SYY-157 and NT-H2 on AMD - dot or spread with spatula?

[tabascosauz]

Waiting for my Foundation block to arrive, trying to figure out whether I should change my paste application process.

I've gotten into the habit of doing a 3-dot application for my 5900X. Seems to work out generally okay and relatively consistently with lots and lots of NT-H1 and NT-H2.

But Optimus apparently recommends manually spreading paste for the Foundation block - probably something to do with their "Ryzen-optimized" flow pattern. I have a small paste spatula that I've never used.

Should I spread instead when I build my loop, and are there any good guides on how to achieve a good spread for thicker pastes like Kryonaut/KPX/TFX/SYY-157?

-------------

For the uninitiated, I try to do 3-dot on my 5900X (2-dot on 3600/3700X) because of the die layout. Whereas most other chips just look like the 5600G, and I do the old 1-dot on those (all similar-looking, just subtract some die area for Ryzen 3000G/4000G, and imagine a skinny rectangle die for Intel 2nd-12th gen).

The delid picture for 3950X is exactly the die orientation for any chiplet Ryzen CPU when installed in a board:

 

[oobymach]

Some people prefer the dot method because it minimizes the opportunity for air bubbles to form but I'm a super thin layer spatula guy, especially with noctua pastes, they spread super thin fairly easy and I find the thermal transfer is better with minimum paste resulting in cooler temps.

The difficult to spread paste I'm using now I had to spread with my finger as it would stick to any spatula I tried to use on it and not the heatsink on the chip.

 

[Deleted member 215115]

It doesn't matter. I always spread the paste just to have peace of mind that everything is covered but all the applications are within 1-2 degrees of each other.

The only thing that would make a real difference would be an offset mount to bring the center of the coldplate over the CCDs but I'm not aware of any such mounts for custom watercooling.

 

[Solid State Soul ( SSS )]

I prefer spread meathood for guaranteed coverage, dot method might sometimes leave corners of the IHS not covered but thats fine, however when applying paste on GPU die always use spread methood for 100% coverage

 

[puma99dk|]

I prefer spread meathood for guaranteed coverage, dot method might sometimes leave corners of the IHS not covered but thats fine, however when applying paste on GPU die always use spread methood for 100% coverage

I have always done the spreading method on all my CPU's to get full coverage.

I usually use my finger except when using Thermal Grizzy or IC Diamond I use a spreader and the Grizzy stuff is a pain to spread out

 

[Solid State Soul ( SSS )]

I have always done the spreading method on all my CPU's to get full coverage.

I usually use my finger except when using Thermal Grizzy or IC Diamond I use a spreader and the Grizzy stuff is a pain to spread out

It's better to use a spreader or a credit card if you don't have one

 

[puma99dk|]

It's better to use a spreader or a credit card if you don't have one

The maybe but with Noctua or MX4 paste I never had an issue using my finger

 

[maxfly]

I've generally used the spread method to try and achieve the thinnest layer possible. This resulted in the most consistent temps for me vs the pea or drop. I used my finger with a piece of plastic bag stretched over it to spread the paste. I've recently started experimenting with a spatula. I've found that using the spatula takes practice in order to minimize waste. Once I found the correct amount to apply it became much easier to reproduce consistent results.

 

[freeagent]

With AMD I have been using the spatula. With Intel I draw a line down the cores and plant the cooler down on that and give it a titty twister before I give it the clamps. I tried dots with AMD, but continue with the spatula..

 

[ShrimpBrime -retired]

Voted Spread also. I'm accustomed to spreading because most of my chips are or have been de-lidded. I like full coverage.

 

[ryzenmaster]

Waiting for my Foundation block to arrive, trying to figure out whether I should change my paste application process.

I've gotten into the habit of doing a 3-dot application for my 5900X. Seems to work out generally okay and relatively consistently with lots and lots of NT-H1 and NT-H2.

But Optimus apparently recommends manually spreading paste for the Foundation block - probably something to do with their "Ryzen-optimized" flow pattern. I have a small paste spatula that I've never used.

Should I spread instead when I build my loop, and are there any good guides on how to achieve a good spread for thicker pastes like Kryonaut/KPX/TFX/SYY-157?

-------------

For the uninitiated, I try to do 3-dot on my 5900X (2-dot on 3600/3700X) because of the die layout. Whereas most other chips just look like the 5600G, and I do the old 1-dot on those (all similar-looking, just subtract some die area for Ryzen 3000G/4000G, and imagine a skinny rectangle die for Intel 2nd-12th gen).

The delid picture for 3950X is exactly the die orientation for any chiplet Ryzen CPU when installed in a board:

View attachment 225993 View attachment 225994

I wonder,

Does SYY is basically equal quality and performance to TFX? and similar to krynout?

I did pasting of Gelid GC extreme but I am planning in 6 months or so, repaste my ASUS G513QY again both CPU/GPU, and I wonder from SYY/Fuze which one will perform better?

Amazon had good deals so I grabbed both, cant hurt... always can use in future:

http://imgur.com/GaAZiNo

(The cooling chamber on this laptop isnt very well pressure mounting)

Thank you!