| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Help Conquer Cancer
Thread: Amount of work to still run on HCC |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 86
|
|
| Author |
|
|
mclaver
Veteran Cruncher Joined: Dec 19, 2005 Post Count: 566 Status: Offline Project Badges: 
|
Now, if only I could find a 7950 or 7970 at a good price!  The HD 8000 series will be out sometime in the first part of next year. You could wait for it, or at least the prices on the 7000 series will drop. But to get the most out of those cards, you have to devote several CPU cores to it, so if you are working on other projects, it may not be worth it. More or less, coming from an owner of a highly overclocked 7950, my 4GHz 8 core processor is just barely able to properly keep it fed. If you are to buy one, you'd have to dedicate all CPU cores to helping the GPU stay fed with data. Therefore, you'd want to be running the 7950 with a really high end CPU and overclock it as far as you can so that the GPU isn't spending half the time at idle. Don't even think about buying two, one 7950 per machine (Unless you magically have a 5GHz six core Intel CPU); 100K PPD per 7950 machine (700K WCG PPD). I have a 7970, that is not overclocked, but seems to slow down on a Intel I7 980x with 12 processors. When I run dedicate 12 cores, it produces less ppd than another 7970 runnting on a I7 2600k with 8 cores dedecated.    |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
I have a 7970, that is not overclocked, but seems to slow down on a Intel I7 980x with 12 processors. When I run dedicate 12 cores, it produces less ppd than another 7970 runnting on a I7 2600k with 8 cores dedecated. What are the processor speeds of each CPU and how many GPU tasks are you running on each?
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
----------------------------------------  [Edit 1 times, last edit by nanoprobe at Dec 14, 2012 10:31:06 PM] |
||
|
|
cehunt
Senior Cruncher CANADA Joined: Oct 10, 2011 Post Count: 172 Status: Offline Project Badges: 
|
Hi:
I am glad there is at least 180 days left in this project. I am using an Intel 8200 cpu in one PC and an Intel i7 cpu in my laptop in my effort to amass 720 days of CPU time to this project. Glad to know I am not going to run out of work units. So what is the reason behind the NA posting in the Dashboard Chart? Clive |
||
|
|
mclaver
Veteran Cruncher Joined: Dec 19, 2005 Post Count: 566 Status: Offline Project Badges:
|
I have a 7970, that is not overclocked, but seems to slow down on a Intel I7 980x with 12 processors. When I run dedicate 12 cores, it produces less ppd than another 7970 runnting on a I7 2600k with 8 cores dedecated. What are the processor speeds of each CPU and how many GPU tasks are you running on each? I7 980 3.33ghz 6 cpus, with 12 hyperthread running 12 gpu tasks I7 2600k 3.8 ghz 4 spus with 8 hyperthread running 8 gpu tasks Both averaging about the same ppd although the one running 8 gpu tasks is producing slightly more ppd than the one running 12. why would processor speed make a lot of difference? |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges:
|
I have a 7970, that is not overclocked, but seems to slow down on a Intel I7 980x with 12 processors. When I run dedicate 12 cores, it produces less ppd than another 7970 runnting on a I7 2600k with 8 cores dedecated. What are the processor speeds of each CPU and how many GPU tasks are you running on each? I7 980 3.33ghz 6 cpus, with 12 hyperthread running 12 gpu tasks I7 2600k 3.8 ghz 4 spus with 8 hyperthread running 8 gpu tasks Both averaging about the same ppd although the one running 8 gpu tasks is producing slightly more ppd than the one running 12. why would processor speed make a lot of difference? The processor speed makes a big difference from what I've seen. My i7-2600k @ 4.2GHz. far out performs my i7-860 @ 3.2Ghz. It looks like the newer chips are just much more efficient when it comes handling whatever it takes to run this app. Mobo memory also seems to play a part. There was a noticeable difference between 1333Mhz and 1600 Mhz. Just a very slight improvement when taking 1 box up to 1800MHz. Not worth it to me. Maybe the newer 2133 or 2300 MHz. memory might make it faster but my mobos don't support it and I'm not investing in new ones to find out.
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
|
||
|
|
nasher
Veteran Cruncher USA Joined: Dec 2, 2005 Post Count: 1423 Status: Offline Project Badges: 
|
crunching solid on this project untill a new one shows up
---------------------------------------- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
why would processor speed make a lot of difference? The processor speed makes a big difference from what I've seen. My i7-2600k @ 4.2GHz. far out performs my i7-860 @ 3.2Ghz. It looks like the newer chips are just much more efficient when it comes handling whatever it takes to run this app.Notes: *The relay-race analogy is the model I use to characterize performance-areas in OpenCL-computing. The hand-off of the baton between runners is a critical part in winning the relay-race. In GPU-computing, that hand-off is what figures in matching CPU-threads to simul GPU-WUs along with matching the speed between runners (processing-units). ; ; andzgridPost#744 ; |
||
|
|
Jim1348
Veteran Cruncher USA Joined: Jul 13, 2009 Post Count: 1066 Status: Offline Project Badges:
|
More or less, coming from an owner of a highly overclocked 7950, my 4GHz 8 core processor is just barely able to properly keep it fed. If you are to buy one, you'd have to dedicate all CPU cores to helping the GPU stay fed with data. I don't think it is necessarily advisable to push the GPU to maximum. From what I have seen, there is a loss of efficiency in trying to get the GPU up to 100% utilization, as though that were the proper measure of merit. But if the incremental CPU power (say the 4th or 5th cores on a high-end CPU feeding a fast GPU) doesn't buy you much in terms of total throughput, then I would use the rest of the cores otherwise. Of course, that is a matter of personal preference as to what projects you want to work on. Also, the GPU is a very efficient computing resource, and so you don't want to starve it either, but you should be aware of the tradeoffs. [Edit 1 times, last edit by Jim1348 at Dec 16, 2012 2:38:13 PM] |
||
|
|
coolstream
Senior Cruncher SCOTLAND Joined: Nov 8, 2005 Post Count: 475 Status: Offline Project Badges: 
|
coolstream, the topic of this thread is how much work is left for the project, not a technical problem. My suspicion though is, that as HCC tasks come from 1 pool, not pre-assigned to GPU or CPU when generated, and the run time for CPU is set to time out at X times Y Flops. For a CPU task it would be about the regular time. Don't know what time-out factor is used... 5 times or 10 times or something in between [can be equated from data in the client_state.xml] Apologies for going O/T, Rob, and apologies for apologising which takes it O/T again Your reply makes perfect sense to me now that I remember that HCC +ATI is not seperate from HCC, so it is of course only right that the time out is as long as it is (even though 10 days for a task that should take only minutes is a nonsense). The only reason I became aware of it was when I read the logs from 'stuck' units which error with time outs (as linked to in my original post). I will try to bring my reply back on topic by concluding that this anomaly was the reason for me asking the question. I appreciate that you have answered it and understand the reason for it. I know that it has been asked before if the two would be treated as different projects and appreciate that we have had our reply in the negative. Not many people appear to encounter 'stuck units' so I guess I must just be unlucky on these rare occasions. I think I will ask Fred at BoincTasks if it could be posssible to create a rule for sending an alert if these GPU tasks run for over a certain amount of time. I don't want to wander O/T again so I'll leave it there. Thanks for your reply, it has helped me realise what I should focus on now.  Crunching in memory of my Mum PEGGY, cousin ROPPA and Aunt AUDREY. |
||
|
|
rilian
Veteran Cruncher Ukraine - we rule! Joined: Jun 17, 2007 Post Count: 1453 Status: Offline Project Badges:
|
thanks for the news!
---------------------------------------- |
||
|
|
|