ShinEtsu thermal paste

In my quest to find the best non-metal, non-IC Diamond thermal paste, a certain ShinEtsu paste: X23-7921-5 ( PDF data sheet) deserves some attention.

Let me first show you some results before going into any more detail:



The Valley benchmark was run on Extreme HD preset for 10 minutes without using my modded U3's 120mm fans, while Watch Dogs was played for 30 minutes on Ultra + SMAA with my modded U3's 120mm fans (check my sig link to see what I'm talking about). Ambient temp was 25C in all tests. The reason why I didn't use any external cooling with Valley was because I wanted to contrast each TIM's ability to cool in the presence and absence of external cooling. Laptop specs in my sig.

First, take the MX-4 results with a grain of salt, because the temps were obtained before I had cut out some metal in my U3, so the temps are artificially skewed higher, especially for the Watch Dogs test. In any case, MX-4 is complete garbage as it lasted a total of 30 hours before temps started creeping up. I saw a 3C increase on both GPUs after 30 hours of Far Cry 3, so unless you like repasting after every 30 hours of gaming, don't waste your time with MX-4.

The first ShinEtsu paste (X23-7783D) was a HUGE disappointment, as I recorded the highest temps both with or without external cooling. This almost made me want to give up and toss the other ShinEtsu paste I had as well, but I'm glad I didn't, because it turns out X23-7921-5 fared the best out of the 4 pastes when external cooling was applied (and the only one that matters really as I'm always using my 120mm fans when gaming).

Subjectively, MX-4 was the least viscous, and could be squeezed out easily with minimal force. X23-7783D was slightly more viscous, and required just a bit more force. Interestingly according to ShinEtsu's data sheet indicates it should have 4x the viscosity of MX-4, but I can say for certain that was NOT the case. X23-7921-5 was a notch above both, and finally Gelid GC Extreme took the cake as it required quite a significant amount of force before it would come out of the syringe. However this measure of "viscosity" is purely based on ease of squeezing out, and has no scientific basis whatsoever.

There does not appear to be a correlation between viscosity and performance as I had initially expected to see. Laptop heatsinks squarely fall into the "low mounting pressure" category, so my initial hypothesis was that more viscous pastes would fare worse, as they would not spread as easily, but if anything the opposite seemed true. What I can say though is that viscous pastes such as Gelid showed very little pumping out even after 50 hours of heavy gaming, whereas something like MX-4 had already formed pooled itself on the sides of the die.

The only question left now is how will the X23-7921-5 fare in the long term, and that's something I'll just have to wait and see.

Finally to address the white elephant in the room, yes I only tested 4 pastes, but I was only interested in testing pastes that were known to perform well in low mount pressure/poor contact situations, as shown by Tom's and SkinneeLabs. In any case, Tom's tests showed Gelid GC Extreme was second only to liquid metal pastes regardless of mounting pressure, so that was my benchmark, and ShinEtsu X23-7921-5 managed to edge past it.

It would certainly be interesting to see how ICD7 compares against X23-7921-5, but as I am boycotting IC on moral grounds and don't want to scratch my die, any volunteers for testing X23-7921-5 instead? (looking at you Mr. Fox )

 

Interesting results.

When I get my machine back I'm going to hope that they do a great job with the paste but if they don't I'll consider the Shin Etsu. How much was it and where did you get it? I try to buy things on Prime because I hate shipping haha

Sent from my HTC One_M8 using Tapatalk

 

I think ModDIY is the only place that sells it. $15 for 3g and free shipping from HK = pretty good deal.

Just have some patience though because registered mail took 2 full weeks to arrive (you have to sign for it as well). There's an express option but it costs more than the paste itself ($30)!

 

Its on sale lol. Sold!

http://www.moddiy.com/products/(Dai...al-X%2d23%2d7921%2d5-W{47}m.K->-6.0-(3g).html

Sent from my HTC One_M8 using Tapatalk

 

Okay sale doesn't have free shipping and shipping is 10 bucks which makes the sale more expensive, that makes sense haha.

Ordered though, thanks for sharing.

Sent from my HTC One_M8 using Tapatalk

 

EDIT: ninja'd

Also, make sure you dig around for a 10% off coupon.

Btw in case you don't feel like sleeping, here's a link to a catalog by ShinEtsu that pretty much lists all the products they make, and goes into a lot more technical detail.

