Overclocking My 1080ti

[IronMike]

So this might sound stupid but i need help with overclocking my GPU. I've got a EVGA 1080ti FTW3 Hybrid card and I would like to see what it can do.At this moment it is running at stock clocks and i don't know how to start with the overclocking on this card. My 8700K is overclocked to 5.2ghz and now i would like to see if i can get more frames in some of the games I play. Some people say its safe to play with the voltage and some say no. Can someone that maybe has the same card show me the way to a good overclock or can someone put me in the right direction please.

 

[Ageless_ZA]

@Gouhan Can maybe give you some pointers.

 

[IronMike]

@Gouhan Can maybe give you some pointers.

Thank you. Ill ask him now.

 

[Gouhan]

Cooling is what will determine your OC not voltage.
In all honesty you're just not going to get that much more out of the card than you are now. More voltage won't add even 10MHz. You can try the following
- Lower voltage, gives more thermal headroom so you can try -20mV and see if that works.
- Set fixed GPU 3D clock via EVGA Precision X (K-Boost) to tyest above -20mV stability
That's pretty much all you can do. Even a GPU water block won't help much as to get GPU Clock frequency scaling you need to keep that GPU @ 30'C and you should gain 30 to 50MHz.

Not much more to add sorry, perhaps decrease your GPU mem frequency so as to lower IMC load and that can give you a few extra MHz, but honestly it is not worth it. Best way to find your cards max is to set 100% fan speed, - 20mV and add to the clock 20MHZ~ at a time (with K-Boost on). Run TimeSpy Extreme if you have it if not try HWBOt Unigine Heaven Xtreme Preset. Regular Firestrike is useless for testing max clocks.

 

[IronMike]

Thank you

 

[CRE4MPIE]

This is how I go about this usually ..

Note: GPU Boost automatically de-clocks your GPU as it gets hotter - so the longer you can cool your GPU the longer it will keep a higher clock.

Beginner stuff:
Increase power target to max
Increase temp target to max
Set a custom linear fan profile ( the default fan profile is pretty poor )

This alone should give you a higher average frame rate by even up to 10% across the board.


Intermediate Stuff:
Once happy with that - you can start fiddling with the standard core and vram clocks. There isn't much of a science to this, increase your core clock by 10Mhz - do some GPU stress testing with 3dmark or heaven on the extreme preset.

When the card fails ( either BSOD or app crash )- roll back and try again. This would most likely be the point where your GPU says no more and you can safely run that core without any issues.

Onto the memory ...
The memory on the 1080Ti's clocks pretty good for the most part ... You could probably start at 100Mhz and increment this by 100Mhz until you start seeing artifacts or again the above scenario.


Advanced Butt Stuff:
Voltage tweaks - This is where things start getting a bit more risky, you could increase the voltage on some cards, the result here is as stated - you make more heat, so if you had the cooling you would see a gain in core Mhz by doing this. So a bit of a double edged sword really - you could end up losing performance due to heat. You MAY get a higher boost clock off the bat, but the drop in clocks will be more significant.

Hope this helps !

On my 1080Ti STRIX OC I am current running +25Mhz Core ( best I could get without issues ) and +250Mhz Mem ( I could push mem more, but it was happy here).

This results in a boost clock of 2025Mhz and drops down to 1999Mhz ingame.

CP over and out !

 

[Gouhan]

This above. Unfortunately NVIDIA took all the OC headroom out of the silicon and sold it at max clocks.
It's a little better with upcoming GPUs but you're still looking at 100MHz max - and on a GPU that does over 2G stock, 100MHz is 5% which is nothing as that translates to 2 to 3% in game performance if you're lucky (most games now are GPU compute bound which do not scale in the way fixed function operations do)

The best thing you can do is try keep stable clocks in game so for instance, instead of having your clock go from 1999 to 1977MHz~ you can keep it at 1999MHz to max OC.
I've been asking around about this and it's largely due to smaller nodes and the thermal dissipation constraints they impose, as the heat simply can't escape fast enough to yield the typical gains we used to have. It gets worse with every node shrink and 9 to 7nm will be worse. It's precisely why on CPU side we can't do much better than 5GHZ~ which we've had since Sandy Bridge's 32nm node, even though we are at 14nm.

I shouldn't say this as that is what my bread and butter is based upon, but there's literally no difference between GPUs anymore. There's no cooler that will give you even 20MHz more than the others. The difference has moved right up to exotic cooling with DICE or LN2. Other than that, leave your GPU as is and play with memory as there's leg room there for all things greater than FHD as said by @CRE4MPIE; memory is what you may want to play with.

