Best cooler other than stock?

By rogue12 ยท 17 replies
Jan 1, 2010
  1. im after something better then the stock cooler for my lga 775 Q9550, from the link below or or any sites please recommend one to me, price limit is $60 AUD

    im looking for a good balance between noise and cooling, good cooling and not noisy.$60.00&rbSrchType=ALL&recMax=500&Page=1

    thanks for the help.
  2. dividebyzero

    dividebyzero trainee n00b Posts: 4,891   +1,264

    Are you planning to overclock?
    Will your chassis accept a tower cooler based on a 120mm fan design ? (You will need approx 160mm of space between the motherboard and the side of the case)
    From the Techbuy site these are good coolers:

    The Cooler Master Hyper 212 would be the best performer.

    If you limited to approx <130mm of space then this:

    the 3R Systems Iceage 120 (especially the Boss II variant if within your price range), Thermalright Ultra-90, Sunbeamtech Core-Contact Freezer (120 and 92mm fan versions) would also be good choices.

    The Arctic Cooling Freezer 7 range are cheap but not a huge step up from the stock Intel cooler- ok if you plan on no (or mild) overclocking.

    My second system runs a tower cooler (Sunbeamtech Core-Contact Freezer- the 120mm version) on top of a Q9400 running at 3.6GHz ( 425 x 8, 1.25v in BIOS) 24/7 at 33-37 C idle, 45 C (approx) normal load, 47-52 C using OCCT torture test (Prime95 small/blend FFT). These are summer temps when the ambient (room) temp is 26-34 C.Most good coolers in a well ventilated chassis will give similar results.
  3. CMH

    CMH TechSpot Chancellor Posts: 2,039   +9

    I personally have the first edition of this heatsink, and many review websites use this same heatsink as a comparison to any other new heatsinks. Its better than any of those mentioned up there, and as far as I know, its only been beat by the noctua NH-U12P consistently (which costs quite abit more).

    It costs 69 bux, and doesn't come with a 120mm fan unfortunately, but the 3rd edition (which is the one linked) has been optimized for lower cfm fans (better cooling with less noisy fans!!)

    And with my first edition heatsink with a low cfm fan, I run my E6420 24/7 without any problems at 3.2ghz (stock 2.13ghz) (including a 48hour torture test)
  4. dividebyzero

    dividebyzero trainee n00b Posts: 4,891   +1,264

    rogue12 explained that his limit was $AUD60.
    Otherwise I would have included any number of cooling solutions including my own:
    Swiftech Apogee GTZ, Laing D5, Swiftech Micro res v2, Feser TFC XChanger 360.

    While the TRUE is a great cooler (I have used the original version) it also requires the use of 1 or 2 good fans- which can be rather expensive-especially if a low noise level is required.
    The Sunbeamtech CCF is in my opinion (and , as well as a damn fine option and comes paired with a 91 cfm fluid magnetic bearing fan at no extra cost, whereas to get similar performance from a TRUE you can add $25-30 for one Noctua/Scythe fan.
    You can also add- if $60 isn't the limit....
    Thermalright IFX-14
    Prolimatech Megahalem (+ Rev.B)
    Prolimatech MegaShadow
    Thermolab Baram
    Titan TTC-NK85TZ Fenrir
    Noctua NH-D14
    Corsair H50
    Zalman 9900LED
  5. CMH

    CMH TechSpot Chancellor Posts: 2,039   +9

    I'm pointing out options, and 69 isn't far from 60 really. Even though it doesn't come with a fan, some people (like me) have spare fans sitting around doing nothing. Everything from 40mm to 120mm fans.

    Most of the others on your list cost at least 80. Not to mention all the watercooling solutions you mentioned, and sub-performing TRUE ripoffs.

    Anyway, its about suggesting some solutions, and not arguing if we're supposed to stick to the $60 limit like crossing it is the end of the world.... so I'm no longer going to post on this particular thread.
  6. dividebyzero

    dividebyzero trainee n00b Posts: 4,891   +1,264

    Hi rogue12,

    If you are still following this thread.
    As CMH has stated- if you indeed have a good quality fan of low noise characteristics "sitting around doing nothing" then the TRUE does need to be included in the shortlist of coolers to be considered. If you have no 120mm fan already then you will need to add $20-30 to the
    budget (i.e. $89-99).
    Contrary to CMH's beliefs the list of coolers I listed (both in the sub $60 category, and in the subsequent list) are more than capable of cooling a Q9550 at both stock and overclocked levels ( dependant upon cpu voltage, chassis cooling and layout and airflow). Differences in cooling effectiveness would amount to a few degrees and are heavily dependant upon the type of fan (and indeed number of fans) used.

