ARCTIC MX-6 (4 g, incl. 6 MX Cleaner) – Ultimate Performance

As a techie who usually builds their own computers, I’ve known about Arctic for awhile but only recently became a customer after purchasing a tube of MX-5 (then a P12 fan, shortly after). My experience of applying another company’s paste on an install is limited to only a couple of other brands as thermal paste application isn’t exactly one of my pastimes, so I’m going to compare & contrast this w/ MX-5. Besides, what better way to help inform readers whether a product is worth purchasing or not than to scrutinize it against its own predecessor? While partly out of convenience, it’s still a pertinent assessment, especially so when considering this is not only the successor but also the replacement, after all. This review will also discuss what seems to be the consensus “best” paste by overclocking enthusiasts, YouTubers, and “influencers”, etc. (should note this is exclusive to non-conductive TIM, so no liquid metals). A product which despite its YUGE following and noteworthy sponsorship, has a FAR too little discussed caveat (a massive dealbreaker, IMHO, and one that has significant implications for the IHS/die it’s been applied on, and the potential post-install ‘maintenance’ risk it comes with). Also, I don’t have any info in regards to performance data (CPU upgrade so unfortunately no baseline), so if you’re looking for metrics, you’ll have to look elsewhere (Tom’s Hardware is a good start).

After researching way too many thermal paste comparison tests, MX-5 was at or near the top of just about every single one it was included in and for those that it wasn’t (due to being too new), MX-4 consistently ranked as one of the better pastes with a solid if unspectacular showing but when taking into account its age, durability, price and how much newer the products it was competing against were, MX-4 was a compelling package offered at an affordable price. A couple of contenders at the top included Thermal Grizzly’s Kryonaut, and the two offerings from Noctua (H1/H2). Personally, the main appeal to Arctic’s formulas for me was its longevitiy, which they claim lasts as long as 8 years, and the only other option which can even come close to approaching that long, comes from Noctua. Even then, (while still impressive) they could only reach 5 years. It’s no coincidence these two belong to what I consider to be the top three cooling companies in the industry, IMO. In no chronological order, the third is Scythe, a Japanese company known for making EXCELLENT heatsinks and case fans that are quiet (if not silent), and whose Fuma 2 happens to be the cooler I chose in this case (no pun intended), which sat on top of these pastes in my build. While I am def not gonna pay Thermal Grizzly’s ridiculous prices ($10/gram for an application, maybe 2 if conservative) out of priniciple, most importantly was for every 5 star review of Kryonaut (including its appropriately named “EXTREME” version) by someone claiming they achieved their lowest recorded temps with it, there would be a few 1-2 star reviews complaining about it drying out, and even turning hard after removing the heatsink. What was most concerning tho, was the fact that this was happening in under a year, usually within months. Their own product page states “Kryonaut uses a special structure, which halts the drying out process at temperatures of up to 80 Celsius”, while also mentioning “that this thermal grease was created especially for cryogenic applications.” So yeah, even tho they admit this crucial detail themselves, to say that this is a frequently overlooked aspect (judging by the consistent surprise in negative reviews, yet still seeing it recommended by others all too frequently) would be putting it kindly. I don’t mind paying a few extra bucks for a quality product, as evidenced by MX-6’s price increase, particularly compared to MX-4 (at launch this was $10 for the 4g tube of MX-6 + 5 wipes), but there is certainly a limit, and ESPECIALLY so if the product has to be replaced at least ONCE a YEAR. Ain’t nobody got time for that.

