I guess this question goes mostly to owners of hybrid kits since we obviously had to take the cards apart. When I installed mine I just used the pre applied stuff that came with the kit. But as we all know there's always room for improvement and better Temps, so I am just curious to know if anyone has tried this and has good comparison data. I ordered some Thermal Grizzly Kryonaut for my cpu, and if I have enough left over I will probably do one of my cards to see if it helps any. Since the two cards are very close in Temps when running both at idle and under load, this should make for a great comparison.

It would be nice to know what results other people have had though before I waste my time taking my card apart again.

Thanks!

We don't recommend IC Diamond for a GPU; keep in mind there is no heat spreader, and thus you'll be on bare silicon. IC Diamond is highly abrasive, and you risk damage to the GPU itself if used. Instead, we recommend a non-conductive thermal interface material. 

Guys keep in mind that company's usually choose a thermal paste that is highly resistant to thermal pump-out and will last the life of the card.

They don't choose 'crappy' compound to save money ...well some do, but you can tell because it will just be normal white silicone grease.

Make sure to pick a thick non-conductive TIM for the GPU as it will be going through a lot of extreme thermal cycles daily.

Yeah, EVGA has been using Shin-etsu for the past couple of card generations. You might see 1 to 3C difference when using the really good, and expensive, thermal interface material.

Unless, that is, there wasn't enough TIM applied in the first place - then the temperature differences can be a lot more.

bcavnaugh
MX-4 for me but will be testing Thermal Grizzly Kryonaut on my 1080 Ti Black Card.

Not impressed with the kryonaut on the 1080ti's I have currently. You should try Hydronaut or stick to MX-4.

Scarlet-Tech

bcavnaugh
MX-4 for me but will be testing Thermal Grizzly Kryonaut on my 1080 Ti Black Card.

Not impressed with the kryonaut on the 1080ti's I have currently. You should try Hydronaut or stick to MX-4.

Dang and I just spent $11 bucks to do enough one CPU or GPU.
I though Hydronaut was the low end of the Thermal Grizzly Products.
Kryonaut Thermal conductivity 12.5 W/mk
Hydronaut Thermal conductivity 11.8 W/mk

MX-4 (8 Years) Thermal conductivity 8.5 W/(mK)

kelvinteh
Hi
I just purchased a EVGA GTX 1060 6GB FTW+ card 2 weeks back and got it few days ago due to shipping from USA Amazon to Singapore where i stay.
i have just replaced the stock thermal paste with Thermal Grizzly Kryonaut.
My BF1 in 64 players MP gaming drop from 77-78degs with auto fan speed of 1400rpm to 74degs 1200rpm.
I stay in Singapore with average ambient temp of 28-30 degrees.
3 degs drop in temp with a 200rpm in fan speed reduction.
 
Is it worth to replace with the stock paste?
I say its a yes!!!
If you have good thermal paste on hand.
 
Anyway EVGA cards run damm silent
So happy with this purchase.
1st time owning a EVGA card.

Not that much, I was hoping 14-16 Degree Drop.

bcavnaugh

kelvinteh
Hi
I just purchased a EVGA GTX 1060 6GB FTW+ card 2 weeks back and got it few days ago due to shipping from USA Amazon to Singapore where i stay.
i have just replaced the stock thermal paste with Thermal Grizzly Kryonaut.
My BF1 in 64 players MP gaming drop from 77-78degs with auto fan speed of 1400rpm to 74degs 1200rpm.
I stay in Singapore with average ambient temp of 28-30 degrees.
3 degs drop in temp with a 200rpm in fan speed reduction.
 
Is it worth to replace with the stock paste?
I say its a yes!!!
If you have good thermal paste on hand.
 
Anyway EVGA cards run damm silent
So happy with this purchase.
1st time owning a EVGA card.

Not that much, I was hoping 14-16 Degree Drop.

Well 3 degs drop is better than nothing
Next will try Liquid Ultra on it
:)

Scarlet-Tech

kelvinteh
 
Well 3 degs drop is better than nothing
Next will try Liquid Ultra on it
:)

Liquid Ultra is the definition of a bad idea unless you are 100% will to risk your warranty.  

Hmm thanks for the advice
I also don't wish to spoil my evga card.
Its great.
Silent and runs fast enough for me in gaming.
Maybe i just still to Kryonaut paste on it.
 

cloiselle1
"  pump out.  "
 
LOL.
 
You notice how two years NO ONE was talking about this nonsense, 
 
now every aut on the internet keeps spewing misinformation because they read it on some forum somewhere. 

