• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

3080 Quick question

BluD

BluD

BluD

Soldato
Joined
6 Jan 2013
Posts
3,774
Location
Yorkshire
Hey there, just a quick one.

My Palit 3080 just arrived so i thought i would do a stock bench run to check everything is running ok, i compared it against 3dgurus review and i seem to be getting much lower scores(10-20%).

I had a quick look through the forums and saw people suggesting inadequate power supplies or using max performance settings why don't apply here, i was hoping im just been stupid and missing something. I did a fresh install of latest drivers and ran a few CPU benchmarks just to check thats performing ok, anyone got anything obvious i could try? Cheers.
 
Yup, i had SLI in there before, and they were pulling lots more power thats why i wasnt concerned about the power supply.
 
I did Timespy first which i only got 13k on and Guru3d got 15k so then i bought Port Royal just to make sure and i got 10k and they got 11.5k, i used to do a lot of benchmarking years ago but i havent really done much since i got my last cards up and running so i'm just making sure that im not missing something stupid since a lot could have changed in the last few years.

First i tried using all stock nvidia settings, then i went into the drivers and did switching to high performance, max performance and all the other little tweaks but it only made a slight difference as i expected it would.
 
Oh thankyou, thats put my mind at rest, maybe they did a driver change since they posted their scores and its nurfed the cards byt that amount. Im guessing your timespy result is much higher than mine because your CPU is a lot more powerful? I got 15k GPU score and 9k cpu i think if i remember correctly.
 
First thing I'd do is graph the GPU and vRAM clock speeds during several benchmark runs, as well as the temps. See what you're actually getting in terms of real world clock speed for the entire run.

Performance not being as high suggests the clock speeds are dropping, which is plausible because the regular clocks for these GPUs are on the low side and you get a decent percentage of your total horsepower from the boost clocks exceeding them. But boosting is based on available power and temps at the time, if one or both hit their limit it will drop clock speeds and you can measure that. If it's temps then you can play with the fan profile to make it more aggressive so it's louder but cools more, therefore better temps and higher clocks. Or if its power and your PSU is good then could just be bad silicone lottery and you have a GPU that has high power draw. Although to such a degree to make 10% difference seems a little high.
 
BluD said:
Oh thankyou, thats put my mind at rest, maybe they did a driver change since they posted their scores and its nurfed the cards byt that amount. Im guessing your timespy result is much higher than mine because your CPU is a lot more powerful? I got 15k GPU score and 9k cpu i think if i remember correctly.
Click to expand...

my cpu is 1k score bettter yea:) but even overclocking the card in afterburner to+15 gave me much betetr then stock numbers it seems to hold the boost more even witha simple oc on it don't know why
 
Ok i guess ill start overclocking now, i just wanted to make sure my stock scores were good before i started cause i've been seeing reviwers getting way higher scores so i was confused.
 
Just increasing the power limit to 109% gives a 1k boost in scores, i guess its hugely power limited, time to go to maplins i guess :)
 
Hehe yeah there is one in my nearest town, i only go there every few years when i get a new GPU tbh though.
 
You must log in or register to reply here.
Back
Top Bottom