Working in concert with ASEtek, a worldwide leader in liquid cooling systems for high-performance computing, Corsair has developed an easy-to-install CPU cooling solution that outperforms stock CPU coolers by more than 30%. Utilizing a copper CPU cooling plate and integrated pump connected to a 120mm radiator and high-efficiency, low-noise fan, the Hydro Series H50 delivers superior cooling for higher overclocking performance and greater stability. Unlike traditional liquid-cooling solutions, the H50 is a closed-loop system that is supplied pre‑filled and fully connected using low permeability tubing to virtually eliminate evaporation and leaking.

“Until today, most innovations in CPU cooling revolved around how to fit increasingly bulky and unmanageable heatpipe-based air coolers into a chassis,” said Jim Carlton, Vice President of Marketing for Corsair, “The new Hydro Series H50 changes the game, packing an incredible amount of cooling into a much more compact solution and delivering better performance than even high-end, dual-fan CPU coolers that barely squeeze into a system.”

Tests performed in the Corsair Lab show significant improvements over traditional air-based CPU coolers. Using a test system built using a popular gaming case and an Intel Core i7-965 processor overclocked to 3.46GHz with a vcore of 1.3V, the following results were obtained:

The Corsair Cooling Hydro Series H50 comes complete with the copper CPU cooling plate and pump unit, 120mm radiator and fan, and mounting brackets for both Intel LGA775 and Intel LGA1366 socket motherboards. Brackets for AMD socket AM2 and AM3 motherboards are available for a nominal shipping and handling charge by contacting Corsair Technical Support.

The Corsair Cooling Hydro Series H50 will be available by the end of June from Corsair authorized resellers and distributors worldwide, and is backed by a two year limited warranty. Complete customer support via telephone, email, forum and Tech Support Express is also available.

The word Apogee means the farthest or highest point according to Merriam-Webster. With this definition in mind Swiftech gave this name to their top of the line water block. The Swiftech Apogee under various sub-names (GT, GTZ etc.) is one of the best water blocks out.  But can the Apogee keep Intel’s newest i7 965 cool running full out?
Read on to see if the Apogee GTZ can maintain its orbit or if it crashes to earth in flames.

Product: Swiftech Apogee GTZ for Core i7 (with H20 220 Apex Ultima)
Author: Sean Kalinich
Reviewed on: December 15th 2008
Product cost: $$69.99 (plus $9.95 for i7 mounting kit) $300 for Complete H20 220 Apex Ultima Kit
Manufacturer: Swiftech
Spelling and Grammatical editor: Planetx64 Staff

{mospagebreak title=Differences}
Differences:
The Apogee GTZ, according to Swiftech, was designed to improve thermal interface material joint thermal resistance by up to 300% compared to older designs. This is an impressive number but is also based on the 775 and 771 sockets from Intel. The i7, socket 1366, although similar has a much larger area to contact and to maintain equal cooling across.  The difference in CPU and socket size also mandates the need to a different mounting bracket and back plane.
In almost all other respects the Apogee GTZ for i7 is identical to the one for the Socket 775.
For more information on the Apogee GTZ visit Swiftech.com

{mospagebreak title=Performance}
Performance:
I know this is what you really want to know about so I will dive straight into the performance side, can the Apogee GTZ keep an i7 cool?

To test this I used a full H20 220 Apex Ultima kit complete with the Apogee GTZ for i7.
This was installed into a Cooler Master ATCS 840 case with the following other hardware:
Intel Core i7 Extreme 965 @ 3.2GHz and 3.97GHz
Asus P6T Deluxe OC Palm Edition
Kingston Tri-Channel  DDR3 2000MHz kit (3x1GB)
Seagate 147GB 15k RPM SAS HDD
2x Zotac AMP! Edition GTX 280 1GB (SLI)
Cooler Master Real Power Pro 850 PSU
Windows Vista Ultimate x64 W/SP1
Temperatures were taken with Core Temp.

I compared the Apogee to the Thermalright Ultra-120 eXtreme RT Air cooler. This high end tower cooler has been able to keep the i7 965 happily running at 3.9GHz for over a month with no thermal issues. I felt it was a good comparison for the Apogee and should give you an excellent idea on performance.

The Tests:
Idle temps were logged after the system was powered on and sitting for 45 minutes.

