Arctic MX-5 Thermal Compound Listed Ahead of Launch

[Axman]

LTT video title next week:

With a 40-minute follow-up sermon from Tech Jesus that could be summed up with "It doesn't."

 

[chithanh]

Top of the list was "Water based lubricant".

They didn't test long term.

IgorsLab indeed warned against using any of these long-term, but said there can be a specific scenario where one makes sense. He tested the lubricant for up to a work day (8 hours) and it was fine. Mind that it is electrically conductive and take precautions. The gearbox oil has been tested for a few weeks. For a temporary benchmark build, or while you are waiting for your order of TIM to arrive, they will probably be ok.

 

[cybereality]

Not sure about you guys, but I sure as hell ain't putting lube in my computer.

I'll wait for my TIM to arrive, but I have loads of free ones that come with coolers around if I really need.

 

[Armenius]

IgorsLab indeed warned against using any of these long-term, but said there can be a specific scenario where one makes sense. He tested the lubricant for up to a work day (8 hours) and it was fine. Mind that it is electrically conductive and take precautions. The gearbox oil has been tested for a few weeks. For a temporary benchmark build, or while you are waiting for your order of TIM to arrive, they will probably be ok.

Definitely would not use anything water-based in my PC that isn't contained. I'd be interested to see what the CPU looks like with the gearbox oil after weeks of use. I bet it's going to look and smell amazing.

 

[pendragon1]

first tube came early BUT im off until next monday so ill have to wait to get back to work for a comparison...
box was a bit smooshed and qr codes on it dont seem to be active either.

 

[Jandor]

MX3 and MX4 are both good. MX3 is very greasyy. The good thing is those thermal paste last very long. Means if you get off the heatspreader in 10 years from now, you will get still greasy paste not some solid stuff as with other thermal pastes, including arctic silver.

 

[cybereality]

MX3 and MX4 are both good. MX3 is very greasyy. The good thing is those thermal paste last very long. Means if you get off the heatspreader in 10 years from now, you will get still greasy paste not some solid stuff as with other thermal pastes, including arctic silver.

Not sure about that. I've been using MX4 for a while, and I recently rebuilt my spare computer and it was brittle flakes. I think I pasted it 4 years ago. I thought it would last longer.

In any case, I switched to thermal pads. So far, so good. They are inert so should last forever.

 

[MrSneis]

This mx 5 stuff is very thick and is also blue.

 

[w4ffles]

I have a tube of MX-5 arriving on Saturday. I'm sure my Arctic Silver 5 is still fine to use but I wanted something non-conductive for re-pasting my graphics card.

 

[pendragon1]

got my second tube this morning. will test it when im back to work next week.

 

[Falkentyne]

I got my 50g tube of MX-5 yesterday. Haven't tested it since I don't think it beats Thermalright TFX and I don't want to tear down my RTX 3090 or my AIO to try it right now.
Maybe I'll toss it on my laptop MXM GPU or something (which also has TFX on it)

 

[pendragon1]

fresh AS5:

View attachment 323685

vs fresh mx-4:

View attachment 323686

if/when i get some mx5 ill compare it too.
ps: those are 45min into the test, it counts down from 60min. on a i5-2400 with stock dell heatsink.

had a chance to test it today, room temp is 2c higher than the above tests. so a slight improvement. putting it in a 2013 mac pro next....

 

[w4ffles]

I used MX-5 on my dad's 11700K and a Raijintek Themis Evo. I haven't done any kind of stress testing or benchmarking but idle temps were good. I was NOT expecting it to be such a thin/runny thermal paste. I knew they said it was more spreadable but dang.

 

[pendragon1]

re-ran the test today while at the original room temp and cores were 67-70. so there is maybe a 2-3c difference between 4 and 5. accidentally deleted the pic though and dont feel like running the test again...

 

[pendragon1]

did another test on a newer dell system with a 4th gen I5, same ~2c difference. ive done 6-8 systems with it now, desktops and laptops and havent reached the first indicator hole either, so thats a plus too.