    On a personal note- I offer advice based on the input of the person requesting the advice- as CMH stated "Anyway, it's about suggesting some solutions...", unfortunately when that altruism is tempered by rampant fanboy statements such as "...sub-performing TRUE ripoffs" it is rather more a case of undermining what we are trying to achieve here.

    If anyone reading this thread can substantiate CMH's assertions that any of coolers I listed for rogue12's benefit are "sub-performing" at their price point then by all means add your voice- I stand to be corrected.
  7. Moltar

    Moltar TS Rookie Posts: 55

    I know I might get shelled for sudgessting this but I am currently running Corsair's H50 in a push/pull configuration. I have a Q9550 running at 4.1GHz ( 470 x 8.5, 1.28v in BIOS) at 35-38 C idle, 43 C normal load, 48-56 under prime95 stress testing. Its not much better than a standard air cooler but it does give your case a much cleaner look.

    You can routinly find these coolers around the net at around 60-70 USD.
  8. dividebyzero

    dividebyzero trainee n00b Posts: 4,891   +1,264

    I personally thought it would be a good possibility so included in the second list I made mainly because the kit is complete-all you need is included.
    When you say push/pull - you have a fan either side of the rad pulling air into the case? or the one included fan pulling air into the case and a top (or side) mount fan exhausting?
    Nice temps. Some quads I've had wouldn't reach 470 core freq. let alone at 1.28v!
  9. Moltar

    Moltar TS Rookie Posts: 55

    I do have a fan on either side of the rad, but I also have a fan shroud between the intake fan and the rad. This shroud increases the air pressure allowing for better cooling. After i installed the shroud I noticed a drop in my temps by 2-3 degrees.

    I'm actually pulling air from the inside of the case and exhausting it out through the top. I have not tried reversing the airflow yet to see what kind of performance it yields.
  10. rogue12

    rogue12 TS Member Topic Starter Posts: 47

    ive decided to go with this cooler due the lower noise and good cooling based on reviews and dividebyzero recommending it.

    another question to, heres a picture of my setup, would adding any sides fans help with cooling in anyway, to lets yous know front fan bottom in case pulling air in, other one at back is exhaust. the case is a thermaltake m9.

    thanks for helping.
  11. dividebyzero

    dividebyzero trainee n00b Posts: 4,891   +1,264

    Mounting a window fan in the lower of the two available positions would help keep your graphics card and northbridge cooler- dependant upon the fans ability to move the air.
    I presume you're ditching the P4 duct in the top fan position-and mounting a fan there would depend upon clearance being available between the fan and the top of the heatpipes.
    I don't know much about the chassis thermals and when faced with a new case/cooling solution I usually mount a fan controller (at least temporarily) and stick the temp probes at various points in the chassis (harddrive cage, northbridge, mosfets etc.) then run combinations of fans (in both exhaust and intake when dealing with side and top mount) to see what works best.
    Sometimes having side mount fans can hinder cooling as the crossflow creates turbulance that doesn''t allow the airflow to run optimally.
    If you do decide to use side fans then it would be a good idea to invest in fan filters as well (some manufacturers like Silverstone package some of their fans with fan filters).

    I think the Xigmatek is a good choice of cooler-and if you are considering upgrading to a LGA1156 or 1366 socket at a later stage and keeping the cooler then the Crossbow adapter is available for those mounting solutions.
  12. rogue12

    rogue12 TS Member Topic Starter Posts: 47

    im a little worried about the weight of these coolers, my mobo is a asus p5kpl-cm the Xigmatek HDT-S1283 wont break it in half will it?
  13. dividebyzero

    dividebyzero trainee n00b Posts: 4,891   +1,264

    You should be fine.
    Most tower coolers weigh upwards of 700 grams with a fan attached. The motherboard will flex slightly when you move it around but once it's secured to the case you should be all good.
    If this is your first aftermarket tower cooler, you're probably worried that the cooler is going to wrench itself free and fall off. Rest assured that wont happen.

