There was a thread a while back about this, and in case folks don't know. C Board has a team. I recently joined it but we are slipping and we need YOU! *points in an uncle Sam manner* links:
EOC Ranking page for the team: http://folding.extremeoverclocking.c...php?s=&t=43476
Home Page of the project: http://folding.stanford.edu/
The team number for "The Cboarders": 43476
Make sure if you are on a multicore system, grab the SMB version, it is made for up to 4 cores.
This will make your computer create a bit more heat, but most computers are fine to handle this, it is primarily a concern on laptops, make sure you either have enough cooling near your laptop to run it, or only run it at 50% of the processor so you do not create as much heat.
Yes, task manager will say you are running at 100% CPU usage, that is because it is taking up every spare cycle it can get. But the work thread for the folding program is set at the lowest priority, so rarely will you see usage issues, I can game (processor intensive games like Supreme Commander) and not see any detrimental effects, granted folding@home runs extremely slow during that but it is ok.
Oh and this helps humanity! But really, making Cboard at least a top 500 (maybe even higher!) team is more imporant!
EDIT: Oh, I forgot to mention that anything less than 500Mhz(or whenever the SSE/3dNOW! stuff was implemented) is too slow for this, they wont complete the work units before the deadlines. At least 1Ghz is recommended.
Don't worry Wraithan, I'm using this again, finally.
>> Yes, task manager will say you are running at 100% CPU usage, that is because it is taking up every spare cycle it can get.
F@H can run without being in the way, if you actually need some processor speed. You can also turn down the processor usage to 75% - 50% so your fan doesn't turn on all the time (who wants to listen to that?). That's what I do, and it seems to be plenty effective.
It's really easy though, folks. If you have a multi-user computer, you can copy the shortcut into the Shared Desktop folder and it should run all the time.
It's been a long while since I last contributed. 4 years I reckon. My laptop became however too old for it since. I'm afraid I have to pass.
My harddrive started to do some clicks a couple of months ago and the display seems to have lost glare 6 months ago. Even when it's idle for hours on a row, my processor fan is constantly buzzing.
I know my poor baby is dying on me. Its time is coming and I can't blame it. Been with me through snow and sunshine. So I prefer to extend its life a little longer until I can manage for a replacement, if you don't mind (which is being planned for a while now).
When that replacement comes, I'll make sure it dies a glorious death crunching in cprog's team.
It's a deal.
It's been almost two years for me. I got the smb version of FAH and already on the second unit.
seems to really like 64 bit linux with a dual core processor. :D
My machines are always folding. At one point I think I had 39 cores running FAH. Never saw the point of the competition, though. Unlike a lot of other distributed parallel algorithms, FAH works by throwing away the vast majority of the results computed by the clients. Ultimately only a single client will make the "critical step" which allows the entire algorithm to advance forward a click.
Running more clients slightly increases your chances of being that one person who finds the "critical step." But you are not making measurable forward progress to the goal at any other time.
<Checks to see if his FAH's are running...>
FAH is a distributed brute force search, looking for a very small needle in a very large haystack. It's no different (from 50,000 feet) than most other DC apps, Just because they optimized it by not having the clients keep repeating "Sorry, nothing but straw" over and over, doesnt mean you arent contributing.
That said, I don't run FAH anymore for various reasons. I run Seti@home and Primegrid when SAH is down.
If I believed I wasn't contributing, I wouldn't run 38 or whatever cores, would I? I just think that the exact way it works makes the idea of a competition kind of silly. You can put in thousands of hours and never make an iota of useful progress. That doesn't mean it's not worth it to run as many clients as possible.
Originally Posted by abachler
And the clients don't repeat anything to the server. Most of the time the clients never even send their results to the server because they already know they aren't good enough. There's a paper about the algorithm out there somewhere. The solution time scales as O(n) but it does this by discarding almost all of the results. I just think the idea of a competition to measure how many hours you spend spinning your wheels doing nothing is sort of silly. It's not like the time has been wasted, but it's hardly something to be proud about either.
With distributed.net the situation was different because each block you processed was a definitive refutation of a block of keys. A failed FAH block is... nothing, really.
And switching from FAH to Seti makes me wonder about your membership in the human race, honestly ;)
Nice we stopped dropping in position as fast, and it looks like we may actually start to go up. Especially if we can get a few more people folding for our team! It is hardly noticable, the only reason not to run it is if you are on a REALLY old computer (pre-SSE instruction set) or if you don't have install rights on your computer.
Brewbuck: what is your name on F@H or are you going anon (in which case it wouldn't hurt to through our team number in there ;) )
38 cores? I thought I was doing well with 5 running F@H. What the heck do you have access to? That can't be your home network. :)
If I were still living at my parents, I could bring 9 cores online on my home network, as is I am just running 4 cores.
Meh, don't insult me by comparing me to those humans, they are a worthless barely intelligent species whose only purpose was to evolve into something that could evolve into whatever the hell I am. I'm just trying ot get off this ball of mud before they irradiate the whole thing and screw up the gene pool. Seti is the best bet for that since all the space agencies are too in love with rockets and robotic missions to actually do any serious space research.
Originally Posted by brewbuck
The three-trillion dollar war in Iraq made it difficult to fund space programs in the U.S., so at least from a local perspective, I really don't think the human race is doing much of that anyway, abachler.
Well, there has been more programs the past twenty years then ever before for the same period of time. I don't think however that looking for extra-terrestrial life was ever a priority... or even seriously considered outside a few momentary lapses of reason, of which the SETI project is an example. Money is usually better spent where it can show results.