Arctic Announces MX-7 Thermal Interface Material

Surprisingly the MX-7 seemed less thick and easier to get out of the tube than the MX-6 that I bought at the end of November last year. Tested running Cinebench R23 for 10 minutes, my CPU cooler is a Thermalright PS120SE with Noctua NF-A12x25 G2 fans. Ambient temperature was slightly cooler when I tested the MX-7 so it's within margin of error, not worth upgrading.

MX-6.png
MX-7.png
 
mark 11:11 said:
i have to admit to being tempted by the mx7 myself ive got a tube of mx6 sat here but from what i gather the mx6 needs repasting fairly often
and as stupid as this sounds 0.5c lower temps is actualyl more than i was expecting i get quiete excited over a 0.2c difference byut ive got an obsession with ott aircooling and seeing
how far i can get on iar which is kinda a pointless side quest of a hobby in of itself
i also need to look into getting a better keyboard at somepoint
but the only ones ive used that i actually like i bought the last box of ten that they had
and ive got one on the desk im using and one in the box left the other 8 have been destroyed by now LOL!
keyboards are a really personal subject ive noticeed
i prefer mine to have as little resistance as possible so a laser projected faux touch screen type thing would be perfect lol
Click to expand...

If you bought 10 of my keyboard, it'd still be cheaper than 128GB of DDR5 :cry:
 
Eddie99 said:
not worth upgrading.
Click to expand...

The difference MX4-MX6 only really becomes apparent under load on high wattage CPUs, possible MX6-MX7 is the same - a 7800X3D doesn't chuck a lot of heat at it relatively speaking. I wouldn't even bother using MX6 over MX4 personally on anything other than 14700/14900 or AMD 79xx and 99xx CPUs.
 
You should have run Prime95 small FFT test.

That gets CPU max temps to places nothing else does in my opinion.

No real world scenario will ever get it that high but it’s great for seeing worst case scenario.

Gets my 5800X3D with a Peerless Assassin up to and bouncing of the 90 degree limit in no time.
 
Interesting that he did the spread method, seemingly without too much issue, after Arctic say you can't spread it.
I'm not a fan of the cross method because like he said in the video it does seem that paste will ooze out the edges, which I like to minimize. I used to do the dot in the middle method but that can miss the corners if you don't put enough on (and if you do it can again ooze). I've recently started spreading (sometimes) but I'm not sure what I'll try with this paste.
 
People have been arguing for years that the spread method could cause microscopic air bubbles, not sure if it has ever been proven?
 
Last edited:
Ice Tea said:
People have been arguing for years that the spread method could cause microscopic air bubbles, not sure if it has ever been proven?
Click to expand...
No, but I've also never seen compelling evidence that spread was superior. Or any method was tangibly superior. It's always been margin-of-error differences. Too little TIM is much more of an issue than how you apply it.

There is of course tidiness and peace of mind reasons, especially with AM5's weirdo IHS, but the better results I've seen have been more to do with ensuring hotspots have decent coverage, especially with the move to P-core and chiplet designs.
 
I got mx4 and due for a repasting, was eying up mx6 but seen this mx7 now.
Worth getting? Not seen any reviews yet how much improvement over mx4?
 
You must log in or register to reply here.
Back
Top Bottom