 

[OutOfPhase]

Thanks for the reports Pen.

That sounds pretty solid, as MX-4 was actually a really nice goop on it's own. Improvements are cool - assuming it still has that longevity I loved about MX-4 (STILL have an overclocked 3700 running it just fine). Not sure how to verify that yet.

 

[pendragon1]

Thanks for the reports Pen.

That sounds pretty solid, as MX-4 was actually a really nice goop on it's own. Improvements are cool - assuming it still has that longevity I loved about MX-4 (STILL have an overclocked 3700 running it just fine). Not sure how to verify that yet.

no prob!. yeah the 8 year claim is gonna be a hard, or long, one to prove. most dont have systems that long and the ones ive used it on will age out or die long before then.
ive also put it on my 5700 when i modded my g12. i took it apart to add some vram heat sinks and one thing i noticed while cleaning it up, is that when you hit it with iso it kinda solidifies/firms up and you can kinda push it together and scoop it up all at once. its kinda weird...

 

[Mizzer]

Put some on my 5900x and Arctic LF 2 420. Very easy to work with.

 

[OutOfPhase]

a hard, or long, one

<eyebrow> Go on...

But yeah, what I've loved about this series is just how resilient it is to time / conditions. I have even older systems on MX-2, but those are full stock and could be cooled with cheddar / cat fur probably. My own personal data point is that 3700 which is overclocked within an inch of its life has been running the same paste and cooling since launch. Still zero problems, and it is serving a bunch of ungrateful kids a wonderful minecraft server, which I may or may not alter slightly for my own entertainment randomly.

Tangent - parents which don't do this, my lord you're missing out on some subtle mind-fun. I digress.

The performance I've experienced over the years - pretty damn cool. Literally!

I may be in the minority here, I buy things to last a long time, and not so much absolute thermals right-now. Not [H}, or different [H}?
I have a long tail of kids using hand-me-downs, and I don't want to fix/update them every iteration. I have been VERY well served by AC and the MX line to this point. In particular, the MX-4 has been bulletproof. Maybe not tops in ideal cases, but just freaking works, for years upon years.

YMMV, and all that. This is my mileage.

 

[GoodBoy]

Just built a new system about 2 weeks ago, used an IC Graphite pad. Couldn't be happier. Rated at 35W/m-k, which is more than triple what the high performance pastes are rated at. Arctic silver 4 is rated at 8.5W/m-k, the other Arctic Silvers were all less than 9 iirc. The liquid metals are rated 36 to 38W/m-k but are a huge pain in the ass, incompatible with copper or aluminum, conductive, and need reapplied annually (fuck that). The IC Graphite pads are also conductive, so it isn't surprising that the performance rivals the liquid metals. Compared to normal compounds, this pad has been very impressive.

i9-10900x air cooled with a Noctua nh-d15s (single fan), prime95, temp settles at 62C. Going to mount a second fan soon as I get the parts.

Probably some reviews someplace, you all owe it to yourselves to check it out.

 

[pendragon1]

Just built a new system about 2 weeks ago, used an IC Graphite pad. Couldn't be happier. Rated at 35W/m-k, which is more than triple what the high performance pastes are rated at. Arctic silver 4 is rated at 8.5W/m-k, the other Arctic Silvers were all less than 9 iirc. The liquid metals are rated 36 to 38W/m-k but are a huge pain in the ass, incompatible with copper or aluminum, conductive, and need reapplied annually (fuck that). The IC Graphite pads are also conductive, so it isn't surprising that the performance rivals the liquid metals. Compared to normal compounds, this pad has been very impressive.

i9-10900x air cooled with a Noctua nh-d15s (single fan), prime95, temp settles at 62C. Going to mount a second fan soon as I get the parts.

Probably some reviews someplace, you all owe it to yourselves to check it out.

so hows the mx5 working for you?