Who's on First?
We talk about Thermal Paste all the time on this Forum.

cloiselle1
"  pump out.  "
 
LOL.
 
You notice how two years NO ONE was talking about this nonsense, 
 
now every aut on the internet keeps spewing misinformation because they read it on some forum somewhere.

Uh?  Thermal Pump-out is a thing.
 
The smaller and smoother the die/heatsink the faster the liquids will separate from the solids and potentially "pump-out" during thermal cycles.  That leaves you with only the solids from the TIM (I.e. dry paste).
 
You don't have to take my word for it, read some of the testing/testing methodology and spec sheets from company's and standard's body's (skip the forums).
 
https://www.google.com/search?q=thermal+pump-out

Over the years, I kept an eye on the TIM testing by many folks who I felt knew what they were doing and were well respected in the community for their efforts.  I personally have seen a difference in temps using MX-4, after it's cure time which isn't too long for about 5+ years now.  I've seen a difference anywhere from 3c-5c+ all around.  Nothing extreme like 10c+.  If you ever did get that kind of a difference, it wasn't because of the TIM used but because of a bad seat.

I also have used Gelid Extreme as that was one of the best performers in majority of the tests I saw across the boards.  Maybe 1c-3c difference from MX-4 I saw at times.  I tried Gelid Extreme and wasn't too happy with it's thickness, let alone you're recommended to run it under water to get it loosened up inside.  I didn't want to deal with that so I stuck with MX-4 and haven't looked back.  Gelid Extreme is great for extreme cooling like Chillers and LN2 for the most part as it really shines during those extreme environments.
 
I know some have brought up AS5 as amazing and I can recall seeing it in the middle to the bottom of the lists in many tests.  To be fair, we're all talking about a few degrees here and quite possibly within the margin of error but if you're looking to squeeze every bit all around in your system, this can surely help and add up nicely.

People need to realize and do their research on these TIMs if they should decide to change theirs since some can be conductive and liquid like to the point it can spill out and short your components.  Make sure people take the proper precautions and put the right amount on there since the whole point is to come as close as possible to a thin layer between the base of the cooler and the CPU.  A pea size does work for all occasions while tightening the cooler down in a "cross" pattern (up, down, left, right, repeat) while twisting each screw half circle rotations.  You keep going till you feel the resistance.  Don't have to go all the way down.  You can than test and see if the cooler is nice and sturdy tight down or not.  And remember, each TIM can have different cure times so find out what your specific TIM's needs are before benchmarking and recording temps. 

Here's a nice Roundup a few years back I remember reading up.  Here is a database of different roundups.
 
While were already here, I'm wondering if EVGA is using EVGA Frostbite on all their TIM based products.  I'm not sure if it's still being sold as I'm having trouble finding it on their site.
 
 

cloiselle1
 
 
I build over 30 systems a year on the side next to my job as a hobby, half of those get delidded and have been doing this since ivy bridge.
 

 
I'd like to see the documentation of you building 30 systems a year and delidding half of them as a hobby .  Do you have a website or youtube page?  I enjoy seeing images of people's work and would like to see your work if I may.

EVGA_JacobF
EVGA GPU's use Shin Etsu thermal paste. It actually is quite high quality stuff.

Awesome.  Thanks for that info.  

My main concern as mentioned above is my CPU.  One of my core temps is always higher than the others, and while under load by a significant amount.  I just ran CPU Z and did the stress test and all cores stayed around 58 to 62C except the one problem core which went to 71C.  Also, my CPU package temp is always higher than any of the cores at idle, as I type this, the cores are between 36 and 39C while the package sits at 45C.  Under 100% load, the package temp and the problem core temp are pretty much the same all the time.  I read about Kryonaut and since I wanted to reseat my H115i on my 6850k anyhow, I figured I'd give it a shot.
 
If I feel overly ambitious I will still try it on the 1080 Ti, however, this raises one more question.  Jacob said that they use Shin Etsu on their EVGA Gpu's, I have a Founders Edition, so I am guessing something else may be used there?  I also just watched a video of a guy putting Thermal Grizzly Hydronaut on a 1080 TI Fe.  He used all stock settings including the stock fan curve.  In every game he tested, the temp was reduced by 3C, however the fan speed maintained a constant 50% with all games.  With the stock paste, it went as high as 63% in certain games.  Makes me wonder how much more of a difference there would have been if he had set the fan curve more aggressively.  The better paste may conduct the heat better to the heatsink, but if the fan is blowing at a lazy 50%, you're probably leaving some degress on the table.  I guess by writing all this, I am really talking myself into trying it out on the GPU after all.
 
Sorry for rambling.