Any word on Linux folding for -bigadv??

Is there any time schedule for -bigadv to again be supported by Linux?

Ahhhh, ask and ye shall receive (from the folding forum):

Re: Is Big advance ever coming back to Native Linux....

Postby kyleb » Fri Jan 28, 2011 5:24 am
I am currently testing version 2.27 of the A3 linux64 core. I believe this core should solve the remaing Linux Bigadv issues. Once testing is complete, we should be able to enable bigadv on linux.

SWEET!!!!

Thanks for info orion456

Very nice. Just upgraded my CPU cooler and I'm pushing my x6 up to 4GHz (150Mhz shy as I type this). I've been wanting to try the bigadv units in a VM set with 8 cores. It will be nice not having the overhead of windows running at the same time.

Zerix01 said:

Very nice. Just upgraded my CPU cooler and I'm pushing my x6 up to 4GHz (150Mhz shy as I type this). I've been wanting to try the bigadv units in a VM set with 8 cores. It will be nice not having the overhead of windows running at the same time.

Click to expand...

You could use the Windows SMP client through WINE.

LOL, I found that out about an hour ago. I believe I was following a thread that you were commenting on. Man people were getting really upset over the x6's running bigadv even though it looks like as long as they are setup right they perform pretty good (about a day or more left before the deadline). Really when it gets down to it a low end i7 not setup (OC'd) correctly won't make the deadline, so it is really a matter of being a responsible folder.

Here is the guide, if your native OS is Linux then just skip to step 8 I think.
http://www.overclock.net/folding-home-guides-tutorials/817550-windows-7-setting-up-bigadv-amd.html

Also, when it really gets down to it, the X6 is 20% slower than what Pande Group considers to be the minimum spec machine to run -bigadv. If you were really a responsible folder, you wouldn't be fooling the client into believing you had 8 cores.

I don't know, from the numbers I am seeing from other people the x6 is comparable to the i7 860.

Isn't the 860 an acceptable chip for bigadv?

The numbers I looked at show the i7 860 @ 3.9 GHz to be about 6% faster (TPF) than an x6 at 4.0 GHz. Adjust to clock for clock, since the chips have the same OC potential, and that would be 9%. I admit that it is close and one could certainly make the argument that the x6 folding -bigadv without a gpu will produce at a TPF on par with the i7 8xx running -bigadv with a GPU. You have to draw the line somewhere though. The X6 isn't that much faster than a Q9550 at the same clock, so why don't we drop the requirement and let them fold too? They'd make the deadline on all but one WU.
Personally, I'm not happy that the first generation i7 is allowed to fold -bigadv. It totally screwed the points system. Now the 2600K is another matter. It's faster than a lot of real 8 core servers.

Well not going to happen for me for a while anyway. Looks like my system is still unstable. I should have returned a WU at 6 and was holding out for 9 but I don't see any point updates. I'm not trying bigadv BTW, just smp and I had the system stable all day.

Edit: Nope just finished later than I thought, stable @ 3.864GHz!

ChasR said:

Now the 2600K is another matter. It's faster than a lot of real 8 core servers.

Click to expand...

It looks like that works too. I saw someone with a 2500K @ 4.5GHz got bigadv up and running. It gets 25,000 PPD on project 6900. I've been looking at a lot of numbers tonight but I think the x6's were around the same PPD for that project. No idea what a 2600K would get in comparison. Still not bad for a quad.

48,771 PPD. 25:03/Frame on p6900 @ 4.8 GHz

Nice. I missed that the 2600K has HT and the 2500K does not.