• 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.

4070 power draw/power % help

Associate
Joined
28 Mar 2007
Posts
2,498
Location
Edinburgh
Hi all,

Just getting the new PC up and running and have been testing the 4070 with Heaven 4. I'm noticing (via Afterburner) that the 4070 is pulling a max of 178w and about 82% of GPU power. I've checked windows power plan and set NVCP to performance.

Any obvious reasons the 4070 would be at those power levels?

Thanks!
 
I can replicate this on my 4070. Checking a 6 month old youtube vid, there is a 4070 pulling more power at stock and getting a much bigger score at 1440p extreme, no aa. I scored 181 fps vs 208. Can only guess it's a windows or driver issue.

 
Last edited:
What are the temps like? Can you post your spec please including make and model of psu?

Temps are good, GPU never gets over 54 deg at 30% fan.

PSU is MSI A850P ATX/PCIe 5.0. I have the 12VHPWR connector direct to the 12VHPWR on the PSU, rather than 2 x 8 pin from PSU into a 12VHPWR connector.

13600K @ 5.5GHz, MSI Tomahawk Z790 & 32Gb Corsair DDR5 6000MHz.
 
Graphics card identified it was running Heaven and decided to have a cig.

Serious answer: would you mind running a benchmark that is ..., more recent.
LoL, fair point. Just used to using Heaven. Any current benchmark's you'd suggest?
 
Temps are good, GPU never gets over 54 deg at 30% fan.

PSU is MSI A850P ATX/PCIe 5.0. I have the 12VHPWR connector direct to the 12VHPWR on the PSU, rather than 2 x 8 pin from PSU into a 12VHPWR connector.

13600K @ 5.5GHz, MSI Tomahawk Z790 & 32Gb Corsair DDR5 6000MHz.
You what? 2x8 pin and an adapter? I have a 4070 it's just a single 8 pin? It's a 200w gpu at peak? It doesn't need more than the pci-e slot power + a single 8 pin?

Also your fan shouldn't be even turning on till 65c stock, atleast my asus doesn't. But I do have dual bios/silent/performance, so who knows. It never goes above 64C though where i've undervolted it.
Never seen a game go above 196w when stock and not undervolted however... TDP is 200w.
 
Last edited:
You what? 2x8 pin and an adapter? I have a 4070 it's just a single 8 pin? It's a 200w gpu at peak? It doesn't need more than the pci-e slot power + a single 8 pin?

Also your fan shouldn't be even turning on till 65c stock, atleast my asus doesn't. But I do have dual bios/silent/performance, so who knows. It never goes above 64C though where i've undervolted it.
Never seen a game go above 196w when stock and not undervolted however... TDP is 200w.

Yep, it's a 4070 FE. Comes out of the box with a 12VHPWR connector on card and supplied with a 2 x 8 pin PCIe to 12VHPWR connector.
 
Yep, it's a 4070 FE. Comes out of the box with a 12VHPWR connector on card and supplied with a 2 x 8 pin PCIe to 12VHPWR connector.
Crazy! I never knew that was even a thing, what a silly idea considering it's a 200w tdp and a pci-e slot does what 70-90w IIRC and a single 8pin can do the rest? Very strange, I hate those adapter connectors!

rolled back to 4070 release driver 531.68 same as vid. Power draw up and score up by 20 FPS. Driver issue.

Strange, I haven't had an issue on any of the releases since buying/installing my asus 4070 on the 25th of July, I must have completed around 20-30 games now too. I've genuinely forgot to update sometimes and even the latest games haven't grumbled with older drivers, then I've bothered to update the drivers and noticed the difference nor gain/losses?

Silicone lottery? Or maybe I'm just lucky? Illuminati?
 
Sure you will have the same issue on heaven as it's driver related. Other games I play have no issues.
Oh I don't run benchmarks mate, I just play games, I don't see the point?

I know what the hardware can do when I research it before I buy it, so I don't get the whole running benchmarks thing, bar when testing an overclock/undervolt, but even then actual gaming is usually a better way of testing stability.
 
Last edited:
Running benches can confirm your Rig is running optimally. No big deal.
Yeah I did say that mate, but unless you're overclocking or undervolting like I was, it's pointless really isn't it, as if your hardware is all new, then it'll just work?
I still prefer to just use games as real world testing though for stability.
Same as a dyno run versus actual driving conditions/use.

My point was more towards those that obsess with benchmarking and 'competing' and posting their results/videos online about it :rolleyes: :cry: :D I find that hilarious those people (not claiming the OP is, just speaking in general) those people could be gaming but they're filming themselves running benchmarks and posting it on forums/youtube haha, how the other half live eh!
 
Last edited:
Crazy! I never knew that was even a thing, what a silly idea considering it's a 200w tdp and a pci-e slot does what 70-90w IIRC and a single 8pin can do the rest? Very strange, I hate those adapter connectors!



Strange, I haven't had an issue on any of the releases since buying/installing my asus 4070 on the 25th of July, I must have completed around 20-30 games now too. I've genuinely forgot to update sometimes and even the latest games haven't grumbled with older drivers, then I've bothered to update the drivers and noticed the difference nor gain/losses?

Silicone lottery? Or maybe I'm just lucky? Illuminati?

Yeah, I'm having no issues in games, performance is what I'd expect. I just found it odd that power draw was so low when benchmarking using Heaven. About to find another benchmark and see if power draw changes at all.
 
Oh I don't run benchmarks mate, I just play games, I don't see the point?

I know what the hardware can do when I research it before I buy it, so I don't get the whole running benchmarks thing, bar when testing an overclock/undervolt, but even then actual gaming is usually a better way of testing stability.

I only really do it when I have a new PC, just to check all is stable with the hardware and the settings I'm using.
 
Yeah, I'm having no issues in games, performance is what I'd expect. I just found it odd that power draw was so low when benchmarking using Heaven. About to find another benchmark and see if power draw changes at all.
I'd say it's correct, it's a 200w tdp, and no games seem to ever push that, I think Control or CP PL with RT on Pyscho at 1440p maxed out with DLSS3.5/FG/Reflex on uses about 196w as with Control at 1440p native maxed with RT on? But I forget now as I've been undervolted within an hour or 2 of owning this gpu, it sips power at 110-135-145w the latter being with RT on, on average it's around 115-135w at maxed out settings at native 1440p or something that requires DLSS/FG/Reflex like CP PL.

So yeah enjoy your new card mate, it packs quite a punch for the IMHO reasonable price of £530 these days + it's nice addition of future proofing feature set options like DLSS3.5/FG/Reflex etc etc, unlike the more expensive featureless 7800XT (a 3 year old 6800XT in a dress being rehashed)
 
Last edited:
I only really do it when I have a new PC, just to check all is stable with the hardware and the settings I'm using.
Yeah that's fair enough mate, quick way to stress it, as I say my point is more this :)
My point was more towards those that obsess with benchmarking and 'competing' and posting their results/videos online about it :rolleyes: :cry: :D I find that hilarious those people (not claiming the OP is, just speaking in general) those people could be gaming but they're filming themselves running benchmarks and posting it on forums/youtube haha, how the other half live eh!
 
Back
Top Bottom