thermal paste Which Thermal Paste to buy and apply (Traditional and Liquid Metal)

[yoodog]

On 10/30/2022 at 6:05 PM, Etern4l said:

My .02: Phobya Nanogrease Extreme. Immediately blows Kryonaut out of the water, both on a laptop and desktop.

Tried CoolerMaster CryoFuse Nano (14 W/mK) sourced from Amazon US on a GPU, looks like a waste of time - minimal improvement over stock.

damm really? that is a rare case to blow it away significantly. TF7 looks strong and not as hard to spread as TFX.

have you tried CYY 157?

 

[Etern4l]

1 hour ago, yoodog said:

damm really? that is a rare case to blow it away significantly. TF7 looks strong and not as hard to spread as TFX.

have you tried CYY 157?

 

 

Actually, on laptops most people would agree Kryonaut sucks. I found it underwhelming on a desktop CPU too, although I didn't have a contact frame back then so maybe the contact was bad. Even if it worked fine on a CPU, its longevity is limited, so I would cross it off on that basis. Potentially a similar concern about SYY 157. The 15.8W/mK spec is appealing, but it's very cheap. Would it last? There are some doubts. I guess at this price point it's worth a shot, thanks for the recommendation.

 

I also wouldn't mind trying the new Alphacool Apex if I could get my hands on any.

 

 

 

[jaybee83]

11 hours ago, Clamibot said:

 

A little heads up: the paste is extremely hard to spread. Don't bother doing this since the paste is very thick. Just squash it down using your heatsink and use the mounting force to spread it for you.

 

People who say Nanogrease sucks say it sucks because it's so thick and hard to spread. There's no need to spread thermal paste yourself though, and I never understood why people do that. The heatsink will spread the paste for you when you screw it in.

 

11 hours ago, Etern4l said:

 

I always spread it no probs, without preheating. Takes a few min. Clearly recall Phobya packaging has a recommendation to spread the paste, comes with a spatula for a reason. 

 

so far ive been the type to let the heatsink do the spreading, ive only dabbled once or twice in manual spatula spreads. im guessing its just a different application philosophy. GN for example seem to prefer the manual spatula spread, so might produce more consistent results in the hands of an experienced technician.

i imagine the spatula method would allow for varying heatsinks with differing surface properties (porosity & bend), as well as mounting pressure. whereas the heatsink spread method is easier to implement if u stick to the same heatsink and mounting pressure. just a hyothesis on my side though.

in the end, such discussions will likely have much less of an effect than the utilized paste or cooler itself 😄

[User]

I'm on a 4G tube of SYY 157 now and so far it's looking to be the best results I've seen yet with a direct-die laptop. Core to core load temps are only 1-2C whereas before I was anywhere from 10-15C between cores.

 

Pump-out with pastes like NT-H1, Kryonaut, Nanogrease, MasterGel Maker, etc all happened after weeks or months. I'd remove the heatsink to see a big empty spot in the middle of the die where there was no paste, and it was all bordering the sides. Load temps gradually crept from low 80's to 100C triggering PROCHOT flags/behavior.

 

SYY 157 spreads thick and has a 15 minute "cure" time before reinstallation of your heatsink. Like I mentioned the core to core temps are so much better than my thinner pastes offer and the load temps are in the mid-70C range which is better than any paste I've tried to date. For $10 thru Amazon I am impressed, and it came with cleaner and multiple application tools.

[cylix]

I also used for the last application on the Ryzen 7900x SYY 157. Until now it looks good.  Max was 78C and the avg is 42.7.

[jaybee83]

On 11/1/2022 at 12:26 PM, cylix said:

I also used for the last application on the Ryzen 7900x SYY 157. Until now it looks good.  Max was 78C and the avg is 42.7.

ambient temps? what kinda load tests did u do?

[raptorddd]

On 10/31/2022 at 12:16 PM, KING19 said:

 

A paypal account is easy to create and its a safer option than the other payment options on their site.

 

Also never use a debit card when buying things online especially when buying things from international websites for obvious reasons. Always use a credit card when doing online shopping.

thanks for the heads up.

ive had paypal many many years ago.  last time i tried to associate my card i couldnt. so i gave up.

and already ordered and decided to try again and this time i did associated card and account. so thanks for the tip. by the way paypal gave me around 10 dollars free for next purchase.

 

let see how long it takes to arrive to mexico.

[cylix]

4 hours ago, jaybee83 said:

ambient temps? what kinda load tests did u do?

