Folding with 2 970's

[Andrewconc]

Hay guys I was thinking about getting a couple of 970's for gaming and of course folding.

 

Curently I get about 110k ppd with my 680 and my cpu, any ideas about the ppd I could expect with an additional 2 970's?

 

Cheers

 

Andy

[Extingquish]

AFAIK the 970 gets around 210k PPD,so two of them could give around 420k

[Andrewconc]

nice 530k ppd would be a nice amount!

 

Thanks for the info!

[Grendle]

Hay guys I was thinking about getting a couple of 970's for gaming and of course folding.

 

Curently I get about 110k ppd with my 680 and my cpu, any ideas about the ppd I could expect with an additional 2 970's?

 

Cheers

 

Andy

The 290s are cheap, but you would save money in the long wrong going with the Geforce 970s becuase they draw ALOT less power.  

[Lays]

I think I was getting around like 500k PPD with my 2 970s when I had them.

[DevinR]

yeah i get 225-270k ppd on my single 970 at stock clocks. depending on the workunit

[Samfisher]

AFAIK the 970 gets around 210k PPD,so two of them could give around 420k

 

That is extremely low.  With a bit of an OC, my 970 is doing 310k PPD.

[blackadder]

That is extremely low. With a bit of an OC, my 970 is doing 310k PPD.

depends on wu, core 15 can suck. I got closer to 250k to 280k with a nice mix if cores.

[Samfisher]

depends on wu, core 15 can suck. I got closer to 250k to 280k with a nice mix if cores.

 

15 does indeed suck.  17 though

[msvelev]

On core 15 I get from 50-70k,and on core 17 I get 210-300k with my 970.

[ProdigyzMined]

The 290s are cheap, but you would save money in the long wrong going with the Geforce 970s becuase they draw ALOT less power.  

It's a marginal addition to power draw, you're talking a few dollars a year. To say that 970's draw A LOT (two words btw) less power than an R9 290 is quite the overstatement. It also depends on the wattage and the efficency of the power supply.

[Imakuni]

It's a marginal addition to power draw, you're talking a few dollars a year. To say that 970's draw A LOT (two words btw) less power than an R9 290 is quite the overstatement. It also depends on the wattage and the efficency of the power supply.

 

Taking a look at the graphs, we get 39W of difference (let's ignore the "uber" mode). This is already the wall draw, so no need to factor efficiency. If your PSU isn't as good as theirs, this number would only be higher, but let's stick to those 39 W.

 

Anyway, 39 * 24(h) * 365 (d) = 341,64 kWh. Using the average cost of 12,29 cents per kWh (data retrieved from here), we get a grand total of U$ 41,987556, which is almost 42$ per year! Considering the price difference between the cards, I think this pays itself off pretty quickly. Not to mention the much lower temperatures, once you consider multiple cards being used at a time.

[bcredeur97]

moar ppd always nice. I cant wait for the day I can afford to get a new GPU cause im kinda needing one now  

[Vavetti]

snip

I don't believe that graph, and actually think the difference is bigger. 

I mean a 280x and a 290 use more power than a 290x sure <_<

[Imakuni]

I don't believe that graph, and actually think the difference is bigger. 

I mean a 280x and a 290 use more power than a 290x sure <_<

It makes sense, once you consider the usage of uber and non uber mode. I also tried to be gentle on the math, I didn't want to risk getting into the 3 digit dollars

[ProdigyzMined]

 

Taking a look at the graphs, we get 39W of difference (let's ignore the "uber" mode). This is already the wall draw, so no need to factor efficiency. If your PSU isn't as good as theirs, this number would only be higher, but let's stick to those 39 W.

 

Anyway, 39 * 24(h) * 365 (d) = 341,64 kWh. Using the average cost of 12,29 cents per kWh (data retrieved from here), we get a grand total of U$ 41,987556, which is almost 42$ per year! Considering the price difference between the cards, I think this pays itself off pretty quickly. Not to mention the much lower temperatures, once you consider multiple cards being used at a time.

That's assuming this card is running 24/7, 365 (it's folding, I know) but it doesn't sound like he'll be folding 24/7, likely more like I do, overnight sometimes and i let it run in the background while I do homework. 

[RedOwen1177]

It makes sense, once you consider the usage of uber and non uber mode. I also tried to be gentle on the math, I didn't want to risk getting into the 3 digit dollars

What is Über mode? Never heard of it.