While I can unequivocally say that this is better than MX-5, that is a pathetically low bar and hurdle to pass, one that is more indicative of its predecessor’s failures than anything else. I’ve already mentioned why I’m unable to measure performance, but what I CAN compare are the physical characteristics, especially as it pertains to MX-5, and to a lesser degree MX-4, which I haven’t applied on an install but have owned a tube of, and thus have seen its consistency on paper. This MX-6 seems almost like a hybrid of MX-5 and MX-4, but leaning significantly more towards the former. It’s still very sticky, just not AS sticky, which helps with the application. MX-4 was much more akin to a normal paste, smooth and easy to move around. This is a THICK paste, substantially thicker than its predecessors, but still stringy enough to spread around unlike MX-5, yet not nearly as easy as MX-4 (but at least still doable, and can be achieved with some time). It’s hard to gauge MX-5 because it’s simultaneously viscous and watery, which I believe is due to the separation (and reason for its recall) if not entirely then at least partially, which made it weird to spread and resistant to emulsify. Spreading it with the spatula was almost like dipping a recently used paintbrush in water, then attempting to paint a clean surface with it. Or, like using a basting brush (aka pastry brush, TIL), the ones that come with BBQ sauce in those gift sets, to glaze meat but w/ barely any sauce left. And that’s if you’re even able to spread at all (it’s somewhat doable, depending on your experience ofc, but you’ll def need a tool like a spatula or credit card and a lot of patience, along w/ time). Another difference is the color. MX-4 is the typical gray, while MX-5 was like a light, almost turquoise blue. MX-6 return back to the (more pleasant if not dull) gray, a welcome change IMO. Something I noticed and thought was interesting is that Arctic advertises this by stating it has 20% better performance than MX-4, although this could be simply due to MX-5 no longer being in production means the former is the closest product in their lineup to compare with (excluding MX-2). One aspect different from its predecessors is that it does NOT come with the spreading spatula for some reason. Due to the chiplet design of Ryzen on AM4, I used that tool to spread a very thin layer, referred to as the “spread method”, predominantly around the edges of the IHS. Then, I use a miniaturized version of the “X” method, that doesn’t go all the way to the edges, and thus is almost half as small. Finally, I place a decent sized blob (aka the pea method) in the center. I have no scientific knowledge or verifiable proof that this works best, and I’m sure you could remove one of the three options and still obtain good coverage. Everyone has their personal beliefs, opinions and philosophies on thermal paste, but due to the processor design I find this is what works for me, as the previous application had very good coverage after taking it off. AMD actually recommends a pentadot which I’ve used and resulted in decent coverage, and which formed the basis of mine. If it were an Intel (Alder/Raptor Lake) or even a Threadripper, I’d prolly just use the X method due to its rectangular shape and call it a day.

MX-5 was like straight up glue, even during application, and ESPECIALLY AFTER. This was exactly what I feared after install, was the day it had to come off. I ran stress (more like torture) tests, specifically Prime95 until my processor reached 95C and stayed there for about 30 minutes before shutting down and then was very careful removing the cooler, while making sure to twist it side to side before pulling. Yet despite all that, it still took the CPU off with it, so after removing off the cooler, I was still unable to separate the CPU from it using anything, including dental floss (both Glide and regular), eventually warming it with a hair dryer (heat gun should work, just be careful w/ the distance at very high temperatures and move it around) on it for about five minutes or so. I gave it another try w/ the regular floss and finally, I was able to remove the CPU w/o any bent pins.

While I’m mentioning all this in case anyone else reading this used glue (aka MX-5) or is stuck in a similar situation, it’s also for those considering the wipes option for this purchase. So if wondering whether you should purchase 5 of these wipes for about a buck (and ESP so if worried about removing stuck thermal paste that’s been glued onto your cooler)? If it’s safe to assume the fact that you’re contemplating this likely means you have a use for it, then my answer would be an emphatic YES! They’re pretty large and good sized once removed from packaging, have plenty of liquid that won’t immediately dry out so you have a few minutes, and each one goes a long way. Just be warned they have a weird and pugent smell. Personally, something as minor as that is never a reason to avoid a product if it performs better than a non-smelling product, but I always see this mentioned in reviews and even factored into the score, so thought it should be noted. I only used one even after that nightmare of a removal mentioned above, until I remembered there was still some gunk left on my Wraith Stealth after replacing it w/ the Fuma, and opted to open a 2nd one. A single wipe goes a long way, which is why I believe that if you have a need for it, it’s worth purchasing due to the value provided for a single dollar and some change.

Anyway, if you made it all the way here to the end, I hope this was helpful for you. Good luck.