If you want you can consider different drivers as well as clock limits will differ slightly, but I'd not bother with that honestly as it's not worth the hassle.

 

[IronMike]

This is how I go about this usually ..

Note: GPU Boost automatically de-clocks your GPU as it gets hotter - so the longer you can cool your GPU the longer it will keep a higher clock.

Beginner stuff:
Increase power target to max
Increase temp target to max
Set a custom linear fan profile ( the default fan profile is pretty poor )

This alone should give you a higher average frame rate by even up to 10% across the board.


Intermediate Stuff:
Once happy with that - you can start fiddling with the standard core and vram clocks. There isn't much of a science to this, increase your core clock by 10Mhz - do some GPU stress testing with 3dmark or heaven on the extreme preset.

When the card fails ( either BSOD or app crash )- roll back and try again. This would most likely be the point where your GPU says no more and you can safely run that core without any issues.

Onto the memory ...
The memory on the 1080Ti's clocks pretty good for the most part ... You could probably start at 100Mhz and increment this by 100Mhz until you start seeing artifacts or again the above scenario.


Advanced Butt Stuff:
Voltage tweaks - This is where things start getting a bit more risky, you could increase the voltage on some cards, the result here is as stated - you make more heat, so if you had the cooling you would see a gain in core Mhz by doing this. So a bit of a double edged sword really - you could end up losing performance due to heat. You MAY get a higher boost clock off the bat, but the drop in clocks will be more significant.

Hope this helps !

On my 1080Ti STRIX OC I am current running +25Mhz Core ( best I could get without issues ) and +250Mhz Mem ( I could push mem more, but it was happy here).

This results in a boost clock of 2025Mhz and drops down to 1999Mhz ingame.

CP over and out !

Thank CP. At this stage its running at 2025MHz in Game with a +250 on mem clock and +50 on core and it's running like a Boss.

 

[IronMike]

This above. Unfortunately NVIDIA took all the OC headroom out of the silicon and sold it at max clocks.
It's a little better with upcoming GPUs but you're still looking at 100MHz max - and on a GPU that does over 2G stock, 100MHz is 5% which is nothing as that translates to 2 to 3% in game performance if you're lucky (most games now are GPU compute bound which do not scale in the way fixed function operations do)

The best thing you can do is try keep stable clocks in game so for instance, instead of having your clock go from 1999 to 1977MHz~ you can keep it at 1999MHz to max OC.
I've been asking around about this and it's largely due to smaller nodes and the thermal dissipation constraints they impose, as the heat simply can't escape fast enough to yield the typical gains we used to have. It gets worse with every node shrink and 9 to 7nm will be worse. It's precisely why on CPU side we can't do much better than 5GHZ~ which we've had since Sandy Bridge's 32nm node, even though we are at 14nm.

I shouldn't say this as that is what my bread and butter is based upon, but there's literally no difference between GPUs anymore. There's no cooler that will give you even 20MHz more than the others. The difference has moved right up to exotic cooling with DICE or LN2. Other than that, leave your GPU as is and play with memory as there's leg room there for all things greater than FHD as said by @CRE4MPIE; memory is what you may want to play with.

If you want you can consider different drivers as well as clock limits will differ slightly, but I'd not bother with that honestly as it's not worth the hassle.

@Oj0 has helped me with the over clock as well and @CRE4MPIE . I just kills me that my CPU clocks like a beast and i was thinking that having a EVGA FTW3 Hybrid would make a difference because it was Water-cooled and i would be able to get better speeds. But at the end im happy with the frames that i get now in games even with the small overclock on it now i can see the difference.

Thanks guys

 

[Oj0]

I bet you miss your 2140 MHz + Gigabyte card now

(I'm probably gonna get a smack for reminding you about it again )

 

[IronMike]

I bet you miss your 2140 MHz + Gigabyte card now

(I'm probably gonna get a smack for reminding you about it again )

You are🤣

 

[baasgene]

honestly just overclock your memory, only 1080 Ti worth overclocking the core is the reference edition fitted with a water block. Because at stock it only goes like 1.8ghz or so, by adding 150mhz +- to the boost you can at least hit 2ghz (most of them can). Or like me just add 50-60mhz and see if it's stable, then forget about it.

My Strix Ti is set at + 60 core, +400 memory. Memory starts artifacting at 490 memory (11980mhz) so I brought it down well enough to ensure happiness even at extended gaming hours. My core is stable at +80 but once again I brought it down for good measure.