Page 1 of 1
Titan V PPD low
Posted: Tue Mar 03, 2020 11:26 am
by CommanderLake
I've got a watercooled Titan V thats cool as a cucumber at around 1980MHz core and 945MHz memory while running F@H and completely stable but I'm only getting around 230kPPD with project 11737 on core 0x22 with an estimated TPF of 28s, the utilization is 90%.
I got it second hand good as new to do some CUDA programming of my own and it performs stunningly with a peak of about 19 TFLOPS.
After looking around the forum it looks like I should be getting 10x more PPD than I am so why only ~230k?
Re: Titan V PPD low
Posted: Tue Mar 03, 2020 4:29 pm
by Joe_H
Welcome to the folding support forum.
The most common cause of low estimated PPD is not using a passkey. If you did not configure your client with a passkey, you can apply for and get a passkey to enter.
Re: Titan V PPD low
Posted: Tue Mar 03, 2020 4:37 pm
by CommanderLake
I figured that out eventually, now I've seen over 3.4M PPD!
Re: Titan V PPD low
Posted: Tue Mar 03, 2020 4:45 pm
by JimboPalmer
The more powerful the GPU, the more the Quick Return Bonus helps the Points Per Day!
Re: Titan V PPD low
Posted: Tue Mar 03, 2020 4:58 pm
by CommanderLake
Over 4M PPD with 11739!
Re: Titan V PPD low
Posted: Tue Mar 03, 2020 9:02 pm
by HaloJones
what does a Titan RTX get??!!
Re: Titan V PPD low
Posted: Tue Mar 03, 2020 9:13 pm
by Nathan_P
I would say nearly the same as a Titan V, less cores but higher clocks.
Re: Titan V PPD low
Posted: Sun Mar 08, 2020 8:53 pm
by bruce
Nathan_P wrote:I would say nearly the same as a Titan V, less cores but higher clocks.
FAH performance is reasonably closely related to the GFLOPS numbers. GFLOPS are proportional to NCores*ClockRate.
Within a given generation, Watts/heat is sort of proportional to GFLOPS, but later generations (almost) always improve that relationship.
Re: Titan V PPD low
Posted: Fri Sep 25, 2020 6:18 pm
by Duce H_K_
CommanderLake wrote:Over 4M PPD with 11739!
Excellent, my congratulations !