Apparently the X23-7783D has a solvent carrier, which needs to be evaporated by burning in for 30 minutes at 60°C, otherwise thermal conductivity is cut in half. Perhaps that's why it performed so terribly in the Valley benchmark as that was done straight after a fresh application. X23-7921-5 doesn't need to be burned in, and in addition to having a higher thermal conductivity, also has a lower thermal resistance.

For the first time I actually wish I had a tube of IC Diamond so I can test to see whether this ShinEtsu paste is indeed superior.

 

I had already ordered it but 15 bucks is cheap, no big deal.

If its superior to GC Extreme, its superior to IC Diamond.

When I was prepping my machine to send it back to Sager, I repasted the slave with GC Extreme and its about 2-3C lower under load than IC Diamond was.

More and more I'm starting to think IC Diamond isn't the best paste to use in these machines. If it wasn't for what it did to my CPU die, I'd try it in my desktop machine but nahhhh.

 

I doubt it's superior to GELID Extreme, the SkinneeLabs guys were extremely thorough, the only thing they didn't test was longevity but the performance is pretty much unquestioned. I would be interested if it achieves a similar degree of longevity to ICD if the bulk thermal conductivity is the same.

And good lord, why so much hate for IC Diamond? we should be thankful it has so much diamond/particles in it for it to have abrasive properties. It's to do with the filler loading, ICD just simply has much more filler and less carrier hence its where the stability and bulk thermal conductivity comes from (albeit at the expense of spread ability and wetting). It's the same as toothpaste (which btw is an excellent thermal paste) in terms of abrasion power.

From what I can gather, Shin Etsu is targeted at the same application as ICD, notably bulk thermal conductivity (i.e. poor interface contact) and high stability. This means, Shin Etsu and ICD are poor candidates for desktop parts which usually have better mounting and easier access (thus rendering wettability and spreadability as more important features than stability or bulk conductivity).

 

Why the hate? Well mostly because of this. The die/heatsink etching certainly doesn't help it in my mind either. As long as a TIM has enough longevity that I only have to repaste every 6 months at most, then I'm happy enough. And if performance is superior then of course that's a bonus.

ICD7 has 4.5W/m-K of thermal conductivity, while the thermal resistance is 0.25°C cm²/W at 100 µ bond line thickness (BLT). X23-7921-5 has 6W/m-K thermal conducitivty, and thermal resistance appears to be around 15mm²/KW at 100 µ BLT by extrapolating from one of the graphs in the ShinEtsu catalog.

Not sure how to equate the units for a direct comparison, but assuming 1 cm² = 100 mm², if this calculation holds up then ICD7 would have a thermal resistance of 25 mm²/KW at 100 µ BLT. But once again I don't know if this is a legit conversion.

I think another interesting question would be, if ShinEtsu is indeed superior to ICD7, how long it would take before it would degrade to the point that it matches ICD7's performance. And again if that answer turns out to be >6 months, then at least for me, there's absolutely 0 reason to use ICD7.

 

You guys ever see this impressive review? Not only did he test 80 TIMs, but he also seems thoughtful on application methods and testing methods.
80-way Thermal Interface Material Performance Test | Thermal Interface Material,Thermal Paste,Heatsink Compound,80-Way Thermal Interface Material Best Thermal Paste Heatsink Compound Cooling Performance Comparison Benchmark Tests

Look at how little the various pastes vary. I think when people see big differences in their paste results it's caused by application variation.

 

Yes I've seen that review, but there's one problem: they only test desktop CPUs, which have MUCH HIGHER mounting pressures that laptop heatsinks can only dream to achieve.

Additionally, the desktop heatsinks mount to an IHS instead of the bare die. For a much more relevant test, see user Idontcare's testing of a delidded desktop 3770K.

What's even more impressive however is just how little (relatively speaking) applying excessive paste does to the temps, as seen in this ridiculous overapplication of Noctua NT-H1.

To quote Idoncare:

P.S. If you feel like jumping ship to NT-H1 after reading that, DON'T. It has some pretty serious pumping out issues if you read through that thread.

 

I'm using gelid extreme now. I must say it seems to be superior to ic diamond.

 

Agreed. I bought the Shin-Etsu to try it but I'm really impressed with GC Extreme, I just hope it holds up.

 

I was surprised at its performance as its kinda runny when applied

 

I think there's something to be said about viscosity of the paste in low pressure/poor contact situations, which is basically where the laptop heatsinks fall into (particularly for GPUs).

