8th Annual Holiday Season Folding@Home Race! The Race is on!!

[petrusbroder]

From now on I should contribute some 170K - 190K ppd.
Lets see if that is true ...

 

[Markfw]

Mine's factory OC but I'll let you know.

can you reply with a link to your exact model ? And I assume you will be using the V7 client ? V6 won't do zeta units.

 

[petrusbroder]

Here are the stats as of December 3, 2013, 18:58 UTC:

Below is the production for the first two days:

 

[Ken g6]

Thanks for the stats, Peter! Your first stats were done just before I sent in three big WUs. So this might be closer to average for me.

 

[theAnimal]

can you reply with a link to your exact model ? And I assume you will be using the V7 client ? V6 won't do zeta units.

http://us.msi.com/product/vga/N780-TF-3GD5-OC.html#overview

With the core clock showing as 1071 in Afterburner it gave around 151k for an 8900 WU (V7 yes), I turned on the factory OC for 2nd unit and it shows core clock now as 1110 so I don't expect any major difference. Have yet to try manually OCing.

 

[Markfw]

http://us.msi.com/product/vga/N780-TF-3GD5-OC.html#overview

With the core clock showing as 1071 in Afterburner it gave around 151k for an 8900 WU (V7 yes), I turned on the factory OC for 2nd unit and it shows core clock now as 1110 so I don't expect any major difference. Have yet to try manually OCing.

For $500 @150k ppd vs $700 for 170k ppd, I think the regular card is a better bang/buck.

Thanks for the update.

 

[biodoc]

For $500 @150k ppd vs $700 for 170k ppd, I think the regular card is a better bang/buck.

Thanks for the update.

Ok, for a fair comparison, I clocked up to 1100 MHz and reduce mem clock by 100 MHz and am now getting 185K PPD on an P8900. $200 is a lot though.

 

[Fullmetal Chocobo]

Beautiful stats Peter. And by beautiful, I mean it looks like we are winning.

I can't remember if I have asked about this before, but has anyone done any crunching in a VM?

 

[PCTC2]

Thanks Peter.

Also... MARK IS BEATING ME AGAIN? I have to step up my game! This is what happens when I have to go on 2 business trips in 2 weeks. I was unable to set up my other four crunchers! (and no, they are not all 32-core SMP like my current 3 crunchers are, and no, I am not borging like last year for crazy up-and-down outputs. These are my servers and gaming machines I haven't used in a while).

 

[Ken g6]

Wow, some kind of muscle car just zoomed by me. It looked a little like a Dodge or Chrysler, or maybe a Plimogz.

Although if some of the above stats are right, one GPU can now do that kind of speed? I may have to upgrade. :hmm:

 

[Ken g6]

Hi Scientists (or Thinkers if you want to help a fellow folder out )-
Need a bit of advice. Got approximately $100 or so to spend on a new GPU. Any suggestions? Got a GTS450 and an old GT240 (not in the race) currently. 7790 seems to be around that price range. How will that do folding? I've seen a few deals where the 7850 drops into the just over $100 range. I'm guessing that will do better than the 7790, but how is the power efficiency on that? There's also the 650 series on the Nvidia side of things. Thanks.

After further review, if you can get a mail-in rebate (I can't with a PO box), this $90AR 7850 looks like a really good deal! You can't OC much with that single fan, but you can't have everything for <$100. Edit: Oh, and it only needs one power plug.

It looks like 7850s may get up to 55kppd! While 7790s only seem to get ~37kppd.

 

[petrusbroder]

Here are the stats as of December 4, 2013:

 

[ZipSpeed]

Thanks for the stats! I should be able to achieve a minimum of 200K PPD starting today, if not, tomorrow.

 

[Assimilator1]

What ppd would a Radeon 4870 give??
I'm reluctant to come off Asteroids@H atm with my CPU.......

 

[biodoc]

What ppd would a Radeon 4870 give??
I'm reluctant to come off Asteroids@H atm with my CPU.......

I tried some google searches but couldn't find an answer to your question. Why not give it try? The "thinkers" could use another TeAm member!

