Zalman LQ Series CPU Liquid Cooler Fan (LQ-310) $29.99 after $35 Rebate + Free Shipping at Newegg

I personally would not worry about GPU cooling as most good GPU's already have decent cooling and trying to liquid cool the core will only be a hassle. You will have to buy VRM and Memory coolers or chop up your stock cooler and apply it to those areas.

I have 2 GTX 465's max voltage in MSI afterburner overclocked from 627MHz to 751MHz core for daily usage. By the time you kill the GPU's from heat/voltage it's already time to upgrade. Mine run in the high 70's°C under full load.

Anyway I just got finished lapping my LQ-310 to 3000grit sand paper and it does work better than my lapped Corsair A70 air cooler. I didn't know my ambient temps though so no definitive results...AMD phenom 975@ 4.2GHz dropped ~5°C from the Corsair A70 at full load (Intel Burn Test)

Thanks, Chris

Quote from flashfir : This is a good deal... given the fact that you're probably not a high-end enthusiast it'll do just fine. However, to claim that FrostyTech is reputable causes me to doubt whether you should be advising other slickdealers on the relative performance of this CLC. Anyways... Overclocking with "liquid" systems is different. Just because it's liquid doesn't mean it's better. There's a whole slew of talk about Closed Loops and how they perform compared to similarly priced Air-coolers, Tom's did one with the Noctua D14 (dual massive heatsink) and some of the newest H100i + other closed loops. What it comes down to is air coolers almost always provide better cooling/noise/$. Highend CLC's like the H100i have to have fans which spin a lot faster (400-800RPM on top of 1200RPM) to get the few degrees of lead they get over high end air. Enough about that, if you want come swing by ocn and take a look around in terms of threads on closed loop coolers. Tons of those all the time "should I get H60... etc..."

Yes, I've read several reviews on the Corsair closed loop systems at multiple sites and that's the other reason I've hesitated - as the benefits versus the "taking out my motherboard, removing my Xigmatek Gaia and the mounting bracket, removing my rear-mounted 120mm fan, and then installing this..." is a major caveat as to why I'm never puchased one of these. I'm not really much of an Overclocker, I pretty much am only interested in what I can do without a voltage increase, as you increase heat you decrease component lifespan. Until I had kids, I ran my cases without the side panels.

Every site has it's flaws - for example, Tom's Hardware had some hard case reviewers pushing the G-series and Celeron socket 1155 CPU's over anything sold by AMD. They chose very specialized tests to "prove" they point - I posted on that back in 2012 with some of the Llano deals, at a low price point with inexpensive HD 6670 the A6 could provide a very nice price/performance ratio for budget gaming. I didn't discount the reviews, but like some Consumer Reports reviews back in the 80' and 90's where the Japanese car was always better - you had to look at the models they chose. There was an element - actually a strong element - of truth to it, but there were multiple times they chose comparisons that never gave the US built car a chance.

I'm saying this so you could see my own biases - what's the problem with FrostyTech? I could research it but I get the impression this is something you've paid attention to for a while, and there is no sense reinventing the wheel. I've posted multiple times over the years on CPU cooler deals here, and I've never noticed anything hugely anomalous on their results when I pull up other reviews for multiple sourcing, which I strongly believe in, or for interpolating results for comparing two coolers that aren't compared directly in any one review.

FrostyTech's relative rankings pretty much match my own results, and I've had - OMG let me try to think - well, I have eight different coolers in service in the house, at least five different OEM coolers I removed still somewhere, and this doesn't include all of my past ones. I like the FrostyTech process, with the fixed wattage heating plate technique (my lab experience is chemistry and biology). I realize that can have limitations, especially in more recent years with multi-core processors with auto-throttling individual cores based on work load and temps - but I haven't bothered to research if the heat plate test has become useless. Is the difference between a heating block and a CPU running at 100% that significant these days? Or does FrostyTech get a little too cozy with certain manufacturers if they take out ads, i.e. no ads or free cooler and you stuff won't get tested? Curious, as I can use other sites but FrostyTech's huge list of coolers is really handy at times.

I just received this and did a load test before and after -

i7 950 stock clock
CPU under 100% load with a Cooler Master Hyper N520 - Average Core Temp = 70'C
CPU under 100% load with LQ310 - Average Core Temp = 50'C

after about an hour under prime95 torture test the average temp did go up to 60'C, but for $30 i'm pleased with the results

have anyone got their rebate from the last time?