    Some coolers do include rear motherboard brackets and /or cross bracing but generally thats for coolers that are heavy to start with (or taller), or because they have options to mount 2 or possibly 3 fans, thus adding more weight, or because their method of securing to the motherboard requires the backplate to anchor the screws/studs to stop them pulling through the motherboards PCBand possibly damaging the socket.
  14. CMH

    CMH TechSpot Chancellor Posts: 2,039   +9

    Just a sidenote for anyone who wants to run overclocked quad cores: a very quick and simple comparison you can do between 2 heatsinks is the number and size of the heatpipes involved. Even better if you have data on how much heat each heatpipe is rated to.

    Using the analogy of a heatpipe is a water-pipe; the amount of heat you can dissipate from your processor is directly proportional to the number (and size) of heatpipes the heatsink has. U-shaped heatpipes don't effectively double the amount of heat that can be dissipated by the heatpipe, but they would spread the heat out more evenly around the fins. (FYI, just about all tower style heatsinks use U-shaped heatpipes, if not all). The shape and make of the heatpipe would make a difference as well, but of course, that would go into more detail than I want to start explaining.

    On a 45W CPU (or GPU), anything more than 3 heatpipes would not make much of a difference. Any review site which puts benchmarks which are dependent on Dual-core CPUs running stock I would immediately shun (specialty sites like SFF are exempt of course).

    Once you're talking overclocked quad-core, the difference between 3 heatpipes and 6 heatpipes is the difference between heaven and earth. Find me a review which says different.

    I agree that I might be a thermalright fanboi, but I stand by my views that some of the heatsinks mentioned in your post is absolutely rubbish, especially when the TRUE rev.C is available at AUD$49 right now (on special).

    (BTW, if anyone can be bothered digging through all my posts, you'd find that I have recommended other TRUE-style heatsinks which are cheaper, with the same performance, but unfortunately at this time, no longer available.)
  15. dividebyzero

    dividebyzero trainee n00b Posts: 4,891   +1,264

    I love a challenge....
    Here goes...

    and here's one with the staunch 3 heatpipe Red Scorpion going up against the mighty TRUE (Q6600 @ 3.3GHz)

    As a passing note. The TRUE is a good heatsink, unfortunately the performance is not out of the box in many cases. Bases on many Thermalright heatsinks are as flat as Kim Kardashian's rear end. To get the best out of them (not Kim's rear end...maybe) they need lapping- time consuming and not a job to be undertaken lightly for an average computer user. Most TRUE owners I know lap the base in any case as it comes Nickel plated, which has a quarter of the thermal conductivity of Copper at the 30-70C temp range (Copper 398 W/mK versus Nickel 90W/mK).

    Number of heatpipes on it's own means squat. Heatpipe cores come in three main varieties;
    Sintered Copper Power - the best
    Groove - middling
    Multilayered Metal Mesh - Wal-Mart special.

    3 Sintered Heatpipes beats 6 mesh heatpipes (but not a straight flush at Vegas)

    It's also best to take into account fin surface area, fan number and actual (as opposed to stated) airflow, whether the cooler is optimized for CPU cooling only or vents airflow over the MCH/Chipset, mosfets and RAM, and for those who swap out componentry regularly, ease of install/uninstall- including coolers that use adhesives to secure backplates.

    The moment you typed "I agree that I might be a thermalright fanboi" your argument became invalid.
    I presume you can also provide reviews, and not just your anecdotal personal thought, on these supposed "rubbish" coolers? Without providing some degree of empirical proof of your assertions it just seems you are unfairly maligning a range of products.
    Also, please don't let our little contretemps dissuade you from answering my plea should I have recourse to ask for assistance in the forums. If I were in desperate need of advice on say, what colour novelty flash drive should I purchase I should expect you to weigh in with the full force of your knowledge.
    Cheers and happy Googling.
  16. dividebyzero

    dividebyzero trainee n00b Posts: 4,891   +1,264

    A quick list of reviews....
    Criteria used: Review used quad-core CPU at stock and overclocked levels. Review included either direct testing against the Thermalright Ultra 120 or against coolers of approximate performance that can be cross-linked to the Thermalright solution itself.