 

I'm a Gelid Extreme guy...

 

For those Gelid Extreme users, do you recall how long your application lasted before the temps started deteriorating? I may necro this thread every now and then to report on how my ShinEtsu application is holding up, but it'd be good to have a baseline for reference, thanks.

 

What I have seen is that there is a small difference between the best cooling paste. It is under long-term use that you see the real difference. At higher temperature over time and an overclocked processors you will quickly know which one is the best quality (1-3 months). The high temperature of the CPU break Down the thermal grease. A laptop will break down the thermal grease much faster than one that is used on a desktop PC. It is much more important with a good thermal grease on a laptop than on a desktop PC. A desktop computer, often have a much better cooling solution than a laptop. :cry:

 

CPU I would not recommend anything but Liquid Ultra now, especially for Haswell chips that run like nuclear reactors during meltdown LOL.

GPUs I think Clevos may not benefit as much from Liquid Ultra due to the poor die contact, which is why I'm trying to find the best non-metal non-IC Diamond paste.

Btw ShinEtsu in combination in combination with my U3 cut out mod is so effective that temps dropped by 7C when playing Far Cry 3. Before my master GPU would max out at 77C and slave GPU shot up to 82C, now my master maxes at 70C and slave at 74C. What a day and night difference!

 

I have the same mod u3 that you have. You can see the PCH temperature, it goes down by almost 8 to 10 degrees with the modification. During normal use, the temperature of the pch not higher than 47-48 degrees Celsius.

 

The PCH is hidden under the keyboard in the P370SM unfortunately, so even at idle it runs around 55C.

 

The PCH isn't really a big deal as far as cooling goes though, they're built to handle heat.

Sent from my HTC One_M8 using Tapatalk

 

There is no technical explanation that I can find that supports the idea that ICD is any more abrasive than any other compound when used under normal application and removal procedures.

If you have one I would like to hear it.

Of course one could always use the product as directed by the manufacturer.

Removal

It is not necessarily a done deal that IC Diamond caused your scratches. I would note that in the following 1-1 comparison with IC Diamond and Shin Etsu and in point of fact our experience is that there is no intrinsic reason for it to have done so. Scratches can occur due to a number of factors outside our control.


In the polishing test set up below IC Diamond is compared to ShinEtsu. Copper plate was prepped with .001 steel wool to establish a uniform texture grain from left to right, paste in then rubbed top to bottom against the texture grain.




Application pressure was approx 8-10 PSI measured on a shipping scale with about about a ¼ of an inch contact area this works out to be 30-40 PSI. This is 10 times the pressure you would use if you were lapping a CPU and 200 times our recommended removal pressures and so is a fairly extreme test.



Final result after 10 to 15 seconds application rubbing. The result shows no appreciable difference between the two, obviously all thermal compounds are polishing compounds. Shil Etsu contains aluminum oxide (same as most compounds MX4, Arctic Silver etc.) which is the most commonly used abrasive and is the main component in most sand papers



Below note the highlighted area which was enlarged and pasted in the upper left. These scratches were incurred with a plain soft paper shop towel purchased from Lowes hardware. Paper towels contain wood fibers and Numasists recommend not using paper envelopes as just removing and replacing coins the paper will scratch them. The soft paper towel scratches are deeper and more prominent than either of the compounds.




You do not have to be aluminum oxide or diamond hard to make scratches on many surfaces as textured surfaces are often weaker than the underlying material. The ugly scratches below were made with just a fingernail (Fingernail MOHS 2.75, Copper MOHS 3).




Misconceptions Regarding Diamond Hardness

In the nomenclature of the abrasion trade when they say “diamond cuts better than anything else” what is really meant is that it lasts longer. In the abrasion literature nearly all specifications relate to movement, RPM, tool feed, heat the particle material can endure under friction etc. So the assertion that diamond cuts better than anything else in what is essentially a static non cutting application one would be attributing characteristics that do not apply to the application and removal of thermal compound.

Example

Take two sheets of sand paper, one aluminum oxide the other diamond each with a 800 grit, approx the particle size of most compounds. Now scuff a couple CPU's with the sandpaper and you note no difference between the two and the result would be the same as our example above, Why? Because they are the same particle size and density as well as both being appreciably harder than the material the material being abraded.
The only time that hardness comes into play is after some number of cycles the aluminum oxide(MOHS 9) will begin to wear sooner that the diamond (MOHS 10) and that's it, the only advantage is that diamond has a longer life due to hardness, there is no “magic” to hardness as implied, no increased probability of it scratching a component over another compound, simple.

