Let’s first isolate the data into groupings of an individual TIMs and look at the specific mount data.
Arctic Silver 5:
AS5 has a noticeable break-in, as advertised. My guess is that the increase in mounting pressure in this test has probably lead to the slightly lower severity of the break-in. In my opinion, at the end of twelve hours, it is still not fully cured, though it is close.
Arctic Cooling MX-2:
MX-2, despite being noted as “non-curing,” definitely has a break-in period. I estimate it at roughly 3-4 hours in this test (down from 6 hours in the previous tests). The incredible consistency is probably due to how forgiving MX-2 is and because I have so much experience with it (hundreds of mounts).
Cooler Master IceFusion:
Huge break-in, inconsistent performance (largely due to inconsistent consistency of the paste itself, sometimes it’s dry and other times it’s oily), and overall poor performance. Guess this TIM is a bust as a great value for people who use a lot of TIM. Still remains a good reference of how a ‘bad’ TIM performs, however.
First mount I totally botched. I turned the pump back on too soon and IX had only reflowed to about 50% of the IHS, limiting its performance. A great overall showing from IX again–great temps, great consistency (when there wasn’t user error), and no cure time.
X23-7783D has a more noticeable cure than MX-2 in this go-round, meaning it probably is less sensitive to mounting pressure. Great performance again from this paste.
Here we can see a good summary of the performance, Indigo Xtreme is again the winner with Shin-Etsu X23-7783D coming in a close second (and being the best paste). Interestingly, AS5 comes in much further behind than it did last time. Not sure exactly why, it could be difference in IHS/waterblock mating or it could be from the bead method that AS5 recommends for AMD processors. Based on how the data lined up in the last test (where I showed how close the threshold of perfect performance is), I don’t think this is a case of MX-2, X23-7783D, and Indigo Xtreme getting better; AS5 is simply performing worse in this setup than it did in the last test. IceFusion is simply an inferior paste in terms of performance and consistency.
With a simpler view it’s even easier to see just how tightly packed the top three TIMs are compared to IceFusion and even AS5. Maybe it’s the larger IHS, the higher mounting pressure, or the waterblock base mating better to the IHS, but the top pastes come a lot closer to IX than they did last time.
Looking at the best mounts of all the TIMs, we see that Shin-Etsu X23-7783D gets really, really close to Indigo Xtreme. Almost close enough where it’d be impossible for an enduser to see the difference, but not quite (I put that threshold at half a degree). If Shin-Etsu X23-7783D were an inexpensive paste, it’d easily be the best TIM here, but overall you get what you pay for from these TIMs–the best TIM is the most expensive, the best paste is the most expensive, and the worst is the least expensive, with AS5 and MX-2 falling between the extremes.