    Thermolab Baram
    Noctua NH-D14 (head-to-head comparison between NH-D14, NH-U12P and TRUE)
    Sunbeamtech Core-Contact Freezer (CCF v TRUE)
    Titan Fenrir
    (includes results from most of the coolers mentioned in the thread)
    Zerotherm NV120 Nirvana
    Thermalright IFX-14
    Prolimatech Megahalem (Megahelem v True)
    Cooler Master Hyper 212

    Reviews for the Zalman 9900LED not included as it is not a tower cooler, and by CMH's definition, not a sub-performing TRUE rip off
  17. CMH

    CMH TechSpot Chancellor Posts: 2,039   +9

    Lol. Just because I'm a fanboi makes all arguments invalid?

    Like I mentioned before, I'm not gonna get into a pissing contest with you. FYI, I find those 2 links interesting, and they do prove my point, especially in the second link where they used a much more powerful fan on the Xigmatec than on the TRUE. I'm sure such seemingly minor inconsistencies would be found in every single link mentioned in your last post.

    However, since I'm not interested in a pissing contest, I withdraw everything I said about your choice of coolers being subperforming.

    fyi: the thermalright IFX-14 is basically a TRUE with an extra component.

    Okay, pissing contest now over, I'll just clarify something about the different types of heatpipes available.

    What you mentioned is true, those are the 3 main varieties, but AFAIK, all the better coolers use the sintered kind, due to the unpredictable nature of heatsink placement (upright, upside down, etc) by the end user. Didn't personally bother to check on sub-$50 heatsinks.
  18. dividebyzero

    dividebyzero trainee n00b Posts: 4,891   +1,264


    You (and others that read the reviews) would have noticed that in many of the reviews that the TRUE comes out on top. No accident, and I certainly wasn't looking to cherry pick reviews that cast the TRUE in a bad light. As I mentioned before, I hold the cooler in high regard.
    The purpose of listing the reviews was simply to show how close the coolers were to each other in cooling ability. I use the stock Intel HSF (as it's free) as my baseline then all other coolers are on a sliding scale regarding temps. 2-3C at idle and 0-5C at full load with a 30% overclocked quad core I would deem, for all intents and purposes, the same in performance.
    From Day 1 these coolers are going to start collecting dust over their fins' surface area, and the denser the fins and the higher the surface area (one of the trademarks of a top-of-the-line cooler) the quicker their cooling efficiency degrades- A couple of months in an average chassis is likely to add a degree to the CPU temp. And given the state of most systems and how often the coolers are removed and cleaned (or even blasted with liquid air/contact cleaner) the coolers' differences usually average out in my experience.

    As I said earlier, I hold the TRUE no grudges, if it had been in the price range at the time I checked the site then I would have certainly recommended it- But with novice computer users I tend to look at all-inclusive setups first. The simpler the better as a relatively inexperienced person,, given the multitude of fans, thermal grease, fan clips, anti-vibration mounts likely to turn a good cooler into something a little less than I envisaged for them.
    The TRUE, especially in it's Rev.C incarnation I would heartily recommend to someone with a reasonable grasp of system building-if not in practice, then at least willing to jump in. Someone whose happy balancing a motherboard while affixing a backplate, who can shop for a good fan or two and not just one that says 60, 70, 80...cfm. I'll happily point out that the Thermalright range are better constructed than most, the fins are soldered to the heatpipes and not just thermal glued and pressed, and that from an aesthetic viewpoint the design looks much more elegant than most of it's competition- but if the prospective buyer is not overclocking, then I'll just as happily tell them that the stock cooler works, or that there are other solutions that are viable (even if not the best) from a cooling, fitting, maintenance and sound point of view.
    The TRUE is turning into a more cost effective option for those looking for a little bling or a mild-medium overclock now that it's successor, the Venomous X is starting to become available, and will likely be selling a lot cheaper than it's direct competitors.
Topic Status:
Not open for further replies.

Similar Topics

Add your comment to this article

You need to be a member to leave a comment. Join thousands of tech enthusiasts and participate.
TechSpot Account You may also...