 

[biodoc]

The radeon 4870 is on the "white list" of supported GPUs

0x1002:0x9440:::RV770 [Radeon HD 4870]

https://fah-web.stanford.edu/file-releases/public/GPUs.txt

 

[Markfw]

dang.... If they would stop sending me 8101 units to my 16 core, I could make some points. It won't make the bonus for that unit (dual 2.93 ghz quad cores w/HT), but it will do any other bigadv unit, and get 90k ppd. And I had to downlock my dual 5639's to 2.66 to be stable. (but they still beat my dual 6234's on ppd on all units)

Maybe time to suck it up, and get one or 2 780 GTX video cards, I already have enough PSU to handle them. (850 Corsair's)

 

[theAnimal]

http://us.msi.com/product/vga/N780-TF-3GD5-OC.html#overview

With the core clock showing as 1071 in Afterburner it gave around 151k for an 8900 WU (V7 yes), I turned on the factory OC for 2nd unit and it shows core clock now as 1110 so I don't expect any major difference. Have yet to try manually OCing.

Edit: Nevermind, seems to be working OK now.

 

[Plimogz]

Though I can't pretend to fully understand the stats spreadsheet helpfully provided by petrusbroder (Thanks for the stats, Peter!), I have to say that I feel there's something screwy about the scores for Dec 2nd.

Try as I might, I can't see where the 400K points difference came from.

 

[petrusbroder]

Though I can't pretend to fully understand the stats spreadsheet helpfully provided by petrusbroder (Thanks for the stats, Peter!), I have to say that I feel there's something screwy about the scores for Dec 2nd.

Try as I might, I can't see where the 400K points difference came from.

Check the stats of December 2, 2013, here!
The spreadsheet only shows the raw stats of the given date; the older numbers are "hidden" on other pages in the same excel "book".

 

[Plimogz]

Yep, I was just looking at those numbers. And after having added up the produced points for both teams, everything does look alright.

I guess what I find odd is that when you look at the sum PPD for both teams on Dec. 1st and 2nd, the teams are almost even (952 900 ppd & 1 019 621 ppd for the Scientists and 984 625 ppd & 1 012 750 ppd for the Thinkers) and yet there's this large 416K difference in production at the end of the day...

I can't quite see how that works.

 

[theAnimal]

Yep, I was just looking at those numbers. And after having added up the produced points for both teams, everything does look alright.

I guess what I find odd is that when you look at the sum PPD for both teams on Dec. 1st and 2nd, the teams are almost even (952 900 ppd & 1 019 621 ppd for the Scientists and 984 625 ppd & 1 012 750 ppd for the Thinkers) and yet there's this large 416K difference in production at the end of the day...

I can't quite see how that works.

The PPD is calculated based on the last 7 days of production so that number was off significantly for those racers who just started folding & had zero production for the previous days.

 

[biodoc]

I've got the 780Ti installed on my 3930K which is overclocked to 3.8 GHz. SMP WUs are running on 10 cores, 1 core is free and 1 core seems to be locked to the GPU running core 17 WUs. The PPD alternates by frame like clockwork. One frame is about 183K PPD and the next frame is about 192K PPD and the next frame is 183K PPD. The GPU is overclocked by 100 MHz. The target power is 100% and the target temp is 82C. The GPU temp and fan speed slowly creeps up to 82 C and 60%, runs for a while and then the GPU load drops to nothing for a few seconds (GPU temp and fan speed go down during those few seconds and then GPU load kicks in and the process starts again. Weird, huh? Maybe the temp is hitting the limit and then the GPU takes a break?

 

[Fullmetal Chocobo]

Ugh, what a day. Monitor literally breaks off of it's stand (and it wasn't the stand that broke, but the monitor), and now that I'm on a single monitor, FAH isn't coming up anywhere... :|

This is why I hate F@H

 

[Plimogz]

The PPD is calculated based on the last 7 days of production so that number was off significantly for those racers who just started folding & had zero production for the previous days.

Ok, that certainly would do it. Thanks for clearing it up for me.