Thermalright tfx. New top 1 paste?

Shenhua · Apr 18, 2019

Looks like it..... it has higher stats than the kryonaut and seeing some flawed reviews, in all of them seems to be better than the kryo.

Srry if there's another thread open for it.

Enviado desde mi Redmi Note 3 mediante Tapatalk

doyll · Apr 20, 2019

I did a couple of tests and found Thermalright's new TIM to be very good. Applies and spreads easy, and gives great heat transfer.

Owterspace · Apr 20, 2019

^^

Did you use TFX or TF8? Id like to try either of them. Leaning more towards TF8 as I'm on a bit of a budget with an upcoming move. But if its worth it to get he TFX well you know.. I would just quit listening to my sensible side as usual. I did notice the performance of my CF3 to be diminishing slightly. I read because they want you to replace it every 3-6 months or something like that. I'm at that point. I'm sure under normal circumstances it would be fine, but I've been hitting Linpack Extreme fairly hard lately.

doyll · Apr 20, 2019

Owterspace said:
^^

Did you use TFX or TF8? Id like to try either of them. Leaning more towards TF8 as I'm on a bit of a budget with an upcoming move. But if its worth it to get he TFX well you know.. I would just quit listening to my sensible side as usual. I did notice the performance of my CF3 to be diminishing slightly. I read because they want you to replace it every 3-6 months or something like that. I'm at that point. I'm sure under normal circumstances it would be fine, but I've been hitting Linpack Extreme fairly hard lately.

I did some quick testing of both and found them both to be very good. Actually I tried TF5, TF8 and TFX. I like TFX best, but all are quite good.

Where did you read that garbage about needing to replace every 3-6 months? I have 2 systems running CF3 that were mounted 3 and 5 years ago with no heat transfer problems yet.

Alastair · Apr 20, 2019

How does tfx spread? Currently using CM Mastergel Nano. But if i can get an extra degree or two Ill take it.

Owterspace · Apr 20, 2019

doyll said:
I did some quick testing of both and found them both to be very good. Actually I tried TF5, TF8 and TFX. I like TFX best, but all are quite good.

Where did you read that garbage about needing to replace every 3-6 months? I have 2 systems running CF3 that were mounted 3 and 5 years ago with no heat transfer problems yet.

I thought I saw it on the Thermalright.de site, but its down where I am. It must be a mistake, I am probably noticing ambient temps changing thanks to spring.. I notice they play with info from time to time, for instance in the past, I have seen them lower the tdp rating of LGMRT from 320 down to 300 and then back up again. Although at last glance Thermalright.de continued to list LGMRT as 300w, while .com is still listing at 320w..

Ok, well thanks for the tip, I think I will just go ahead and from the TF8, as with my coupon its about 10 bucks on Aliexpress.

I have the same paste on LGMRT same mount for over a year, and I gave it the beans a few times, and its still ok. Holding a 2-3c spread on my x5690. My 3770k just plain runs hot, with a 10c spread over the 4 cores, and a 3c spread between 3 cores.

Alastair · Apr 20, 2019

I really want to buy some tfx from aliexpress. Had i have known tfx existed when i ordered my block i would of ordered some with the block. Now i need thermal compound. But it takes like 8 weeks for something from china to get to me. So now i need to find something worth my time on aliexpress to bring up with some tf8 or tfx.

Shenhua · Aug 4, 2019

Tested the TFX against the kryo.

1800x with 1.344v at 3.9ghz.
Cooler nh-d15

Prime95 small fft during 5-6min, then reset prime95 and hwinfo64 at same time, then results took at 5:30mark.

Application method: small pea at the center
Fan curve stock (i know I should've left it static, but this wasn't a benchmark, but more of "let's try it", and anyway i didn't expected a big variance in temperature, so fans RPM were very similar).

Kryo with 26.3°C ambient temp got 72.4°C
TFX with 26.2°C ambient temp got 69.0°C

Considering margin of error, i guess the TFX is better, but not by a margin that can be considered relevant.

Hwinfo64 pics http://imgur.com/a/WQIqpwA

maltamonk · Aug 4, 2019

Shenhua said:
Tested the TFX against the kryo.

1800x with 1.344v at 3.9ghz.
Cooler nh-d15

Prime95 small fft during 5-6min, then reset prime95 and hwinfo64 at same time, then results took at 5:30mark.

Application method: small pea at the center
Fan curve stock (i know I should've left it static, but this wasn't a benchmark, but more of "let's try it", and anyway i didn't expected a big variance in temperature, so fans RPM were very similar).

Kryo with 26.3°C ambient temp got 72.4°C
TFX with 26.2°C ambient temp got 69.0°C

Considering margin of error, i guess the TFX is better, but not by a margin that can be considered relevant.

Hwinfo64 pics http://imgur.com/a/WQIqpwA

3.4c is about the difference between kryonaut and bundled CM paste that comes with 212s. In other words if 3.4c is margin of error then all thermal pastes are basically the same.

ozlay · Aug 4, 2019

Also a lot more expensive. I can get 11 grams of kryo for $28 or 6 grams of TFX for $40.

Shenhua · Aug 4, 2019

ozlay said:
Also a lot more expensive. I can get 11 grams of kryo for $28 or 6 grams of TFX for $40.

