1. Is there any way to limit how much CPU to allocate to the project? eg say I would rather it was limited to 80% rather than 100% could I do this?
2. I know it is "safe" to have my CPU running at 100% all the time but will it significantly reduce my computer's lifespan? I have a desktop with i5 3570K processor with Akasa Nero fan. I've been monitoring the temperatures and so far the max temp for either of the cores is 56C.
edit: If you want to know where I first heard of Folding@home, it was from the 2009 magazine "101 Computer Projects"
Last edited by UseroNameo on Fri Nov 09, 2012 11:56 am, edited 1 time in total.
Yes, there's a way to limit CPU by percentage; it's a flag somewhere (sorry, can't tell you where, wiki perhaps). I've found that CPU folding is really unobtrusive (OS scheduler), even when gaming.
Technically, I suppose your CPU's lifespan can be reduced from, like, 50 years to 45 years but effectively speaking, no.
You have a K model at only 56C? Are you running stock speeds?
If you fold with 3 cores, you'd be using 75% of your CPU. You can change from the default smp:4 for your CPU to smp:3 by clicking Configure --> Slots --> Edit.
Some client versions have an adjustment for CPU utilization but it doesn't work when you're running SMP -- only when you're running a uniprocessor client. Napoleon's recommendation to use less than the maximum number of CPUs is the only real option for SMP.
One more thing does doing this use up much bandwidth? Do you need to be connected to the internet continually or do you just need internet connection for the software to download a new project then when the project is finished so the results are uploaded from your computer?
I know what setting to use. Please confirm that you actually can achieve 80% total utilizaton by setting some unknown value for cpu-usage.
You can use cpu-usage as a global value if you choose, but you can set it in each slot, too but if you can't actually achieve a total of 80% then you have not answered the OP's question.
Mea culpa: it does only limit one core's utilisation and it only keeps the other cores from being pegged. Originally, I used 90% and it appeared to affect all cores since they were all hovering between 90 and 100% (with other background tasks running, of course).
I used 50% and it is quite clear that only one core is throttled, leaving the others bouncing in the high 90's; with no value specified, all cores remain flat at 100%.