Room temperature around 21 22 C, load tests was mostly gaming, around 1 Hour of Battlefield 2042 and right after 2 hours of Cyberpunk 2077, so 3-3,5 Hours of gaming. The peak was of course in Battlefield 2042,  in Cyberpunk the CPU is not that hot

[Vasudev]

On 10/31/2022 at 7:31 AM, Rofa1234 said:

Nanogrease is complete crap for my notebook.

Wasn't available in India and I went CM Maker Gel paste.

On 10/31/2022 at 4:49 PM, yoodog said:

guys planning to repast my Mac Book Pro 2019 with 2,3 GHz 8-Core Intel Core i9 + AMD Radeon Pro 5500M 8 GB 
- I checked and based on this photo I see 0 danger in using liquid metal on CPU - Right big square! there are no resistors or any components at all on CPU near by the die

Am I correct that I SHOULD use liquid metal for CPU - ordered Thermal Grizzly - Conductonaut  1g

For GPU I ordered SYY 157 2022 8g - will heat it up in boiling water and apply thinest (non transparent) layer on GPU + on heat sink semi transparent layer 

Any other paste that would give better thermal transfer in your experience that I should have ordered instead? Why?

If you want a longer lifetime w/o repaste I'd suggest TG Carbonaut or ICD Graphite HC version (new material like carbonaut). Though I have seen 1-2c higher peaks with ICD Graphite HC version but operating temps with ICD HC was better and fans were not spinning most of the time. Carbonaut is fragile and perfect for warped heatsink and conforms better which I use on alienware even after lapping my heatsink surrounded with insulation tapes. ICD HC is a 5 pack version and costs similar to 1 package of TG carbonaut and 2-3x cheaper on amazon even after imported to India.

[Rofa1234]

On 11/1/2022 at 4:01 PM, User said:

I'm on a 4G tube of SYY 157 now and so far it's looking to be the best results I've seen yet with a direct-die laptop. Core to core load temps are only 1-2C whereas before I was anywhere from 10-15C between cores.

 

Pump-out with pastes like NT-H1, Kryonaut, Nanogrease, MasterGel Maker, etc all happened after weeks or months. I'd remove the heatsink to see a big empty spot in the middle of the die where there was no paste, and it was all bordering the sides. Load temps gradually crept from low 80's to 100C triggering PROCHOT flags/behavior.

 

SYY 157 spreads thick and has a 15 minute "cure" time before reinstallation of your heatsink. Like I mentioned the core to core temps are so much better than my thinner pastes offer and the load temps are in the mid-70C range which is better than any paste I've tried to date. For $10 thru Amazon I am impressed, and it came with cleaner and multiple application tools.

 

 

Is it the 2022 version? The pre 2022 SYY-157 was even thicker and better performing on my notebook.

 

Arctic MX-6 is out. According to first tests it is higher viscosity than the MX-4 which is good news. There is a good chance it will perform better on a notebook, hopefully much better in my case considering the MX-4 was several degrees down. I have ordered MX-6 as well as Corsair XTM70 and Kooling Monster KOLD-01. I'm curious how they will perform. 

[jaybee83]

its good to see that after a subjective eternity of no innovations in the thermal paste market, new pastes are coming out left and right!

[JeanLegi]

30XX Series, 40XX Series, Intel 12th and 13th generation.. AMD with Ryzen and 5000 and 7000, RX 6000 and 7000 series...

We need new innovations for cooling the heat of hell out of our systems 😄

[jaybee83]

17 minutes ago, JeanLegi said:

30XX Series, 40XX Series, Intel 12th and 13th generation.. AMD with Ryzen and 5000 and 7000, RX 6000 and 7000 series...

We need new innovations for cooling the heat of hell out of our systems 😄

either that or just use them as space heaters during winter time. thats exactly what im planning to do this winter 😁

[Clamibot]

6 hours ago, jaybee83 said:

either that or just use them as space heaters during winter time. thats exactly what im planning to do this winter 😁

 

Yep, I'm gonna use my computers to heat my room and use the cold air to increase my overclocks. My computer runs faster for useful computing tasks and gaming, and I get my room heated! Dual purpose! This is what you call efficiency.

[jaybee83]

23 hours ago, Clamibot said:

 

Yep, I'm gonna use my computers to heat my room and use the cold air to increase my overclocks. My computer runs faster for useful computing tasks and gaming, and I get my room heated! Dual purpose! This is what you call efficiency.

u get me! 😂👍😁

[Etern4l]

I gave the SYY-157 a go. Good first impressions, I think it pretty closely matches Phobya Nanogrease Extreme. The applicator kit is quite handy. Keeping it on the CPU. Question is: how long will it last?