 

[cybereality]

Yeah, the IC Graphite Pad is where it's at. I'm getting 32C idle temps on my 4790K, which is great.

Under load it does go higher, 65C while playing Cyberpunk and absolute highest temp I saw was 80C but that was doing a burn test.

So not the most amazing temps in the world, but more than good enough for what it is. It will last forever, you can reuse without a mess, and you never have to worry.

That build I replaced the HSF and the MX-4 had dried to flakes (only after 4 years). With the Graphite Pads, you set it and forget it. Great investment, don't think I will use paste ever again.

 

[pendragon1]

Yeah, the IC Graphite Pad is where it's at. I'm getting 32C idle temps on my 4790K, which is great.

Under load it does go higher, 65C while playing Cyberpunk and absolute highest temp I saw was 80C but that was doing a burn test.

So not the most amazing temps in the world, but more than good enough for what it is. It will last forever, you can reuse without a mess, and you never have to worry.

That build I replaced the HSF and the MX-4 had dried to flakes (only after 4 years). With the Graphite Pads, you set it and forget it. Great investment, don't think I will use paste ever again.

thats nice but off topic.

 

[SmokeRngs]

Yeah, the IC Graphite Pad is where it's at. I'm getting 32C idle temps on my 4790K, which is great.

Under load it does go higher, 65C while playing Cyberpunk and absolute highest temp I saw was 80C but that was doing a burn test.

So not the most amazing temps in the world, but more than good enough for what it is. It will last forever, you can reuse without a mess, and you never have to worry.

That build I replaced the HSF and the MX-4 had dried to flakes (only after 4 years). With the Graphite Pads, you set it and forget it. Great investment, don't think I will use paste ever again.

I can't say I've ever heard of Arctic paste being dried to flakes before. I don't have any long term experience with MX-4 as I only got it a couple years ago but I've had multiple applications of MX-2 last 5+ years and the reason for having checked the paste had nothing to do with the paste. The paste also looked like it had been applied yesterday.

The fact that the MX pastes work well and work for a very long time is the reason why I keep going back to them just like a previous poster. I don't need the absolute bestest performance out of a paste but I do need one that is going to last a long time to the point where I can literally forget about it. It also doesn't hurt that MX pastes tend to be very affordable.

I probably would have ordered some MX-5 when I was getting the parts together for my upgrade but availability was low at the time and prices seemed a bit screwy because of it. Unless Arctic comes out with even better paste soon I'll end up with some MX-5 eventually.

 

[cybereality]

Sorry, I just double checked and it was Arctic MX-2 I used, and it was 5 years ago. In any case, it was all dried and flaky.

 

[w4ffles]

I just used MX-5 on a 5600X and a Scythe Fuma 2 in a Fractal Meshify 2 Compact. Idles at 31C and running Prime95 Small FFTs it reached 55.6C in a 73F/22.8C room. I love how quiet the Scythe Fuma 2 is.

 

[Armenius]

Ja, MX-2 always dried out. MX-4 always looked like it had just been applied even after 5-6 years.

 

[Falkentyne]

Just built a new system about 2 weeks ago, used an IC Graphite pad. Couldn't be happier. Rated at 35W/m-k, which is more than triple what the high performance pastes are rated at. Arctic silver 4 is rated at 8.5W/m-k, the other Arctic Silvers were all less than 9 iirc. The liquid metals are rated 36 to 38W/m-k but are a huge pain in the ass, incompatible with copper or aluminum, conductive, and need reapplied annually (fuck that). The IC Graphite pads are also conductive, so it isn't surprising that the performance rivals the liquid metals. Compared to normal compounds, this pad has been very impressive.

i9-10900x air cooled with a Noctua nh-d15s (single fan), prime95, temp settles at 62C. Going to mount a second fan soon as I get the parts.

Probably some reviews someplace, you all owe it to yourselves to check it out.

