Upcoming release of SMP2 cores http://foldingforum.org/viewtopic.php?f=24&t=13038 Points system for SMP2 work units http://foldingforum.org/viewtopic.php?f=24&t=13039 I'm itching to try these, but won't have time in the next week or so, leaving the field clear for another CPC folder to be the pioneer.
Have updated one of my linux box's to the new core. Have not had much luck trying to put the new core onto my notfred's machine thu. Just have to wait for NotFred to update the VM i guess...
I got one of those new a3 jobs, but what a drop in PPD!!! With older WUs I was getting ca. 3300PPD, and now 990.
That's because FahMon can't deal with the new bonus system. Your figures are similar to those mentioned on the Stanford Forum, so you should get a reasonable bonus. (Anyone know how to look up what a bonus was given to a completed WU?)
Try HFM.NET, although I think that it's only reporting a base figure without the bonus...At least I hope so. Fahmon is reporting 941 ppd. HFM.NET reports 4623. This is on a Q6600 OC'd to 3.4GHz.
HFM.NET is apparently calculating bonus. Hopefully the calculation needs to be fine tuned as it will obvioulsy only have been used for bigadv wu's until now.
It could of course just mean that I dont qualify for bouns yet? I dont know how that could be known, I have folded with this passkey just not for a while.
HFM.NET does report ppd with bonus. Caution, my personal conclusion is that using core A3 is highly likely to result in a ppd drop on a Q6600.
Interesting, it seems that overall credit you get depends on how your CPU is performing. If I leave it and do nothing it reaches 2700 points, but if I for example turn on some other application, thus using some of the cycles that would normally go towards folding, time to return the WU is longer and the credit gets smaller (2300 points as I write this in a Firefox browser with two tabs opened, for example). Overall PPD seems o be around 4950 on Q9400 OC'd to 3.53GHz. That is a massive step up from what I had before the update to 6.29. I will leave it for some time now and see if my overall amount of points I receive every day will increase, confirming what I just wrote, or drop - confirming what Tesla effect wrote.
Reported PPD seems to increase over time or with more completed wu's. My Q6600 is currently back at a2 benchmark level.
It seems, that I got 2023 points for the a3 job that took ca. 12H to complete, yet to confirm this though. But it also looks like it takes ages for stanfort to process the result, add the bonus and put it to update.
I updated and ignored, but now I have seen this thread I went to look. My bog standard Q6600 used to run (SMP under windows) at ~1200PPD but HFM is now reporting 3000+PPD so I'm happy
Yes indeed, an unfounded & rash conclusion on my part. The 6.29 client with the a3 core running 601x projects rocks
Not got any of the new ones yet on any of my 10 CPU's, will be interesting to see which ones get effected more having such a mixture of AMD/Intel Duals/Quads...
Mmmm can't wait to try these out once the current WU's finish... Should be interesting to see how much of a difference it makes running them native on some of the Windows boxes I have (the part machines).
Cool, thanks for that link. I've just converted over 5 of my full time machines to the new A3... 4 Quads, 1 Dual so will keep an eye on the PPD of those and see what happens. One thing I have noticed right away is it seems more efficient in the way it spawns, there's a single process on my Linux boxes rather than the 1 per core that I was getting before. So far it's sat at a flat 400% (i.e. 100%) where as the older setup used to bounce around between 90%-100%... Anyway, time will tell I guess. Will be nice to get my PPD back up to normal