[jaybee83]

10 hours ago, Etern4l said:

I gave the SYY-157 a go. Good first impressions, I think it pretty closely matches Phobya Nanogrease Extreme. The applicator kit is quite handy. Keeping it on the CPU. Question is: how long will it last?

i find it weird and fascinating that such a "no name brand" comes out with such a high performing paste. plus the name, marketing and also pricing(?) are rather indicative of some trashy bulk paste, isnt it?

[Etern4l]

18 minutes ago, jaybee83 said:

i find it weird and fascinating that such a "no name brand" comes out with such a high performing paste. plus the name, marketing and also pricing(?) are rather indicative of some trashy bulk paste, isnt it?

 

I was quite skeptical too (and still am a bit suspicious, but many people claim lifetime is good too). Well, it's a Chinese paste, so things can get cheap. Phobya is German I believe. It seems rather like the 12/13th gen contact frames. Der8auer (invented?) made one costing $30+, Thermaltake made a $7 or so version. Thermaltake version actually outperforms Der8auer's according to Gamers Nexus.... (and works wonders for me).

 

It's tough out there, but it is what it is. As an aside, unless something is patented, it can be copied freely (not suggesting the SYY paste is a copy of Phobya NGE, although they look similar).

 

TBH I prefer this to a bit bogus products with redundant packaging such as some of the recent pastes released by name brands.

[Vasudev]

On 11/6/2022 at 1:35 AM, jaybee83 said:

either that or just use them as space heaters during winter time. thats exactly what im planning to do this winter 😁

In fact my AW BGA made my room so hot even though a heavy rainfall was forecasted. Ambients shot up to 35C at night.

@Rofa1234Let me know if new paste must be added to OP.

i think PCT should be standard paste on all devices and normal paste can't get these mini-Sun/star.

[jaybee83]

13 hours ago, Vasudev said:

In fact my AW BGA made my room so hot even though a heavy rainfall was forecasted. Ambients shot up to 35C at night.

@Rofa1234Let me know if new paste must be added to OP.

i think PCT should be standard paste on all devices and normal paste can't get these mini-Sun/star.

no surprise there my man, ill never forget those 48C days in Hyderabad, just insane!

[Papusan]

Test : Arctic MX-6 Thermal paste overclocking.com | 9 november 2022
https://overclocking.com/test-arctic-mx-6-thermal-paste/5/

 

The remaining question will be... Will also MX-6 suffer from uneven batches? Arctic as you know had to scrap MX-5 into EOL status for the mentioned problem.

[Etern4l]

24 minutes ago, Papusan said:

Test : Arctic MX-6 Thermal paste overclocking.com | 9 november 2022
https://overclocking.com/test-arctic-mx-6-thermal-paste/5/

 

The remaining question will be... Will also MX-6 suffer from uneven batches? Arctic as you know had to scrap MX-5 into EOL status for the mentioned problem.

 

Not very impressed by their choice of comparison pastes.

Also, no W/m*K this time. I guess it's not record-beating then. 

Intriguing though, looking forward to further reviews..

[jaybee83]

10 hours ago, Etern4l said:

 

Not very impressed by their choice of comparison pastes.

Also, no W/m*K this time. I guess it's not record-beating then. 

Intriguing though, looking forward to further reviews..

welp, W/m*K doesnt tell us much anyways, varies way too much based on testing conditions. what matters more is actual results vs other pastes under identical testing conditions.

[Etern4l]

3 hours ago, jaybee83 said:

welp, W/m*K doesnt tell us much anyways, varies way too much based on testing conditions. what matters more is actual results vs other pastes under identical testing conditions.

Well, it tells me quite a bit actually, given that the best pastes I have used are 17 and 16 W/m*K. This number likely indicates the maximum value, which pastes achieve at high temp. Manufacturers of lesser pastes like to argue otherwise of course. Der8auer basically admitted they are gaming the system with Kryonaut by quoting the W/mK at optimal (high) temperature, and other manufacturers probably do the same. That's actually fine, given that the optimum is close to operating temperatures at full load. 

[Rofa1234]

W/m*K is mainly a marketing number, it really doesn't tell much and nobody knows if it's real or just a fake number. That's why some manufacturer won't even share it. Is there a W/m*K number for Corsair XTM70? I haven't found. Some of my best pastes have a low W/m*K number, for example Deepcool G40 or Akasa T5 ProGrade Plus and some with a high W/m*K are poor performer.