 

I don't think it's that surprising. The paste will be compressed and thin out regardless.
.

 

Tell that to everyone who has scratched CPU and GPU dies care of ICD. It's not a problem reported with any other paste therefore it would appear to be an IC Diamond issue.

That test is the heatsink... Everyone knows heatsinks stain, scratch, etc. Nobody cares about that but the scratches left on dies by ICD are impossible to ignore.

Sent from my HTC One_M8 using Tapatalk

 

There is no standard for reporting W/m-K as everybody calculates it differently as well as measurement is done with a variety of different equipment. One man' 15 W/m-K on another test system might yield 4 W/m-K, so unless you know the calculation as well as the test conditions of both the number is pretty well meaningless other than for marketing bragging rights.

 

With all due respect, your removal instructions don't seem special to me. Liquefying/softening the TIM with some organic solvent (be it IPA or acetone) is pretty much standard practice, particularly if you're cleaning off an old application where the TIM has hardened and caked.

And I have to ask, why not just apply ICD to a CPU, then mount the heatsink, and remove it as you would per your instructions? Desktop heatsinks have much higher mounting pressures easily exceeding 30-40 PSI. Wouldn't that be a more realistic and fair test?

In any case die/heatsink etching remains a real and documented phenomenon, and seems to be singly associated with ICD products. We can argue about not following instructions all day long, but then I have to ask, isn't the very fact that you have to specify removal instructions a tacit admission that the paste may have something to do with scratching dies and heatsinks? Nevermind the fact that removing other pastes "the wrong way" don't seem to cause these scratching issues.

 

1. Actually our compound removal page is the 4th most visited page on our site. It must be special to those looking for instruction perhaps?

2. If you had read the above post the applied pressure was in the 30 to 40 PSI range, but we will go one better here.

In any event pressure has minimal effect as you are working with a "Bed of Nails" where the load is evenly distributed

Assume the die is about 1 sq in. you are looking at about 2500 10 micron particles per inch so at full Intel max loading you are looking at a per particle of 0.028 lbs., as nobody is at the Intel max you are probably looking more to the nominal Intel spec of 35-45 psi or .014-.019lbs.

To mark cold rolled copper you would need a single half mm point with 70lbs applied which would equal approx 105,192 PSI vs worst case scenario of .028lbs

And this is an off the cuff over estimate not taking into account lateral or side to side loads between particles which would cut the individual particle point loading in half again.

As IC Diamond has more or less the same particle size as other compounds they would be subject to the exactly the same pressure loadings regardless of the type of particle.

Proof of concept here is on our removal page Removal

For a more "realistic" extreme comparison we tested it above the Intel Max of 75lbs about 30lbs higher than you can get with your avg best aftermarket sink

Simulated Mount - Pony Clamps Applied Pressure of 79lbs @ Intel Max

The copper oxide on our Compound Removal Page is much softer than the underlying copper which has an MOHS of 3 with absolutely no damage evident.

As silicon has an MOHS hardness of 8.5 puts it in the ballpark of hardness of both ShinEtsu's aluminium oxide filler with an MOHS of 9 and Diamond's MOHS of 10.

If it will not mark copper or the softer oxide it's exponentially less likely to mark a silicon die.


3. As for being a "real and documented phenomenon" other than those attributable to carelessness have been investigated by IC were only 5 or 6 complaints in total that got posted hundreds of times over the internet of which IC then investigated and found the claims to be without merit.

We offered to replace any damaged parts even though we are deeply convinced ICD played no part and was declined on all accounts.

We also offered to have them send their compound directly to an independent lab for analysis which was also declined.

We were unable to mechanically duplicate any of the pictured results so had them analyzed.

Here is one example below

AnandTech Forums - View Single Post - Thermal Compound On Core i7 2600k Heatsink

Well thought of forum member with 19,000+ posts

Here is the photo analysis

http://innovationcooling.com/image/icreport3.pdf

The investigator also noted the serial numbers did not match in the before and after pics



Photo Forensics is a well understood scientific discipline, well documented with many technical papers and reports and functions to a high standard, much care is taken in the handling evidence. The analysis software on analysis of the photo introduces no "artifacts" of it's own and makes no changes, only states what's there. Re-compression by the originator does not create any manipulated artifacts and only obscures the manipulation, the standard is such that it is admissible as evidence in a court of law.