The IC pads are *terrible* if you're overclocking. They are 6C worse than MX-5 on a nice flat (clear imprint so good mounting pressure) r9 290X video card, and more than 10C worse (reached 100C so that was it) on an overclocked 9900k vs Kryonaut (using an old Venomous-X heatsink). They are good for stock mount and forget systems but that's ALL they are good for.

They would probably only reach their w/mk rating on the *Z Axis* by extremely high (>50 PSI) mounting pressure and you aren't getting that.

 

[pendragon1]

The IC pads are *terrible* if you're overclocking. They are 6C worse than MX-5 on a nice flat (clear imprint so good mounting pressure) r9 290X video card, and more than 10C worse (reached 100C so that was it) on an overclocked 9900k vs Kryonaut (using an old Venomous-X heatsink). They are good for stock mount and forget systems but that's ALL they are good for.

They would probably only reach their w/mk rating on the *Z Axis* by extremely high (>50 PSI) mounting pressure and you aren't getting that.

their w/mk numbers are way out of wack i think they are using the numbers from the raw elements used not when its made into a pad or paste.

 

[Falkentyne]

their w/mk numbers are way out of wack i think they are using the numbers from the raw elements used not when its made into a pad or paste.

Their w/mk numbers are apparently based on "X/Y" Axis (meaning, through the pad itself lengthwise).
What seems to matter more is Z axis transfer, which is MUCH lower, and even then, those seem to be based on theoretical numbers with massive mounting pressure.
Also, the pads I gave temps for above are the newere Panasonic EYG-R pads, which are more compressible than the Soft PGS+ (EYG-S) pads that the IC pads are a rebrand of (I bought the official Soft PGS+ pads from Digikey years ago, to compare them to the free samples I got from IC, when they were giving out their pads to beta testers over on Notebookreview forums).

I know some people here think I'm pulling stuff out of my ass, but I put in my time.

 

[Jandor]

Sorry, I just double checked and it was Arctic MX-2 I used, and it was 5 years ago. In any case, it was all dried and flaky.

Yes. In fact MX3 is already good at staying stable without drying out. It's only very greasy compared to MX4. Cleaning from MX3 is a real pain. Not sure there is big difference on every the other side.

 

[Sir Beregond]

Nice. MX-4 was always a go-to for me and I have a fairly new tube of it right now. It's always lasted a super long time without drying out for me. NT-H1 was the other one I liked, but don't ever see it locally anymore. Was happy with the performance of Kryonaut, but its expensive for how much you get and definitely dries up faster than MX-4.

 

[learners permit]

I'm gonna try some bacon cuz it smells good when it's cooking.

 

[GoodBoy]

The IC pads are *terrible* if you're overclocking. They are 6C worse than MX-5 on a nice flat (clear imprint so good mounting pressure) r9 290X video card, and more than 10C worse (reached 100C so that was it) on an overclocked 9900k vs Kryonaut (using an old Venomous-X heatsink). They are good for stock mount and forget systems but that's ALL they are good for.

They would probably only reach their w/mk rating on the *Z Axis* by extremely high (>50 PSI) mounting pressure and you aren't getting that.

30's idle, 62C all cores at 100%. These pads are really not that bad.

I don't buy the x-y-z axis... how likely are these pads "aligned" at the molecular level such that heat only moves in one axis? pfft.

Soon as I kill prime95, temp drops to 40 in only a couple seconds... heat is moving through the material very readily, and I am only air-cooling.

I'll get around to overclocking at some point.
It was some time ago that I tested it, but the old cpu/mobo, an i7-6850k, prime95 took that into the mid 80'c, maybe higher. On that cpu I used arctic silver "ceramique". Nearly as good, if not just as good as the "silver" arctic silver compounds.

 

[cybereality]

I did watch the videos on the thermal pads, Tech Jesus did a few long reviews.

The result was that the pads were within a few degrees hotter than the best paste.

Maybe if you are extreme overclocking that makes a difference, but for normal builds they are more than fine.