• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Good air cpu cooler suggestions?

K

kubo.almighty

n00b
Joined
Mar 6, 2011
Messages
52
I currently have a zalman cnps9700 on my 2600K. It gives pretty good temps from what I've seen. My i7 idles around 29 C (which those of you with 2nd gen i7s know the cpu only idles at around 1.6 GHz, so temps are super low) and using prime95 the temps max out around 67 C. Gaming temps are around 43-49 C depending on the game. Overall, I'm not complaining, but I would like to get the load temps a bit lower. Any suggestions for good air coolers, possibly a closed water loop (h70 and the like)? However, the one thing I'm trying to keep down is the weight. I know the Noctua D-14 is awesome, but the thing weighs a ton. Anyway, suggestions greatly appreciated!
 
H50 with dual fan setup.

TRUE or Venomous X with push-pull setup (add a Delta 220CFM if you don't mind the noise)
 
My cooler works very well, but you would be hard pressed to find one. Its a raidmax x3
 
Excellent suggestions thus far. I'm putting this question out there too. On coolers that don't have the flat machine finish on the cpu contact plate, aren't you suppose to put a super thin layer of TIM to kind of fill in the cracks between the heat pipes and the base? That's been kind of a deterrent for me, not too confident in my doing that. BTW,Mannyman, what kind of temps are you seeing with that raidmax x3?
 
You're supposed to add TIM anyways, because no surface is perfectly smooth, especially the CPU surface.

I would also throw the A50 and A70 in there, both good cheap coolers.
 
Tsumi said:
You're supposed to add TIM anyways, because no surface is perfectly smooth, especially the CPU surface.
Click to expand...

I know you're suppose to add TIM. I've got arctic silver 5 going right now. What I'm talking about concerns the cpu heatsink base plate itself. Is the standard TIM application sufficient, or do you have to pre-treat the base plate that have crevices between the heatpipes and the base?
 
When you already have an aftermarket heatsink, and you want an improvement, it makes no sense to suggest budget heatsinks like the CM 212+ (unless THIS is your size limit) ... I would have expected the following answers:

If you can fit it, then the Thermalright Archon (it's VERY tall, so your case width may not be sufficient to close the case door) is the best (esp. with dual TY-140 fans) single-tower heatsink.

Next in line would be the Thermalright Venomous-X, which I favor over the comparable Prolimatech Megahalems rev.B (Both are $55 - $60).

The Thermolab Baram 2010 is also another option, as it makes efficient use of its measly 5-heatpipes (and is generally less expensive). It does not come with a fan, but works well with two.

http://www.xbitlabs.com/articles/coolers/display/6-coolers_8.html#sect0 - here is a review showing the Baram-2010 (not the original Baram) competing against the NH-D14. The same site shows the Archon beating the NH-D14 and Silver Arrow. Note that most reviews show the Ven-X and Megahalems beating the Baram-2010.

http://www.sidewindercomputers.com/thba20ulhecp.html - they have the Baram-2010 for $45

I eliminated the NH-D14, Silver Arrow, and Super-Mega due to size/weight. The Archon may be just as good anyway.

I use a relatively thin layer of thermal interface material, however I notice that if I were to use more, then most of the excess is expelled by the mounting pressure regardless.

Edit: If you're using an HDT-heatsink with separated heatpipes, then I believe the standard method is to apply paste in lines to fill in the center area of those grooves. Here is visual explanation: [Link was blocked, but it's in the following discussion] , and here is a discussion: http://www.overclock.net/air-cooling/442971-applying-thermal-paste-xigmatek-hdt-1283-a.html - which suggests that it'd be fine to apply the lines to the heatpipes. I used the first (lines in grooves) method with my HDT-sink, but I also spread a super-thin layer of residue over the CPU's IHS, just to be safe. It'd be nice if companies followed Swiftech and stuck to gapless HDT designs as employed in the Polaris heatsink.
 
Last edited:
Thank you so much for that link Archmage! That really answers my question about the TIM. Good heatsink suggestions too. I've got a pretty big case, so height shouldn't be an issue. Like you know, i'm just trying to keep a bit of stress of my mobo from the weight of the heatsink.
 
Never spread TIM. Pea size drop in the center and let the heat sink pressure spread it. If you have any cracks or crevices fill those with TIM first but still use the pea size drop. Spreading TIM s the worse application method there is. Check this video out it tells all. http://www.youtube.com/watch?v=EyXLu1Ms-q4&feature=fvw.

I also can get your free sample of IC Diamond TIM PM me your name, email, and address. We just ask you post your results out here.
 
Personally I'd go with a Cooler master Hyper N520 for $34.99 its a beast of a cooler
 
Zalman cnps9900 A or Max. Phenomenal cooler and much much better then the 9700 or cm212+
 
Falcon_CMH said:
Never spread TIM. Pea size drop in the center and let the heat sink pressure spread it. If you have any cracks or crevices fill those with TIM first but still use the pea size drop. Spreading TIM s the worse application method there is. Check this video out it tells all. http://www.youtube.com/watch?v=EyXLu1Ms-q4&feature=fvw.

I also can get your free sample of IC Diamond TIM PM me your name, email, and address. We just ask you post your results out here.
Click to expand...

@ That last part, No Friggen way, I wanted to try ICDiamond Since i learned about it, After reading and post somewhere on the interwebz on how to make your own diamond paste and how it has 3-5 times better heat transfer than silver. PMed!
 
mannyman said:
@ That last part, No Friggen way, I wanted to try ICDiamond Since i learned about it, After reading and post somewhere on the interwebz on how to make your own diamond paste and how it has 3-5 times better heat transfer than silver. PMed!
Click to expand...

