Best thermal pastes 2011?

[hellokitty[hk]]

Well, it's time for a new tube, and newegg's got 10% off ATM.

I've looked and found some very conflicting reviews, so I'm asking techspot to help me sift out the truth.

I've read some reviews about the TX-4, all have been favorable, but there was one that listed the TX-4 as about 12°C lower than TX-2, and I'm highly inclined to call BS there. Also, it was only one round of tests.

This one looks pretty legit but it ranks AS5 as much worse, which I wouldn't doubt alone, but I've ready many reviews that put it much higher. I do think that AS5 is falling out though.

So anyway...I'm looking for some solid results and the best performance. Price will be a factor, but the absolute price differences are not that big in thermal compounds, so it'll be secondary.

 

[captaincranky]

So anyway...I'm looking for some solid results and the best performance. Price will be a factor, but the absolute price differences are not that big in thermal compounds, so it'll be secondary.

"Since price will be factor", I just gotta ask, how much is 10% off too much?

 

[EXCellR8]

ugh, there's TX-4 now?

i've been using the Tuniq pastes for awhile now and i still recommend them highly.

 

[hellokitty[hk]]

"Since price will be factor", I just gotta ask, how much is 10% off too much?

I mentioned that price would be secondary to performance, so it's not a primary factor per se. But of course, the 10% off doesn't at all detract from performance, so the only negative about that is I have to order within a week, so a bit of an opportunity cost, except that I'll be ordering quite soon anyway.
Sales exist to stimulate more sales, and I have to say it works sometimes.

Appreciate the fast responses .

 

[captaincranky]

Of course with low dollar purchases such as thermal compound. "10% off", is rapidly whittled down when shipping costs are factored in.

This might be a dumb question but, "isn't plain old AS-5 as good as any, (anymore)"? Time has regrettably passed me by.

 

[hellokitty[hk]]

Actually ATM many have free shipping, including the TX-4 and TX-2; free shipping/consolidating items into one package is one of the best incentives for me personally to buy online.

Yes that time has past...it's mostly that the price difference is pretty minimal, and it makes me feel better.
Couple degrees...meh, but I'm also planning to use this for some other applications that may go up to 150°C, so differences will be a little more pronounced.

 

[dividebyzero]

A lot of the TIM's these days are within a few degree's of each other. The bigger differentiators are ease of application , curing/burn-in time (if any. AS5 for instance requires ~200 hours of burn-in) and whether the grease is conductive or not.
TX-4 is a pretty good (and widely used ) TIM and applies easily.Coollaboratory Liquid Metal Pro and ShinEtsu X23 are generally considered the best TIM's on the market. Here is a roundup of 80 thermal greases- have fun. This review may be of more use to you since ShinEtsu and Coollabratory may be hard to source in some markets.

I would note that the BMR review is at some odds with some other findings- especially regarding the Liquid Metal Pro (which I tend to use on occasion). Here is possibly a more representative review.

 

[Mizzou]

I've had good luck with Tuniq and find it easy to work with. Didn't know about TX-4 either, may have to give it try one of these days.

 

[EXCellR8]

just ordered a tube of TX-4 because i can't find the TX-3... may have used it all i suppose.

i've used TX-2 and TX-3 on both CPU and GPU cores and it has always been at least 5-10c cooler than generic paste. i was never a fan of AS5 because of the curing and the texture, but the Tuniq pastes have been the most effective. another decent product is the OCZ freeze stuff but i don't know if they have made any revisions to their formula.

 

[Tmagic650]

AS5 still works for all my builds

 

[hellokitty[hk]]

dividebyzero:
It seems to show that AS5 is in fact quite good, just behind the ShinEtsu. That'd be nice since I can get it for $8 instead of newegg's $10, but I hear a lot of people argue that AS5 is very old and isn't as good as the newer ones. Also it's texture is less than perfect, and it may be slightly conductive. Thermal conductivity doesn't really matter too much to me but that's still a nice feature I guess. Anyway I see a lot of different results and opinions about AS5.

Probably no liquid metal pro since I need to use it on aluminum, and I think it's supposed to be detrimental to aluminum.

 

[dividebyzero]

If the TIM will be coming into direct contact with Aluminium then I'd steer well clear of the liquid metal pro -it will start chewing away the heatspreader/heatsink.
AS5 probably only gets a rip because of the fact that people like >>NEW<< stuff, and hype tells you that new equals improved. The fact that AS5 have been a standard for so long should give an inkling to how well it is regarded -it figures in many comparison tests as almost the default standard of measurement.
Other than the curing time (not a big issue unless someone is planning a big voltage increase from the outset) and the fact that it can be very viscous (also not a big issue if you use the rice grain sized blob/thin stripe and use the heatsink securing pressure to spread the TIM) there isn't a lot negative associated with AS5 - it's relatively cheap, readily available and has been proven over time.
By conductive I presume you mean electrically conductive (or capacitive). Unless you are applying the TIM with a trowel you should be ok - just make sure that any excess TIM that squeezes out/dropped doesn't stay on the motherboard, since it can and will spread over time, and can (and will) cause a short circuit.
Thermal conductivity is the measure of how well the TIM transfers heat from a high energy state (CPU) to a low energy state (cold heatsink). This, and how thinly it can be applied whilst still maintaining a continuous mating surface between the two metal layers determines the TIM's effectiveness.

