• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

5870 PPD?

Someone who runs recent Ati hardware and has good GPU architectural knowledge could extrapolate a ballpark figure. I personally have no idea because I don't follow Ati as much as nVidia. It should be interesting. Hopefully Stanford could improve the Ati client for this new generation. ;)
 
~6000-7000PPD assuming it scales linearly compared to a 4890. I doubt it will be that high though. Probably closer to 5000PPD. We'll find out for certain in the near future.
 
Zero82z said:
~6000-7000PPD assuming it scales linearly compared to a 4890. I doubt it will be that high though. Probably closer to 5000PPD. We'll find out for certain in the near future.
Click to expand...


yeah id say some where in that figure.. looking at the benchmarks it makes me think that the shaders are running much much faster then the 4k series.. so its possible we could see a 7k PPD output on these cards.. though it should be more like 15-16k PPD.. but oh well.. who knows one day we might get an ATI client that actually shows the true potential of the ATI cards..
 
sirmonkey1985 said:
looking at the benchmarks it makes me think that the shaders are running much much faster then the 4k series..
Click to expand...
The shaders run at the same speed as the core just like the 4000 series. The 5870 has a core speed of 850MHz which is the same as the 4890.
 
I say between 2 and 2.3 times more ppd than a 4890 just looking at specs alone.
 
over9000-2.jpg
 
Does the direct compute performance have anything to do with this at all? Because the 5870 did really good in that on the Anandtech review.
 
Erasmus354 said:
Does the direct compute performance have anything to do with this at all? Because the 5870 did really good in that on the Anandtech review.
Click to expand...
That doesn't matter. Folding@Home doesn't use DirectCompute, it uses CUDA on nVidia GPUs and CAL on ATI GPUs. Its performance is also directly related to how the cores are written, and right now that is the big issue with the ATI client. It simply isn't written very efficiently and doesn't take advantage of the power of the GPUs.
 
there's a link on RAGE3d that puts it at 4.6k for a 353 point WU. It's piss poor.
 
Zero82z said:
It's not ATI's fault. They aren't the ones writing the F@H cores.
Click to expand...

One could argue it is their fault for picking a shitty programming language and having plenty of time to fix it. ATI was the first to the folding prom and when nvidia arrives late they sweep the folding floor with ATI despite having a time handicap.

Is it microsofts fault for having some many holes in their OS or those pesky kids who keep on finding them?
 
osrk said:
One could argue it is their fault for picking a shitty programming language and having plenty of time to fix it. ATI was the first to the folding prom and when nvidia arrives late they sweep the folding floor with ATI despite having a time handicap.

Is it microsofts fault for having some many holes in their OS or those pesky kids who keep on finding them?
Click to expand...
ATI picked OpenCL, which isn't a shitty API at all. The problem is the fact that the ATI GPU client doesn't get nearly as much support or resources as the nVidia client does, which is why it takes so much more time for anything to happen with it. And ATI may have been first to GPU folding, but that was with the GPU1 client which was a completely different beast from the current GPU2 client. nVidia was actually the first to support the GPU2 client.
 
Zero82z said:
ATI picked OpenCL, which isn't a shitty API at all. The problem is the fact that the ATI GPU client doesn't get nearly as much support or resources as the nVidia client does, which is why it takes so much more time for anything to happen with it. And ATI may have been first to GPU folding, but that was with the GPU1 client which was a completely different beast from the current GPU2 client. nVidia was actually the first to support the GPU2 client.
Click to expand...

Currently they're using Brook+ they're going to eventually go to opencl. I want ATI to have aswesome performance for FAH. The new card doubles computing performance with only a slight power draw increase. It's just frustrating that they can't pull it together for so long.
 
Nearly 5,000 ppd on a 384-pointer. It is, frankly speaking, pretty shit. Even my 4870 does 3k+ on those. 3160ppd at the moment.
 
Mr. Pedantic said:
Nearly 5,000 ppd on a 384-pointer. It is, frankly speaking, pretty shit. Even my 4870 does 3k+ on those. 3160ppd at the moment.
Click to expand...
That's pretty disappointing. My AGP 3870 accomplishes 2400PPD...
 
I think all the people working on the Ati client will feel the need to step things up with this new generation of hardware. It's sad all this technology and horse power isn't being utilized... :(
 
You must log in or register to reply here.
Back
Top