Jump to content
1 minute ago, SageOfSpice said:

How far is it safe to crank the Power Limit on a GPU? I've pushed it up to 50% already, how much further could I go?

All the way

Elemental 

Spoiler

Intel i5 6500 @3.8ghz - 8GB HyperX - 600w Apex PSU - GTX 1060 G1 GIGABYTE 6GB - s340 Black - 240gb Toshiba Q300 - Cooler master TX3i - MSI z170-A PRO.

Old Build (sold for 290€)

Spoiler

Intel i3 540 @ 3.9ghz (On stock cooler, Hits 80c max) - 8gb ram - 500w power supply - P7H55-M LE  120gb SSD - Talius Drakko case

Project Frug 50$ Water loop

 

Laptops

Spoiler

13" Macbook Air - Alienware m14x r2 -  2009 15" Macbook Pro (I was give all of these and would never buy them myself)

 

 

Link to comment
https://linustechtips.com/topic/636980-power-limit-death/#findComment-8219664
Share on other sites

Link to post
Share on other sites

6 minutes ago, SageOfSpice said:

Even though I'm ignoring the software limitations? Default is something like 20%. I've found a way to increase it pretty much indefinitely though.

Power limit is just that... A limit. it does not force the gpu to run at what ever you set it at. Cranking it up to like 150% is just removing that artificial limit.

Cpu: Ryzen 2700 @ 4.0Ghz | Motherboard: Hero VI x370 | Gpu: EVGA RTX 2080 | Cooler: Custom Water loop | Ram: 16GB Trident Z 3000MHz

PSU: RM650x + Braided cables | Case:  painted Corsair c70 | Monitor: MSI 1440p 144hz VA | Drives: 500GB 850 Evo (OS)

Laptop: 2014 Razer blade 14" Desktop: http://imgur.com/AQZh2sj , http://imgur.com/ukAXerd

 

Link to comment
https://linustechtips.com/topic/636980-power-limit-death/#findComment-8219720
Share on other sites

Link to post
Share on other sites

The higher you set it, the more power the card is allowed to draw. It might still not draw that much, that depends on the other settings and what software you run.

 

Anyway, drawing more power can be fine or not fine depending on circumstances. If your power supply is strong enough, then all you really have to worry about is the heat that all this power turns into. And if your cooler is keeping the GPU and the VRMs on the card reasonably cool, then that's fine as well.

 

Just make sure you monitor temperatures and stress test in a few different demanding tests and games, to make sure you've found the highest temperatures the card will reach (also bear in mind it'll get hotter if your room temperature gets higher).

Link to comment
https://linustechtips.com/topic/636980-power-limit-death/#findComment-8219735
Share on other sites

Link to post
Share on other sites

1 hour ago, SageOfSpice said:

Hmm. Well, I've got it up a bit. It seems to be thermal throttling as soon as it gets to about 72c. I re-pasted it and tried again, same thing. It runs perfectly fine up until that point. 72c seems pretty low for a GPU to throttle, though.

 

Did you increase the temperature limit at the same time as the power limit?

Link to comment
https://linustechtips.com/topic/636980-power-limit-death/#findComment-8220949
Share on other sites

Link to post
Share on other sites

3 hours ago, done12many2 said:

 

Did you increase the temperature limit at the same time as the power limit?

Nah, I don't have a thingy for that. I'm thinking something is seriously wrong though. Dunno if it's the VRMs overheating or something, but this thing is just unstable.

 

I wanted to test where it would throttle without an OC out of curiosity because I'd seen the temps rise to 75c before without issue on stock clocks. So, I manually reduce and a set a static fan speed. It'd gotten up to about 82c before my computer just outright froze.

 

Such a weird issue, the fact that it's related to temperature in some way is just throwing me off. The odd thing is though, that if I set a custom fan profile and I keep the GPU under 70c, it keeps trucking no problem. The second it passes 72c, it cuts back the core clock for a second, the temps drop by 3c, and then it goes right back up and does it again.

 

I don't know, maybe I'll try another driver.

Link to comment
https://linustechtips.com/topic/636980-power-limit-death/#findComment-8221600
Share on other sites

Link to post
Share on other sites

9 minutes ago, SageOfSpice said:

Nah, I don't have a thingy for that. I'm thinking something is seriously wrong though. Dunno if it's the VRMs overheating or something, but this thing is just unstable.

 

I wanted to test where it would throttle without an OC out of curiosity because I'd seen the temps rise to 75c before without issue on stock clocks. So, I manually reduce and a set a static fan speed. It'd gotten up to about 82c before my computer just outright froze.

 

Such a weird issue, the fact that it's related to temperature in some way is just throwing me off. The odd thing is though, that if I set a custom fan profile and I keep the GPU under 70c, it keeps trucking no problem. The second it passes 72c, it cuts back the core clock for a second, the temps drop by 3c, and then it goes right back up and does it again.

 

I don't know, maybe I'll try another driver.

 

I wish I could help you more on fan curves and air cooling with the card in general, but I'm no expert in that area.  What card do you have?

Link to comment
https://linustechtips.com/topic/636980-power-limit-death/#findComment-8221650
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×