For me it's 18€ vs 21,5€. 6g for both of them

Enviado desde mi Redmi Note 3 mediante Tapatalk

AlphaC · Aug 4, 2019

TFX is on Amazon but at $20 for 2g it's really expensive.

https://www.amazon.com/Thermalright-TFX-2g-14-3W-m-K/dp/B07MYJK8ZS

I suspect for most people TF8 is a decent option at $11 for 2g.
https://www.amazon.com/Thermalright-Compound-Performance-Heatsink-Interface/dp/B07R6H6N6F/

From Thermaright's site, the TFX has 14.3 W/mK while TF8 has 13.8W/mK.

Kryonaut is 12.5W/mK for comparison, but as it's not the same vendor you have to be skeptical. Gelid GC Extreme is 8.5W/mK ($15 for 3.5g) yet typically performs close, Arctic MX-4 is also 8.5W/mK (budget option 4g for $7).

See https://www.hwcooling.net/en/the-test-of-27-thermal-compounds-part-2-en/2/ and https://www.igorslab.media/waermeleitpasten-chart/5/

speed_demon · Aug 4, 2019

For comparisons sake Thermal Grizzly Conductonaut is said to be 73 W/mk according to the TG site - https://www.thermal-grizzly.com/en/products/26-conductonaut-en

But that's a bit like comparing apples and oranges. And we all know that's not a true comparison. Each has it's own pros and cons.

For my normal paste I use MX-4 but when mine runs out I'm jumping on the TF8. Looks to be an excellent performer for the price.

ntuason · Aug 4, 2019

Looking to replace Conductonaut (Just ran out,) sucks cants even get Thermaright's new TIM in Canada.

ozlay · Aug 4, 2019

I wonder how the tf8 compares to the kryo.

Owterspace · Aug 5, 2019

ntuason said:
Looking to replace Conductonaut (Just ran out,) sucks cants even get Thermaright's new TIM in Canada.

I got mine from aliexpress. A 6g tube for 25-or 30 beans. It came in about a week or so. You should be able to get it from amazon too, u.s. not Canada.

Its pretty good stuff, probably the best I've used. I will order TFX next.

GRABibus · Sep 15, 2019

Where can we get easily the TFX ?
Except on amazon with 2 Japanese sellers and eBay from Korea, nothing in Europe ??

Kommando Kodiak · Oct 17, 2019

I bought some and did a trial run in my laptop that was routinely hitting 100C at full fan speed thermal throttling. Not the best test for the paste but again a trial run and I'm glad I did this because holy crap is this stuff the hardest paste to spread(which matters to me). The consistency and color is that of your standard grey clay for throwing in art class. It would just stick to the spatula rather than the die.

So performance wise it halved my max temps and my fans remain at the minimum levels but this *isn't* the best test for this paste as this laptop barely has any heatsinks.

doyll · Oct 18, 2019

Kommando Kodiak said:
I bought some and did a trial run in my laptop that was routinely hitting 100C at full fan speed thermal throttling. Not the best test for the paste but again a trial run and I'm glad I did this because holy crap is this stuff the hardest paste to spread(which matters to me). The consistency and color is that of your standard grey clay for throwing in art class. It would just stick to the spatula rather than the die.

So performance wise it halved my max temps and my fans remain at the minimum levels but this is the best test for this paste as this laptop barely has any heatsinks.

I hate thick TIM. I know they are often rated as better heat tranfer, but getting a good seat with TIM only filling voids between direct contact surfaces rather than putting a layer of TIM between them makes a huge difference in temps. Layer of TIM print will result in 10+c higher CPU temps than a lower rate TIM with direct contact print.

I wonder how much of change in temps was because your old paste was not working as well as it did when laptop was new.

ciarlatano · Oct 18, 2019

Kommando Kodiak said:
I bought some and did a trial run in my laptop that was routinely hitting 100C at full fan speed thermal throttling. Not the best test for the paste but again a trial run and I'm glad I did this because holy crap is this stuff the hardest paste to spread(which matters to me). The consistency and color is that of your standard grey clay for throwing in art class. It would just stick to the spatula rather than the die.

So performance wise it halved my max temps and my fans remain at the minimum levels but this is the best test for this paste as this laptop barely has any heatsinks.

Well.....which is it?

doyll said:
I hate thick TIM. I know they are often rated as better heat tranfer, but getting a good seat with TIM only filling voids between direct contact surfaces rather than putting a layer of TIM between them makes a huge difference in temps. Layer of TIM print will result in 10+c higher CPU temps than a lower rate TIM with direct contact print.

I wonder how much of change in temps was because your old paste was not working as well as it did when laptop was new.

There was obviously a problem with the old TIM. Given that the best and worst TIMs rarely have more than a 5-6 C difference, we can attribute almost all of the difference to that. So, the test was TFX vs. non-working TIM. While it's good to see that Kommando fixed his laptop, there is really nothing that can be taken from these results.

Thermalright tfx. New top 1 paste?

Shenhua

doyll

Owterspace

doyll

Alastair

Owterspace

Alastair

Shenhua

maltamonk

ozlay

Shenhua

AlphaC

speed_demon

ntuason

ozlay

Owterspace

GRABibus

Kommando Kodiak

doyll

ciarlatano

Top Contributors this Month

Recommended Communities