We enthusiasts are a picky bunch. Not only PC enthusiasts, but enthusiasts of cars, cooking, tools, etc. know that quality equipment is necessary to get the most out of a product. All of us are trying to reach the nexus of superiority for the products we enjoy the most.

We know that PC enthusiasts religiously overclock their systems to various degrees. Overclocking is becoming a well-established art, with a market behind it that is very supportive. Homegrown cooling modifications have, for the most part, given way to solutions that can be purchased at your favorite online and brick-and-mortar retail establishments.

Although I may not be the most skillful overclocker, I have had experience overclocking a variety of systems. My first all-copper CPU heatsink/fan allowed me to reach 2.0GHz with an Athlon Thoroughbred-B 1700+ on a VIA KT266A-based system a few years ago. I moved on to the nForce2 platform and achieved similar results with Socket A processors based on the Thoroughbred, Barton, and Thornton cores.

In the Raw

Moving to the Athlon 64 platform, I initially went the Socket 754 route with the 3000+ and a VIA K8T800 chipset based motherboard. Unfortunately, my overclocking options were limited and I was content using the retail CPU heatsink/fan. Wanting a bit more power and speed, I ventured to the Socket 939 and with this new format, hope to achieve overclocking levels similar to my previous Socket A setups.

This review will focus on a new product from SilenX, one of the most well established names in PC cooling and overclocking. They are best known for their line of quiet and cool running power supplies, as well as their heatsink designs, which also focus on optimal thermal management and low noise levels.

SilenX's iXtrema 120mm Copper CPU Cooler just hit the market. As you can see by the picture, the product is quite large and those who know their heatsinks can easily infer that this visual representation indicates a high level of performance. With a retail price of $49.95, the SilenX iXtrema is neither the cheapest nor the most expensive fan on the market.

The Package

At first glance, I found the sheer size of the unit a bit intimidating, which made me wonder if it would fit in a regular ATX case. The weight was also a concern, but the mounting bracket is very secure.

Moving on to the "noise" blurb, 12-24 dBA should be music to anybody's ears. I must add that this cooler is certified to run both Athlon 64 4000+ and 4.0GHz Pentium 4 processors in QUIET MODE. That is quite a claim, which makes the product very attractive to those who desire effective cooling without having to resort to an exotic, and expensive, water or phase-change setup.

• First heatsink to come with standard 120mm fan for increased airflow
• Large surface area for maximum heat dissipation
• Variable controller to manage fan speeds
• Thermal paste included in package
• Supports all socket 462/478/754/775 platforms



• Dimensions: 140x140x60 mm
• Base Material: Copper
• Weight: 785g
• Noise: 12-24 dBA
• Rated Voltage: 12v
• Speed: 1000~2200RPM
• Air Flow: 37.5~77.5CFM

This heatsink bears a striking resemblance to the Zalman CNPS7700-Cu. There are differences, however as the SilenX is 133 grams lighter, 4mm wider, and the fins feature a ruffled, rather than straight shape. One can infer that the ruffling is used to increase surface area.

The product is packaged securely. On the backside is a little bubble in the packaging that protects the surface of the heatsink. While removing the fan from the box, I was greeted with the following items: various sets of brackets, a replacement backplate for the Athlon 64, all necessary screws, washers, and standoffs, and the product manual and installation guide.

Closely inspecting the heatsink revealed a flat and properly milled surface. No surface divots were present and there was a minimal amount of surface rippling. It is possible, however, that an 800-1500 grit lap job could improve functionality, as I've seen better surface slickness on other units, such as Kingwin's Socket A 70mm Copper CPU Cooler. I didn't modify the SilenX for this review and it was installed as it came right from the package.

Accompanying the heatsink is a small module designed to plug into a motherboard fan header, which provides active control of fan speed. Here's a picture of it in discrete format.

Fan Control

Controlling fan speed has become an integral part of enthusiast overclocking endeavors since noise is typically a concern. Those of us that have used a Vantec Tornado fan can remember how intolerably loud it became. The SilenX iXtrema allows for fan speeds to be adjusted on the fly using a dial on the side. I tested my Zalman 80mm fans, which had the quiet mode resistors unhooked, to determine if I could hear a difference when the dial was turned from the minimum to the maximum. Sure enough, there was a drastic difference in speed and a minor increase in the noise level.

When I say this products is compatible with all current AMD and Intel platforms, I do mean all of them including: Socket A/462, 754/939 and 478/775. The unit comes with all the necessary brackets and hardware, along with clear instructions for each application.

Installation on my rig was absurdly simple. I replaced the plastic backplate with the SilenX-supplied unit, applied the shims to the nipples, screwed them in, and then screwed the heatsink to the nipples. I applied thermal grease to the heatsink before installation, but I used Artic Silver 5 in lieu of the supplied compound.

After securing the heatsink, the fan speed controller was hooked up and mounted using double-sided tape. Once everything was mounted, I installed the motherboard into my "standard dimension" CyberTron ATX case. Worth noting: I did have to pull out my RAM sticks before I installed the heatsink, as there was a clearance problem with the first DIMM slot. Ironically, in my torture testing I found my RAM to be more stable in slots 3 and 4 anyways, so as it turns out for me, there are NO clearance issues with drives, cables, memory or the power supply.

Next Page: Test System, Results, and Conclusion