You can do your own analysis here

FotoForensics

Try it out on these pictures or your own, upload them, make changes to your own pictures then check the result and you will see the exact same effects on your own manipulated area.

 

And that's a flat out lie. Sneekypete sent you his Noctua cooler only for you to refuse to send his old one back because it was "horribly damaged". So what if it was "horribly damaged", that cooler was legally his property, if he wanted it back I see no legit reason not to send it back. Given that precedence, it's no surprise that people were hesitant to send you their hardware.

As for the photo forensics, not even going to get started on this one except to urge people to start reading from page 23 of that thread.

 

I, in fact DID have a Brand New Cooler shipped directly Him. He told me he got the cooler free for review and did not care if he got the old one back. Some poetic license on my character there?"Horribly damaged" as I never said that in any form or manner and you did not hear me say it, you do not really have cause to post events you where you were not personally involved and have no first hand knowledge of and I take exception to that.

Note the "Horribly damaged" was edited out at 8.05- Changing Horribly to too damaged does not change your comment significantly so it stands

in any event

Yes we all know about the brouhaha on TP Kind of yesterday's news, a one off where I bit hard on the baiting which was my mistake and I regret that but in 5 years and 24 giveaways I had Zero problems. Googling is something we all can do and you can believe what you want but it does not make an argumentative point .

Of course we gave the opportunity in case of a mistake for a reanalysis as the armchair experts on TP were adamant that it must have been Microsoft Office Picture Manager's Autocorrect feature introduced artifacts such as traces of Adobe: Creative, marketing, and document management solutions adobe, and World Wide Web Consortium (W3C). and so the original(supposedly) pictures were sent directly along with a Control group one new picture Auto corrected and one to the investigator with the following conclusion.

Report here
http://innovationcooling.com/image/lacreanalysis.pdf


The control group shows no indications of http://ns.adobe.com,adobe,orhttp://www.w3.org.
However, the auto corrected digital images of example 1 and example 2 contain traces of Microsoft Photo and http://ns.microsoft.com/photo.

Whereas the supposedly camera raw photo of damage supplied by the person who took the photo contained in it's EXIF file the adobe Photoshop marker when it should have only contained the camera data, proving our case and undermining his.

Of course keeping an open mind if you have some technical publications as to the invalid nature of photo forensics please enlighten me, but please no google links to an amateur analysis.

It was mechanically impossible to reproduce any of the damage claimed and you can read the reports and try to duplicate on your own and forensic analysis was fairly definitive along with other inconsistencies like multiple types of damage not seen anywhere else limited to 5 tubes which would each have to contain 5 different type particles? and only on that forum? The laws of mixing do apply and the odds would be incalculable. That being said regular paste processing is done elsewhere but on samples for customers and giveaway's I do personally as a convenience and so I personally would have noted any inconsistencies like particles that were too large to to fit through the tip of the syringe.

Shin Etsu is a good paste enjoy your test

 

Just to set the record straight, this is what you posted in your own words in post #454:

I changed "too damaged" to "horribly damaged" precisely because I didn't want to misquote. No "poetic license" on your character whatsoever, so please don't do the same to me. Never once in that thread did the person in question mention he did not want it back, and in fact the tone of several of his posts would indicate quite the opposite, but whatever.

As for me not having any cause to post that, you would be absolutely correct if I posted that out of the blue and without provocation. But no, this thread's main focus was on various pastes, their longevity, and whether their viscosity could be related to their performance and durability in low mount pressure/poor die contact situations. This thread was never about your product to begin with, you don't see me going around in your threads and actively preaching about the fiasco that went down at TPU.

The irony here is, if you actually bothered going through some of my older posts I actually recommended your product to various users despite me boycotting your product. I have never actively actively campaigned against your product, and in every instance I made it perfectly clear I was boycotting your product for purely personal reasons when mentioning why I did not use your product despite recommending them.

Yet simply because I mentioned in passing that your paste could potentially scratch dies and heatsinks in the first post, you felt entitled to completely derail my thread and take it in a completely different direction. I should've also known better than to take your bait, but I did, and this thread is now effectively ruined. Congratulations you win OK?

 

I wouldn't really say he won anything... Those of us reading this in awe won't be buying any of his products ever again after seeing his attitude...

