Folding Forum

My 2P bigadv* rig get around 200K PPD (doing a 8105 in around 28 hours with 17 mins TPF), where as single 7970 gets around 120K ppd on the core 17 beta WUs, so 2 7970s would get higher PPD then a 2P bigadv rig.
So would you prefer we run 2 gpus over a 2 cpu system, what does more science? It seams for the cost and amount of electricity 2P bigadv rigs use they are no longer worth buying for f@H.

*a high end 2P AMD rig with overclocked Opteron 6276's

They all do science and they all are needed from the lowly old single core cpu to the most advanced multiple processor rig; from low end gpu's to the high end gpu's, they all do very important science.

The question should be what can the donor afford to donate to F@H in equipment and power usage.

The standard line is to optimize for PPD, regardless of hardware, for that is optimizing the science done. PG has designed the point system to optimize the science.

That being said, optimize folding for whatever hardware you already have. No one is busy trying to convince you to buy new stuff just to fold on. What is optimum changes over time especially if you are busy looking at beta stuff which is particularly vulnerable to rapid PPD changes. Then there is the fact that constantly changing your hardware is expensive.

So would you prefer we run 2 gpus over a 2 cpu system, what does more science? It seams for the cost and amount of electricity 2P bigadv rigs use they are no longer worth buying for f@H.

I am alsways reminded of the Army phrase from WWII..."smoke'em if you got'em"

The preference is for the most people possible to run on whatever hardware they happen to own.

People who have 2p systems have been well rewarded over the years for their donation and will continue to be rewarded.

There are quite literally millions of single card gaming rigs out there that are not folding. The new cor looks to improve the efficiency of their operation.
The subsequent points is merely the outcome of that improvement in efficiency.
So, I think the question

Bigadv valued less then multiple GPU WUs

misses the point. The Bigadv are being valued the same...there is just more opportunity to donate with more gpu efficiently.

I don't think there is any concerted effort to pick winners or losers...only to get the most work done possible.


EDIT: PS if you find all of the above pointless semantics...then test and find what works best for you...that's the Raison d'être of the point system.