NVIDIA Finally Fixes Multi-Monitor Power Consumption of Turing GeForce 20. Tested on RTX 2070, 2080 and 2080 Ti.

[W1zzard]

Today, NVIDIA released their GeForce 416.81 drivers, which among others, contains the following changelog entry: "[Turing GPU]: Multi-monitor idle power draw is very high. [2400161]". Back at launch in September, Turing was plagued with very high non-gaming power consumption, in both single-monitor and multi-monitor idle.

The company was quick to fix single-monitor power consumption, which we tested promptly. Unfortunately, at the time, multi-monitor power draw wasn't improved and people were starting to get worried that there might be some kind of unfixable issue present on Turing that would prevent NVIDIA from fixing multi-monitor power draw.


Using today's 416.81 drivers I did a few quick runs for power consumption on all three existing GeForce 20 SKUs; the RTX 2070, RTX 2080 and RTX 2080 Ti.



As you can see, multi-monitor power consumption is finally back at normal levels. Good job, NVIDIA! While the power draw of Pascal cards is still a little bit lower, the differences are now negligible, and won't have any significant effect on power usage, power bill, temperatures, or the environment.

I also did a run measuring single-monitor idle power - no changes here, but not needed either. In case you wonder, since these changes are part of the driver, nothing special will have to be done for custom boards from the various NVIDIA partners. No BIOS update needed, just install the 416.81 drivers and multi-monitor power consumption will go down by more than half.

View at TechPowerUp Main Site

 

[cucker tarlson]

when is amd gonna fix rx480 though ? draws 5x more than 1060.

 

[bug]

It's great this was just a driver fix. But these still draw 2x as much as Pascal

Edit: And thanks for retesting.

 

[TheoneandonlyMrK]

when is amd gonna fix rx480 though ? draws 5x more than 1060.

Pointless post of the day though this might pip it.

 

[cucker tarlson]

Pointless post of the day though this might pip it.

vega 64 draws 17w, rx480 is at 40w. that ain't normal.

 

[PerfectWave]

Wizzard can u tell us how you did the measurement? Thank you

btw on all my pc with w10 all license deactivaed.

 

[W1zzard]

Wizzard can u tell us how you did the measurement? Thank you

btw on all my pc with w10 all license deactivaed.

Card only power consumption, physically measured, like I always do for my reviews. Details here

 

[qubit]

I do wonder why something simple like this is a problem in the first place. Surely power issues could have easily been ironed out before the product is released.

 

[Vya Domus]

I do wonder why something simple like this is a problem in the first place. Surely power issues could have easily been ironed out before the product is released.

Key phrase : could have

 

[ironwolf]

Is this the driver that Steve and the gang over at GN hinted at yesterday in one of their RTX card problem vids?

 

[xkm1948]

Man dat Polaris power draw.

 

[bug]

I do wonder why something simple like this is a problem in the first place. Surely power issues could have easily been ironed out before the product is released.

It's about clocks and timings. You don't sync right and all of a sudden components are running at higher frequencies than needed.
A few years back, the video cards didn't know how to sync with displays at all when two monitors were connected so they just ran full-throttle

The thing is, video cards tend to spend at least half the time idling, so idle power draw tends to draw (pun intended) my attention.

 

[R-T-B]

Man dat Polaris power draw.

You need to change your sig to "A polaris asks "Where is my share?"

/troll

 

[Xzibit]

Now they need to mitigate

Researcher found how to use Nvidia GPU to spy on browser activity, steal passwords

 

[TheoneandonlyMrK]

vega 64 draws 17w, rx480 is at 40w. that ain't normal.

Is this the thread for that complaint , no it's not ,if you have a 480 still ,like me and want to moan about it fine , make a thread ,resurect one even but i repeat , pointless to mention here imho , it will spark debate ,but not about this card lineup or it's issues and these fixes.

 

[B-Real]

when is amd gonna fix rx480 though ? draws 5x more than 1060.

Maybe not anyone expressed it to AMD?

Man dat Polaris power draw.

Most relevant info of this thread by a milked owner of a 2080Ti. BTW, what do you think of the RTX cards gaining 0 performance/Watt increase?

 

[Fluffmeister]

Man dat Polaris power draw.

Well played sir, you kicked the hornets nest.... then we get stolen passwords, not this thread discuss it somewhere else.... and you just got milked because the competiton is complete shit!

I salute you.

 

[xkm1948]

Well played sir, you kicked the hornets nest.... then we get stolen passwords, not this thread discuss it somewhere else.... and you just got milked because the competiton is complete shit!

I salute you.

I know I know. It is fun to poke the angry red guards though. Those people always prefer personal attack over state of the matter: in this case the power draw of different GPUs. But I guess logic discussion is not on their list.

 

[R-T-B]

gaining 0 performance/Watt increase?

Considering he came from a Fury, I'm betting he loves it, lol.

 

[xkm1948]

Considering he came from a Fury, I'm betting he loves it, lol.

CUDA accelerated genomic sequence alignment has been amazing froggy! SO MUCH BETTER than trouble shooting the crappy OpenCL stuff.

Just hang in there, soon you will have your secret synthetic frogman chromosome ready for injection as we discussed.

 

[Kamgusta]

Is this the thread for that complaint , no it's not ,if you have a 480 still ,like me and want to moan about it fine , make a thread ,resurect one even but i repeat , pointless to mention here imho , it will spark debate ,but not about this card lineup or it's issues and these fixes.

Sounds like the "don't ask, don't tell" policy here.
We name it "the tongue beats where the tooth aches" symptom. Many AMD fans have that problem. No cure ATM, sorry.

 

[the54thvoid]

I know I know. It is fun to poke the angry red guards though. Those people always prefer personal attack over state of the matter: in this case the power draw of different GPUs. But I guess logic discussion is not on their list.

Yeah, it's relevant for sure - coming from that time when the GTX 480 was the laughing stock, to ATI/AMD having better efficiency with the 7xxx series. Each time Nvidia make a card with even mediocre efficiency improvements, you get AMD dissing it.

I think as well you're in the best position to troll (or semi-troll tbh). I clearly recall how pro-AMD you were about Fury X. Then you got one and were disappointed with it. You then went green and are happy. Actual experience of each top end (at the time) gives you a sheriffs badge to wear when it comes to Nvidia/AMD GPU conflicts.

 

[INSTG8R]

vega 64 draws 17w, rx480 is at 40w. that ain't normal.

Mine says 7W at the moment

 

[TheoneandonlyMrK]

Sounds like the "don't ask, don't tell" policy here.
We name it "the tongue beats where the tooth aches" symptom. Many AMD fans have that problem. No cure ATM, sorry.

No sir I don't feed trolls, try again.

 

[londiste]

My RTX 2080 still reports it is using 50W-ish for dual-monitor, even with the new driver.
Granted, I run 1440p@165Hz and 1440p@60Hz side-by-side with an additional UHD 4K TV attached occasionally.