 

[EXCellR8]

I don't prefer TX-2 or 3 because it's newer... I prefer it because it's better :suspiciou

 

[captaincranky]

A lot of the TIM's these days are within a few degree's of each other. The bigger differentiators are ease of application , curing/burn-in time (if any. AS5 for instance requires ~200 hours of burn-in) and whether the grease is conductive or not.

Um, how could anything made with silver possibly be electrically non-conductive? What actual drawback exists if the TIM is electrically conductive?

 

[hellokitty[hk]]

Silver oxides, IDK.
I've never thought of a situation where that comes into play; except if you mess up and squeeze a tube full.
It might be a problem with GPU heatsinks, or I could be making things up.

 

[dividebyzero]

Um, how could anything made with silver possibly be electrically non-conductive?

I think AS5 (and similar) are electrically conductive to a degree. The carrier (silicon, ceramic, mineral oil) I believe acts as an insulator -how effective that is depends on whether the carrier dries out or remains viable over time.
EDIT: You're pretty much on the money hk regarding GPU's. The voltage/amperage involved is fairly decent (at 1volt vCore a lot of mainstream/enthusiast boards will have in excess of 80 Amps flowing through them).

What actual drawback exists if the TIM is electrically conductive?

For it's intended purpose, none whatsoever. Where is becomes problematic is where the TIM is applied heavily (the old more= better scenario) and the excess is pushed out from the heatsink/CPU heatspreader. If the amount is sufficient (i.e. the tablespoon quantity used to secure stock GPU heatsinks at Foxconn or PC Partner) then it can ooze across the CPU socket to the surface mounted capacitors etc. and short out the board. More likely once the board is vertically mounted and gravity starts pulling the grease downward (usually toward the chipset silicon/traces.

 

[Tedster]

In reality, one is pretty much as good as another. You will never be able to tell the difference. I've always used arctic silver, but stock pastes work just fine.

 

[hellokitty[hk]]

Looks like a very through review, but it's a little outdated.
Based on this, I'm currently considering the TX-2, but I'm still pretty indecisive. I still can't find any solid reviews on the MX-3/4 and TX-3/4.

EDIT: I might be thinking about this too hard.

 

[captaincranky]

[QUOTE='hellokitty[hk]

EDIT: I might be thinking about this too hard.[/QUOTE]Ya think.....? More like "obsessing", or perhaps, "imploding".

To the upside, that poor ten dollar bill burning a hole in your pocket, no longer has to sit there spinning its wheels! It can fly joyously away in its newly found freedom, in search of the perfect TIM. I could go on about it, flexing its buying muscle, despite being blissfully aware of its badly curtailed purchasing power in the "Year of our Lord"...."2011 AD".

Sic; "to boldly go, where no sawbuck has gone before"!

(Wow, I bet Roddenberry is rolling over in his meteor shower after that stinker)....

 

[dividebyzero]

"to boldly go, where no sawbuck has gone before"!

Don't be so sure, those Hamilton's sure get around

Wow, I bet Roddenberry is rolling over in his meteor shower after that stinker

He came up with a few himself

 

[Leeky]

I've used Zalman ZM-STG2 the last few times, purely because it comes in a bottle with a brush and makes applying it thinly and evenly very easy.

I see in that 80 paste review it got an A, so good enough for me.

I can't say I actually noticed any differences in temps between that and AS5 to be honest - Originally my PC had been overclocked after AS5 was applied, and then removed at some point to clean, and re-applied with this Zalman stuff, and then refitted. Temps seemed fine.

I might try out this TX-4 as well, seems to be fairly well recommended verbally when I've spoken to people - I never did like applying the AS5, it just seemed overly messy somehow.

EDIT: Mine is ZM-STG1 in fact, not ZM-STG2 as I first thought.

 

[captaincranky]

I never did like applying the AS5, it just seemed overly messy somehow.

You just need to have considerable experience using single edge razor blades. After which, you'd much prefer it.

 

[Tmagic650]

I agree captaincranky,
I use AS5 on all my builds. No muss no fuss...

 

[Leeky]

You just need to have considerable experience using single edge razor blades. After which, you'd much prefer it.

Thinks of ways in which I can gain considerable experience with razor blades... :haha:

I've never really warmed to the idea of taking a metal cutting implement to a CPU face, and scraping it along it for some reason. lol.

 

[Tmagic650]

A plastic bank or credit card works well too