[Jumper118]

Hay guys I was thinking about getting a couple of 970's for gaming and of course folding.

 

Curently I get about 110k ppd with my 680 and my cpu, any ideas about the ppd I could expect with an additional 2 970's?

 

Cheers

 

Andy

http://folding.extremeoverclocking.com/user_summary.php?s=&u=635691this is what mine get

[Imakuni]

What is Über mode? Never heard of it.

There are 2 BIOS in the card: quiet and uber. You manually choose which bios you want via a switch on the card. I think it's pretty self-explanatory as to what each mode does.... and how much energy they consume.

[Me1z]

 

Taking a look at the graphs, we get 39W of difference (let's ignore the "uber" mode). This is already the wall draw, so no need to factor efficiency. If your PSU isn't as good as theirs, this number would only be higher, but let's stick to those 39 W.

 

Anyway, 39 * 24(h) * 365 (d) = 341,64 kWh. Using the average cost of 12,29 cents per kWh (data retrieved from here), we get a grand total of U$ 41,987556, which is almost 42$ per year! Considering the price difference between the cards, I think this pays itself off pretty quickly. Not to mention the much lower temperatures, once you consider multiple cards being used at a time.

 

Oh FFS. This graph, THIS graph. How many times must I explain this? This chart is completely IRRELEVANT in relation to folding. It's not that hard of a concept, I promise. FurMark, or any synthetic benchmark for that matter, does not have the same power usage as folding. I apologize if it was someone else last time, but without fail someone always breaks out this chart as an argument against folding with 290Xs and I'm getting tired of explaining this. My entire folding rig, consisting of an R9 295x2, only pulls 460 watts max. After factoring in the power consumption of the CPU, Mobo, fans, etc; each core is using just over 200 watts each. Mind blowing right? "But Me1z, how could the power-hungry R9 290x be just as energy efficient as Nvidia's godly Maxwell GPUs? a-derp" I have so little patience for fanboyism, so little. Don't even get me started on how Nvidia skewed their TDP results. There are differences, but for all intents and purposes there is no worthwhile benefit to using one over the other. So I can get actually some sleep tonight, please repeat after me: "In relation to folding, the R9 290X and the GTX 970 are EQUAL." 

[Samfisher]

I don't believe that graph, and actually think the difference is bigger. 

I mean a 280x and a 290 use more power than a 290x sure <_<

 

It's Anandtech, it's as reliable a graph as you'll ever get

[Imakuni]

Oh FFS. This graph, THIS graph. How many times must I explain this? This chart is completely IRRELEVANT in relation to folding. It's not that hard of a concept, I promise. FurMark, or any synthetic benchmark for that matter, does not have the same power usage as folding. I apologize if it was someone else last time, but without fail someone always breaks out this chart as an argument against folding with 290Xs and I'm getting tired of explaining this. My entire folding rig, consisting of an R9 295x2, only pulls 460 watts max. After factoring in the power consumption of the CPU, Mobo, fans, etc; each core is using just over 200 watts each. Mind blowing right? "But Me1z, how could the power-hungry R9 290x be just as energy efficient as Nvidia's godly Maxwell GPUs? a-derp" I have so little patience for fanboyism, so little. Don't even get me started on how Nvidia skewed their TDP results. There are differences, but for all intents and purposes there is no worthwhile benefit to using one over the other. So I can get actually some sleep tonight, please repeat after me: "In relation to folding, the R9 290X and the GTX 970 are EQUAL." 

Yup, my OCed CPU + Gtx 970 folding on both CPU and GPU consumed around 250w. Don't forget, if the synthetic benchmark stresses the 290x a lot more than folding, you can be that it ALSO stresses the 970 a lot more too!

 

I also have little patience with unfair people, defending that you must alleviate the problems on one side, without remembering to give the other side the same discounts.

[Me1z]

Yup, my OCed CPU + Gtx 970 folding on both CPU and GPU consumed around 250w. Don't forget, if the synthetic benchmark stresses the 290x a lot more than folding, you can be that it ALSO stresses the 970 a lot more too!

 

I also have little patience with unfair people, defending that you must alleviate the problems on one side, without remembering to give the other side the same discounts.

 

I did remember to give the other side the same discount, I said that the R9 290X and the GTX 970 were equal, not that the 970 consumed 280w while the 290x was sipping on 200W. I myself had a 970 and its results line up with yours, about 200w without the CPU. Applying the same discount doesn't mean they magically have their power consumption reduced by the same amount.