Question TF7 vs MX-6 vs NT-H2

[consptheory77]

So I was going to ask whether I should use the thermal paste I've had in a tube since my last build in 2015, but since I'm buying a Thermalright bracket for LGA 1700, it comes with TF7 paste, so I have new paste if I want, but if I'm already going to use new paste, should I buy another brand? I'm not convinced it makes much difference overall, but I know there are always some variables. I did see the Tom's Hardware article on pastes. I'm going to use this for a 13600K. My own inclination is to get MX-6, based upon reading this:

Arctic MX-6, complete analysis of this CPU thermal paste

"Empirically, we have not been lucky enough to be able to test all the thermal pastes on the market, but certainly from those that have passed through our hands in the last 10 years (and there have been many), it is the one with the best thermal performance... taking 0.7 degrees from the excellent Noctua NT-H2 and 1.9 degrees Celsius from the CORSAIR TM30, which is not bad thermal paste either. "

 

[Phaaze88]

Excluding gpus, and other direct die applications, the TIM used on cpus matters so very little.
As long as the price per gram is good.

 

[consptheory77]

A review from Amazon Canada notes

https://www.amazon.ca/Thermalright-TF7-Compound-Performance-Interface/dp/B08CGJMMQ9#customerReviews

"I never thought much about different viscosity when it comes to thermal paste but for me, the thickness of this paste is exactly what was needed specifically for my Intel CPU. Thicker is better for Intel because of the iLM problem. This thick thermal paste will fill those pockets created by the iLM bending your 12th or 13th gen CPU. I only figured this out because when I bought the iLM and it came with this exact thermal paste I had the best result, even compared to MX4 which is what I have been using for the last 3 years."

 

[Phaaze88]

The purpose of paste is to fill in pockets. Excess is squeezed out from underneath.
By how much does this 'best result' change things?

That review doesn't really say much.

 

[consptheory77]

The purpose of paste is to fill in pockets. Excess is squeezed out from underneath.
By how much does this 'best result' change things?

That review doesn't really say much.

There seems to be a subjective preference by users for or against various levels of viscosity.

I know I for one don't want to try and clean up excess at the point I ever have to delid, though too much would always be better than too little, right?

I'm not sure that it can be demonstrated the thicker texture is "better" for the CPU, the user might feel it yields a more appropriate spread. Or perhaps believes that low viscosity paste gets "pumped out of its place", resulting in excess spread.

MX4 has been described by one user as "watery" while MX5 has been compared by another user to "glue".

It's also claimed that MX4 requires no curing time but MX5 does.

And that both the Thermalright and Noctua pastes effectively dry out after 2 to 3 years.

 

[Lafong]

I've been using the same application of Noctua paste for 7 years with no change in temps.

Some paste is like mayonnaise; others more like peanut butter..very tacky.

Any paste I've tried has been within the margin of error...little difference.

I wouldn't worry about it unless that last 3 degrees is going to matter a lot......which it might for an extreme situation or for a high-anxiety user.

 

[consptheory77]

I've been using the same application of Noctua paste for 7 years with no change in temps.

Some paste is like mayonnaise; others more like peanut butter..very tacky.

Any paste I've tried has been within the margin of error...little difference.

I wouldn't worry about it unless that last 3 degrees is going to matter a lot......which it might for an extreme situation or for a high-anxiety user.

I am a mid-anxiety user, lol

 

[consptheory77]

How much of a difference will my choice of thermal paste actually make?

"A stable overclock should not be contingent on whether or not you have the right thermal paste. What matters more is your raw cooling power, onboard VRMs, PSU capacity, and the ultimate ability of the chip you’re trying to overclock.

There are ultimately a lot of factors that go into deciding whether or not to replace your thermal paste, but the most important one is just whether or not your system is thermal throttling and performing noticeably worse than it used to. For users who aren’t hardcore cooling enthusiasts, we’d recommend reserving the thermal paste replacements for only when they’re strictly necessary, like when your decade-old laptop is starting to reach 90+ degrees Celsius in regular usage."

 

[consptheory77]

I did wind up trying the Thermalright TF9, it went on like bubble gum, it would not spread, I said screw this, got out my eight year old syringe of MX-4, it spread easy, temps on new CPU are just fine.