News from PG re GPU benchmarking

extide said:
Well, here is the thing, in SINGLE precision floating point, GPU's are quite a bit faster than CPU's. However in DOUBLE precision FP, that is not necessarily the case. A 4P system probably actually has quite a bit MORE DP FP power than a current high end GPU. I dont know the exact numbers off the top of my head. I am not sure if F@H requires DP or if SP is good enough...
Click to expand...

They use SP...
The difference to this point has been the type of calculations...
 
I think its possible they already knocked down the ppd value somehow. I'm in my phone or I'd check.

I finally got some 8057. I'm seeing only 18k ppd on my GTX460 and "only" 60k ppd on my GTX470s
 
I finally got a few of these, only three on three different GPUs overnight. Was about 60k on 460s OC to 1600 shaders according to HFM. Would have been nice to get a few more. :(
 
It seems that anyone who is paying attention on or off the beta team has set their system to try and grab these WU.

This leave the beta team with scraps to try and report on.
 
rhavern said:
I finally got a few of these, only three on three different GPUs overnight. Was about 60k on 460s OC to 1600 shaders according to HFM. Would have been nice to get a few more. :(
Click to expand...
Interesting that your ppd was so much higher than mine. My cards are running at stock. It appears with the QRB the ppd will be *very* sensitive to clock speeds.
 
jebo_4jc said:
Interesting that your ppd was so much higher than mine. My cards are running at stock. It appears with the QRB the ppd will be *very* sensitive to clock speeds.
Click to expand...

Another part (guess) is that these are smaller WU and thus require a smaller number of shaders, but love fast ones.

I will love to see (if the above is true) how larger WU react and if they slow down and the GPU shader and memeory gets saturated.

They shouldn't just use one small quick WU to bench everything off of.
 
Kendrak said:
It seems that anyone who is paying attention on or off the beta team has set their system to try and grab these WU.

This leave the beta team with scraps to try and report on.
Click to expand...

Indeed, it has been rather dry regarding beta WUs, hence my suggestion earlier about implementing a white list for beta testers.

jebo_4jc said:
Interesting that your ppd was so much higher than mine. My cards are running at stock. It appears with the QRB the ppd will be *very* sensitive to clock speeds.
Click to expand...

Yeah, it surprised me too, but in a happy way. :cool:
 
I just wanted to stop by and say that even though I have been following this thread intently, and I really want the QRB work units as I have a very highly clocked 570, I am respecting the beta-team rules and am not using the flag to try and grab these until they are on advanced or regular. I would hope others would do the same, but I'll leave that to another discussion.
 
sticks435 said:
I just wanted to stop by and say that even though I have been following this thread intently, and I really want the QRB work units as I have a very highly clocked 570, I am respecting the beta-team rules and am not using the flag to try and grab these until they are on advanced or regular. I would hope others would do the same, but I'll leave that to another discussion.
Click to expand...

Good man and thanks for delurking to say so. You can always set advanced to get late stage beta units without impacting the beta team or having to join the beta team.
 
as Kendrick said,
do not grab these wu,s.
let the beta team have them,i took off my flags,and it really sucks,just thinking how the extra points would help to catch the horde.
Damn,but this sucks,i could cheat and catch up.
fold on brothers
 
patchesanook said:
as Kendrick said,
do not grab these wu,s.
let the beta team have them,i took off my flags,and it really sucks,just thinking how the extra points would help to catch the horde.
Damn,but this sucks,i could cheat and catch up.
fold on brothers
Click to expand...

Its not cheating but it does hurt the beta teams ability to provide input... they need to be able to fold them.

Thank you patches.
 
My train of thought is the faster and more the beta team gets them to test, the faster they get to advanced and regular fah and then everyone can enjoy the love :)
 
Quite widespread the last 24 hours - EVGA took a 3m point boost from it, might need to redeploy some HW
 
Nathan_P said:
Quite widespread the last 24 hours - EVGA took a 3m point boost from it, might need to redeploy some HW
Click to expand...

Think that has more to do with their typical beginning-of-the-month evga bucks ramp. Look at their production at the beginning of October, same story.

Nothing wrong with kicking out some hardware though ;)
 
colinstu said:
Think that has more to do with their typical beginning-of-the-month evga bucks ramp. Look at their production at the beginning of October, same story.

Nothing wrong with kicking out some hardware though ;)
Click to expand...

The question is why the hardware wasn't folding already in the first place.
 
402blownstroker said:
Is that $10 per 350K or $10 up to 350K?
Click to expand...

It's $5 for 150k, and then another $5 when you hit 350k. So a total (up to) of $10 a month.

Heck, if people could get $10 for every 350k, everyone would be folding for them and evga would be broke! Nope, it's a $10 for 350k max each month limit, which equates to the ramp-up and ramp-down ya see each month for them.
 
One last time guys - Keep it on topic, this thread is not for discussion of EVGA, or the EVGA Bucks program.
 
You must log in or register to reply here.
Back
Top