I've got two 570's running 8018 and two running 8057 not sure on 460 or 450

_Nite_

I got a 8054, 8043, and 7623 at the moment, still no sign of 8057 on 6.42r2

 
is it possible on 6.42?
 

leftygof

_Nite_

I got a 8054, 8043, and 7623 at the moment, still no sign of 8057 on 6.42r2

is it possible on 6.42?

 
according to a few others it is yes, you have to set the -betateam flag where as with v7 its just the -beta flag

My system had multiple failures last night!!  So I shut down and am starting all over again... all 8054's BTW... Just said failed... Logs had been of no help and why I simply just uninstalled and started over...

9:00am PT - 3:00pm PT 11/18/2012!   EVGA Z68 FTW 160-SB-E689-KR LGA 1155 Intel Z68 SATA 6Gb/s USB 3.0 Extended ATX Intel Motherboard
$249.99
Your Price: $84.99


$59.99 After $25.00 MIR

Xavier Zepherious
Thank you for the info. This goes with the shut down of the PS3 that started the gpu folding. Looks like a new future to folding and this will put EVGA back on top. AMD saw this and look at how they reacted.
FP comparison CPU VS GPU
http://www.sisoftware.net/?d=qa&f=cpu_vs_gpu_proc

as you can see GPU is at least 4x more powerful than an i7 in calculating FP (just not efficient doing them)
GPU's suck up lots of power CPU's don't

using that info 3930k -produces around 50k PPD
4x for fermi =200k
which is what we are seeing

see other papers at the time that stated GPU 10x-110x more powerful at FP than CPU
some from Nvdia and some from FAH own scientists that USE FAH

papers
https://dspace.ist.utl.pt/bitstream/2295/581147/2/p14.pdf

http://www.deskeng.com/articles/aaayet.htm
20x more powerful

http://www.nvidia.com/object/what-is-gpu-computing.html
nvidia says up to 100x more powerful

just maybe the QRB is correct


further reading (if you want to read about GPU compute)
http://www.mgnet.org/~douglas/Classes/hpc-xtc/2012su-notes/amcs311.pdf

http://developer.download.nvidia.com/assets/cuda/files/NVIDIA-CUDA-Floating-Point.pdf

http://cs.utsa.edu/~qitian/seminar/Spring11/03_04_11/GPU.pdf

I am weak, zero self control. Told myself back in march I would limit myself to 2 rigs from now on. No more than that ever again
 
So since march just left a pile of parts sitting around. Sold a few things, gave some away, let the rest collect dust. Tried to sell a few more half-heartedly to find no interest in my sales.
 
Now this sits running in a corner
 

 
Someone please help me. Yes that is a twisty tie holding the cards apart for air flow. Yes i fried a PSU and ran to best buy for a new one. I need help
 
Hi my name is -ZS-Carpenter, I am a foldaholic.

-ZS-Carpenter

I am weak, zero self control. Told myself back in march I would limit myself to 2 rigs from now on. No more than that ever again

So since march just left a pile of parts sitting around. Sold a few things, gave some away, let the rest collect dust. Tried to sell a few more half-heartedly to find no interest in my sales.

Now this sits running in a corner



Someone please help me. Yes that is a twisty tie holding the cards apart for air flow. Yes i fried a PSU and ran to best buy for a new one. I need help

Hi my name is -ZS-Carpenter, I am a foldaholic.

 
Good thing I didn't take that board over the DFI Bloodiron, I knew someone with more cards would put it to use
 
I would becareful the way the 460's are sitting loosely btw, it could put stress on the PCI-E slots in my opinion.

asus939

i said the same thing and now have 3 going also gonna hate to see the hydro bill

 
Assuming your mean electricity bill there.
that is one thing to be careful of, especially when folding on multiple gpu's ,
The electric bill can add up real quick.