So I was going to get the hydro copper waterblock from evga but I found out it has been discontinued, so what other waterblock options is there for the 2080ti kingpin?

Damn, im debating on selling my kingpin and get another card , but I doubt I will

Yea , I think a custom loop would benefit more then just the 240 water cooled AIO cooler

Here is a thread i started a while back  and a way to use both the EKWB thermosphere, and the EKWB supremacy universal gpu block on the 2080ti Kingpin and still use the coldplate it comes withfor the memory, and still have the vrm fan used that it comes with, and just add the EVGA LN2 kit for it.  After testing and using the supremacy and coldplate combo it cools better then the hydro copper did and the thermosphere and coldplate did. I am actually going to mod 2 of my kingpins with the supremacy/coldplate version and SLI them.    
 
https://forums.evga.com/Modified-2080ti-Kingpin-Using-EKWB-and-Coldplate-m3002608.aspx  

GTXJackBauer
A universal GPU block might work but you'll have to figure out a way to get good air flow on the VRM section unless you can some how modify the AIO shroud to include the block and fan together along with the housing but any permanent modifications to the GPU would void its warranty.

 

 The VRM has heatsinks that are attached to the card, not the shroud. The larger of the two has its own fan. The VRM is so overbuilt that temps are not a concern. 
 

SpecterSniper1
Yea , I think a custom loop would benefit more then just the 240 water cooled AIO cooler

Not really though. Unless you're doing extended benchmarking sessions where heat soak of the AIO may be an issue. Just max out the fans even in that case and make sure your ambient temp is as low as reasonably possible. The difference between the AIO and a dedicated GPU block for be was at best 10c lower than the AIO, with 7 or 8c being more typical. This won't make a difference in gaming even if the card drops the clocks down. For my card, the difference was 2130mhz in the mid to high 30c range, vs the AIO which was 2100mhz at 45c at the same +80 clock offset. 

They are also doing some QC (at least thats what I was told by a CS), seems like the OLED issue is more widespread than I thought, along with the whole temps compared to other WBs. Thats what I was told over phone when I was getting my RMA done.

Since the coldplate on the AIO is unique what about leaving the AIO pump and block/plate on. Remove AIO radiator. And splice in a regular pump and desired Rads. Remove the AIO impeller if it impedes flow. Could it work?

Thinking of trying it.

altonadavid

emmett
Since the coldplate on the AIO is unique what about leaving the AIO pump and block/plate on. Remove AIO radiator. And splice in a regular pump and desired Rads. Remove the AIO impeller if it impedes flow. Could it work?

Thinking of trying it.

That wouldn't be a really good idea, you would never be able to restore the card back to factory condition by doing that. You'd be better to try one of these mods if a hydro copper isn't available

.https://forums.evga.com/M...oldplate-m3002608.aspx

True.. and that mod looks great. I have a few unused AIO's laying around. So I might try it with one of my other 2080's with the NZXT bracket..

The kingpin memory spacing is closer to the die I believe.

Edit: actually, it's the same as every other 2080ti. Actually it's the 2080ti in general that moved the memory closer to the GPU die.. so the mounting holes would be different if using something made for a 1080/1080ti.

I haven't been keeping myself up to date on the specifics of this card.  I did purchase one a bit late under the assumption that a waterblock from EVGA would be available at some point.  I too kind of questioned the performance value as I can run my card at 2160mhz at 45-46c for a 6 hour bench test.  I did immediately install with a push/pull config on the rad though. This is better performance than my 2080ti in a custom open loop which hits 50c with 35c water temp max.  Why has evga discontinued the waterblock ?

Mcwrah
How many rads do you have in that system and size? Curious about the temps vs rad estate etc.

sorry I edited my original post cause I thought I went off topic.  I have 2 360 rads.  1 30mm , 1 medium thickness (I think 45mm?), 9 fans , 6 on the rads, 2 extra intake and 1 extra exhaust.  So overall 5 intake, 4 exhaust.  A pwm D5 pump/res with the tall reservoir.  An EK waterblock on a 9900k, and an EVGA waterblock on a 2080ti xc.  I am running my fans off of watertemp.  I have 2 temp sensors.  Once coming out of pump, and one coming into the res (after cpu/gpu).  I quickly found out that the water moves so quickly that 2 sensors make no difference in temp.  When benching I have the gpu under 100% load and the cpu under a light load the highest the water temp will see is 35c.  That temp stays very stable.  And the gpu delta is 15c.  So 50c on the gpu.  After putting the system through the worst torture I can I back off the fan speed and have them run at about 80% at 40c water temp (which it never hits, so more realistic 70% in worst case scenario).  The problem with the 2080ti is that if you add a second card in sli to the loop your watertemp skyrockets.  I was seeing 48c watertemp when benching the gpu's together and the die's were hitting in the 60 to high 60c range.  pretty much negating the reason for watercooling when you consider that the ftw3 card can run in the high 60's on air.  I then realized that in SLI the second card is hardly every used and all it did was raise my temps on the entire system even when it wasn't under load. It raised the temp on my utilized card 10 degrees in every condition.  Forcing sli in games never worked well and if it did work it causes your frametime to become very unstable.  So I just ditched the entire SLI concept, the only way I would do it again is if it was supported by nvidia and game developers.