Sent from my HTC One_M8 using Tapatalk

 

OK gentlemen, let's stop. This thread was for exploration of the longevity of ShinEtsu paste, not IC Diamond. Let's please stay on topic and drop the discussion on IC Diamond in this thread. The debate is starting to get a little heated and is derailing the thread.

Thanks, and enjoy the rest of your weekend.

 

If you guys are in the mood for a bit of reading, check out this PDF from Dow that explains how different factors such as thermal conductivity, thermal resistance, and bond line thickness are all related and affect each other.

If you want a more in-depth reading, then have a go at this paper that really digs deep into the interplay of temperature vs thickness, thermal conductivity, and thermal resistance for solders TIMs The most surprising part to me was they showed that below a solder TIM thickness of 0.1mm, increasing thermal conductivity or decreasing thermal resistance yielded no net benefit on temperature. Obviously this study was done on solder based TIMs so I'm not sure if the results could be directly extrapolated to TIMs made with other base materials. But, perhaps this may explain why some people see less of a benefit with Liquid Ultra than others.

 

My ShinEtsu came in the other day.

I'm not going to crack the stock compound unless its running hot when I get the machine today - if its running hot, I'll put the ShinEtsu on it and give it a try.

 

I have some of this and although it works great for the first couple of weeks /month after that its no good goes out fast.

 

How does it compare to GC Extreme though?

 

Xigmatek Xi-3 HDT and some Prolimatech stuff also should be considered

 

I loved Gelid extreme before. Now I'm unfaithful and using Liquid Ultra

 

Gelid is The Best If You don't want stains or scratching and extreme ease of use

 

But does it last? There are plenty of TIMs that start off great and then have to be redone in a week or two because they get burnt up. Hoping that isn't the case with Gelid because so far I really like it.

Sent from my HTC One_M8 on Tapatalk

 

Exactly. MX-4 is GREAT when freshly applied, but after 30 hours of gaming it becomes an epic fail.

I'd rather trade 1-2C in initial temps for a paste that can hold its ground for AT LEAST 3 months. I'm talking about 3 months of "light" usage with not more than 10 hours of gaming per week. Surely that's not too much to ask for.

 

I've been using OCZ Freeze for a number of years now. Easy application and removal, no burn-in time, good performance, and no problems with longevity that I've observed.

 

A number of pastes are durable enough running at ordinary clock speeds. It takes an extraordinary paste to last more than 2 to 4 weeks with substantial overclocking. On my systems that don't have unlocked and overclocked CPUs it doesn't take a fancy thermal paste to last more than a year. My son's desktop is coming up on year 3 with IC Diamond and our M17xR2 typically goes between 1 and 2 years on IC Diamond. Both systems run close to the same temps as when it was first applied. But, neither of those systems have overclocked CPU and GPUs. An old Sager that I used to have with Core 2 Duo and Go 7600 GPU went about 2 years on AS5, then 2 more years on IC Diamond before it became too tired and worn out to be of any further usefulness. It was still running cool on IC Diamond when I stripped it for spare parts and tossed the leftovers.

I still have an old (but never opened) tube of ShinEtus laying around here somewhere. I don't remember which ShinEtsu paste it is. If I can find it and it's not all dried out I might test it on the Alienware 18 to see what happens. ShinEtsu has always had a decent reputation among enthusiasts. I don't recall where I got it. I think maybe it came with a fancy old desktop CPU heat sink that I bought some years ago, back in the day when liquid cooling was still considered an extreme novelty solution reserved only for overclockers.

 

Well IC Diamond is known for lasting a long time though.

I may end up repasting soon, I'm debating whether or not I want to crack the seal on the new paste and pads for what would likely only be a 1-2C reduction in temperature.

Sent from my HTC One_M8 on Tapatalk

 

I wonder why everyone is having issues with shin etsu. I use it exclusively on my laptops and also my son'a desktop and haven't experienced any sort of temperature degradation after extended (3-4 months) use. I overclock as well, though probably not as high as some of you.

 

Which one do you use? That makes me more interested in trying it since the results seem to be slightly better than GC Extreme which is my current favorite

Sent from my HTC One_M8 on Tapatalk

 

2 weeks and 40 hours of gaming later still holding strong, so it's at least better than MX-4 LOL

 

I will admit I'm curious. It's sitting there in the envelope with all the Chinese stickers on it waiting lol

Sent from my HTC One_M8 on Tapatalk

 

I use Shin Etsu X23-7783D.