Is buying the latest and greatest TIM really worth it???

[tachi1247]

Curious to see what type of real world temperature differences people are seeing with using the "best" modern pastes. Putting together a new build and trying to decide if it is really worth it to by some new TIM. From my research it looks like Thermal Grizzly Kryonaut is the best stuff going right now and Gelid GC Extreme is close behind.

Found this testing roundup linked in another thread and it only shows a difference of 1.55° K (same as Celsius) between the best stuff available and the tube of arctic silver 5 I have sitting on my desk. That certainly doesn't seem worth it to spend the cash since I'm not doing an extreme overclock or shooting for any records here.

This review is a couple years old and thus doesn't include Kryonaut, but they got a 3° C difference between GC Extreme and AC5. Since Kryonaut supposedly does better now were talking about 4°+ of a difference maybe it's worth upgrading.

This review also gets a 4° C difference however I would say that I trust the source on the first link more than the latter two showing the larger difference.

Then you have this older review that has AC5 performing 4° worse than PK-1 and other common TIM's from ~2011 compared to the 0.5° degree difference on the techinferno review.

I know a lot of this is dependent on how well the sink is mounted and whether or not you achieve a "good mount" but it seems like there is a pretty varied amount of info on what kind of impact a TIM can really have.

So....what has everyone on this forum been experiencing? If we're talking about a 1-2° difference then I'm not going to waste the money, but if we're talking 4-5° then it might be worth it when I factor in the bonus that new stuff is also non-conductive.

 

[ProgramGuy]

I would also like to hear what others are experiencing with their choice of TIM, and their choice of how to apply it. Some use the "pea" method while Thermal Grizzly recommends spreading it. IMHO... it makes more sense to spread it as it would cover the entire surface, compared to the "pea" and getting a circular pattern on a squarish item.

 

[Mandrake4565]

Worth is only something you can determine tachi1247. As you can see in the links you posted the difference between the very best "non liquid metal" and somewhat decent TIM is 4-5 c. I find as you have also mentioned, getting a good spread and mount is much more important, then the TIM itself. Unless one is doing some hard core benching pushing the temps to the extremes, you really do not need the very best out there. Right now I'm using Arctic Cooling Mx-2 because I bought a 65 g tube on sale. I'm constantly mounting and un-mounting my heatsinks, water blocks and Cpu pot when extreme benching. I also swap processors in and out of my setups often so I go through more TIM then the average user. The Mx-2 works just fine for me, I have also used Gelid GC extreme, AS-5, what ever Cooler Master Tim that came with my CM 212. They all work just fine for me but I've never done any temperature comparisons between them.

 

[tachi1247]

I guess I just want to know what temperature difference people are really seeing. If it's really 4-5 degrees then I'll probably do it, but if people are seeing a degree or two then I won't.

 

[bassnut]

depends on what you are comparing too ...... from most decent tim to the very best again unless you are pushing max ore really trying to stretch your OC you won't find much difference. My big tube of AS 5 thickened and I was looking through review not long ago for Prolimatech PK3 reviews ...... you have to be careful and take some of these reviews with a grain of salt as there were some very notable differences between review of the same product from review to review. I have use AS 5 silver seems forever but recently switched up to Gelid GC Extreme I didn't notice and real temp difference but again didn't look to hard either both were good.

 

[Mandrake4565]

I doubt you will find many people here that have done testing between pastes, in a true scientific way. If you're not planning on pushing to extremes, just get one of the top non conductive pastes that you can find for a good price.

 

[knoober]

Ive done head to head with the "good" pastes against the "average" pastes and seen 1C or so of difference, sometimes more. Sorry I didnt save the shots from HWMonitor for long enough to post them here. I will be the first to admit though that perhaps application might have played its part. As Ive gotten more comfortable with application process Ive noticed that I use less to greater effect. that said - the difference between the diamond suspension paste and liquid metal was 1-2C but when we are talking the Arctic Ceramique the difference was up to 5C.

I guess i just chimed in to agree with everyone that said to use any non conductive top tier paste and to make sure you have your application method dialed in. Someday when I can be sure Ive removed application as a variable I may tear down the rig and do head to heads with what I have left of different pastes ( Ive actually got quite a collection now - Ive even got a tube of no-name paste that came with a cooler that looks like toothpaste!). Ill try to do a decent write up when that happens, but Zen will have been released long before I get around to that kind of shenanigans