It does? i always thought silver was the best lol
 
Diamond is better than gold,silver and bronze when we're talking about heat transfering.
 
Yeah, IC Diamond is considered much better than Arctic Silver. The difference in most cases under load is ~4 C.
 
I love my TR Venomous X Black with IC Diamond and push-pull. Big upgrade from my TRUE Black.
 
ill let you know how much Icdiamond helps i have a sample coming in about a week ill have a thread on it when it gets here.
 
My experience with TIMs:

It's possible that I just got bad batches of these two compounds, but I've had 4 tubes of ICD (2 different batches I think) and 10 tubes of X23-7783D (these were all the same batch I figure). If anyone has any application tips, then please delineate the process, and maybe I'll try again (because I do have a bunch of this stuff).

I've used IC Diamond: It's good, but it's difficult to apply. The performance was almost within the margin of error...

X23-7783D: Also good, and tests generally better than IC-Diamond, but also difficult to apply. It's difficult to apply a thin layer of these thicker compounds (mounting pressure did not consistently produce a thin, uniform layer for me (checking by the contact patch)). If you are able to apply it, this seems to be the best compound available...

...until one steps up to liquid-metal (coollaboratory liquid pro (and ultra, which may not be as good, but is less corrosive to aluminum)) and Indigo Xtreme (~$7-10 per application). These compounds DO produce a repeatable 2-4C improvement over, say AS5 and MX-2... although I notice that in some tests, CLL.Pro loses to X23/ICD...I'm not sure how to explain this other than improper application. It takes a long time to apply, but achieving a thin, uniform layer is repeatable. It is also difficult to remove (sometimes requires SANDING it off the IHS = light lapping every time)).

My experience aligns with Martin's: http://www.xtremesystems.org/forums/showthread.php?t=159752 - and he really knows what he is doing. He's the man behind the best fan/noise tests on the web. One could also say that he has popularized informal testing standards for PC-cooling equipment.

One of his quotes, which I think may be true: "Also testing on my 8800(Stock non-lapped IHS), but I've noticed it doesn't like to bond nearly as well with the nickel plating as it does pure copper, so I'm thinking you see the greatest benefit with a lapped IHS where you end up with a Copper/CLP/Copper sandwich." - I was successful, and all of my surfaces were lapped.

http://www.xtremesystems.org/forums/showthread.php?t=136356 - the review linked in this thread produces similar results, and one guy in that thread also wonders why the compound seems to murder all else for some people, and work no better than regular compounds for other people.

Edit: http://www.xtremesystems.org/forums/showthread.php?t=162440 - according to this thread, I didn't need to sand down my IHS each time... oh well...that still may have been one of the reasons it worked for me.

That said, for ease of use, I now just use thinner compounds (MX4, MX2, Matrix, AS5)...they're easy and they work well enough, but more importantly, they do so EVERY time. I cannot get a bad application out of these compounds. No warming of the tube, no heating of the heatsink to assist in spreading. It just works, and the performance is usually within 1-2C for me, so I'd need to do multiple mounts in a controlled environment to prove that the X23/ICD compounds were actually better.
 
Last edited:
IC Diamond is very very easy to apply I have 11-14 year olds doing it perfect every time. You cannot really use to much the excess squishes out and its non-conductive so no worrys. A pea size drop in the center, tighten the heatsink down your done. Works perfect everytime unless you don't use enough. And for surfaces that have crevices/cracks juwst fill them in first with ICD then still dop the pea size drop afterwards. Here is my youtube channel with me doing it on many CPUs/GPUs. http://www.youtube.com/user/FalconCMH?feature=mhsn

Oh and ICD is actually the best TIM for GPUs bar none IMO. I do not work for ICD I support them because of the high quality product they produce and I have used for many many years. I met the owner a year or so ago online and he lets me submit people for free samples as a courtesy to our community and to get the word out on his product. I do it puely for the community to educate and help folks out on one of the most important aspects of the build the TIM application.
 
After watching your TIM application videos, I have to say that you DO have an incredibly shaky hand :eek: - and you use SO MUCH PASTE... like a whole damned tube...ahah

I must also conclude that ALL 4 of my ICD tubes (I had the early batches from years ago... when ICD was a new product) were bad. They were not nearly as easy to apply as that. This is ridiculous... I'm quite pissed now. Perhaps ICD isn't as bad as I had thought, but that was quite a bit of money (I'm more upset about the TIME) wasted on bad batches that were so damned HARDENED that I needed to nearly BOIL them to get out of the tube, and use a hair-drier on the heatsink to get any sort of malleable consistency.

There is absolutely NO WAY that my paste and yours are similar. I suspected something was wrong as I am highly experienced and I had steady hands at the time (now not so much due to neurological issues)... I shouldn't have experienced that level of difficulty.

I retract my statements about ICD... but good lord... I'm a bit annoyed that there is such a big difference just visually. My paste could NOT be adequately squeezed out by heatsink mounting pressure... it would just create a thick, immovable putty.

My experiences with X23 were not quite so bad... I'm still willing to work with it, though it is still a bit putty-like...but this is what I've seen in pictures of applications anyway. No surprises there...

Edit: nice videos, btw... very thorough
 
Last edited:
Well I can get you a free sample of ICD if you PM me your name, email, and address to help make up for it. ICD used a preserative when it first came out that caused all kind of issues they no longer use it. And yes I have a neurological disorder that cause my hands to shake unfortunately.
 
Last edited:
You must log in or register to reply here.
Back
Top