Load Testing:
HyperPi 0.99b
HyperPi was used as it is able to push both CPU cores up to 100% and keep them there for extended periods. I ran back to back 32M (8 core) runs and recorded peak temps.
AutoGK
AutoGK was used to transcode a full 2 hour movie from DVD source to Xvid at 100% quality; again maximum temps were recorded during the Transcoding.
Lightwave 9.3 x64
For my Lightwave load testing I rendered frame 32 of the moonbase scene (found on the Lightwave 8 content CD1) I used a resolution of 1920×1080 (1080i HD), a 7-Pass PLD AA filter and a Guassian Sharp reconstruction filter; as with all other tests maximum temperatures were recorded.
Gaming:
For Gaming I ran Crysis Warhead for one hour and recorded the high temps for each CPU
Overclocking:
My Overclock was at 3969MHz with the base clock at 189MHz Multiplier at 21. CPU vCore was 1.45 VDimm 1.65 all others set to auto.
Once the stable OC was achieved I ran through all of the tests again.

Thermalright Ultra-120 eXtreme RT

Apogee GTZ

Temps are recorded as Core0/ Core1/ Core2/ Core4. All temps are Celsius.

Notice that the Apogee performs much better under sustained load than the Thermalright Ultra-120 eXtreme. This is due to heat saturation. As I am sure you all know, when a system is under load everything puts out more heat. This causes the air in the case to become hotter, as the internal air temperature increases the cooling ability of any air cooler begins to decline as the fins on a heatsink will only cool down to around the internal ambient temperature. Since the CPU is located very close to other major heat producing parts (GPU, Northbridge, and power regulators) it receives a very concentrated doses of this hot air.
Conversely, since the radiator on most water cooling systems is away from the main concentration of heat (right over the CPU, NorthBridge and GPU) it is able to keep the CPU under the internal ambient temp and lower overall case temperatures.
Of course water cooling effectiveness will depend on the size of your radiator and external ambient temperatures. The larger your radiator is the more efficient the cooling will be.

{mospagebreak title=Installation}
Installation:
The H20 2020 Apex Ultima kit that I used for this test was one of the easier water cooling rigs to install. This was mainly due to the CoolerMaster ATCS 840 which has a significant amount of room to maneuver in. The block itself was very easy to install and was actually much easier than the Ultra-120 eXtreme.
With the Ultra-120 eXtreme I had to remove the fan and then tip the case onto its side to mount it. Due to the need to use a long thin Phillips head screwdriver I had to be very careful not to slip as the pressure needed to ensure the screws connected with the back plate was pretty heavy.
The Apogee did not require any screwdrivers, I was able to use my fingers and still get a good firm mount.

{mospagebreak title=Value}
Value:
The Apogee GTZ all by itself (with the 775 mounting bracket) is only about $70. When you add the i7 mount you push this adds an extra $10 to the cost. In all not a bad price for the extra cooling if you already have a water cooling setup. If you are starting from Scratch the H20 220 Apex Ulitma kit that I used for test will set you back $300.00. Of course you will not have to buy the Apogee separate as it is part of the kit, but you will probably have to spend the extra $10 for the i7 mount.

{mospagebreak title=Conclusion}
Conclusion:
I felt that the Apogee GTZ, combined with the Swiftech H20 220 Apex Ultima kit did a great job of keeping my i7 965 cool. This was very evident under sustained loads especially gaming where the GPUs started to heat things up. The installation was fairly simple and very straight forward. Filling the kit was also easy with the provided instructions even a first time builder would be able to safely setup the kit.
As far as price goes the block on its own is a very good deal if you already have a water cooling rig. However if you are starting from the ground up the $300 is somewhat pricey considering the cost of the i7 and other parts.
In the end if you are looking for a good quiet cooling system and water is your thing, well you really won’t get much better than Swiftech’s Apogee GTZ. When you combine it with the H20 220 Apex Ultima you get very good performance indeed.  

Discussion Link
 

CoolerMaster is fast becoming a regular here at Planetx64. We have had their HAF932 case (which is still one of the best cases I have ever worked with) The UCP 1100 PSU, and the GeminII S cooler. These have all be great good products and have show that CoolerMaster can put out a range of solid products.
Today we have another cooler from CoolerMaster, this time it is the V8 CPU Cooler. The V8 gets its name from having 8 heat pipes to help transfer the heat away from your CPU. Will these 8 pipes be enough? As usual ready your “left click” finger and read on.

Product: CoolerMaster V8 CPU cooler
Author: Sean Kalinich           
Category: Enthusiast          
Reviewed on: September 28th 2008
Manufacturer: CoolerMaster
Spelling and Grammatical Editors: Jason Hambly

{mospagebreak title=Packaging and Accessories}
Packaging and Accessories:
The box that V8 comes in shows promise. Is it s glossy black with artistic images of the V8, as well as embossed red lettering.

Opening the box shows the V8 itself encased in a plastic cocoon.
Inside we also find mounting hard ware for both AMD and Intel CPUs and a small tube of thermal paste.
CoolerMaster has also chosen to include a small instruction leaflet in multiple languages and for multiple CPU types. Be careful of the illustrations though, they show that you should mount the V8 with the lettering upside down in relation to the top of the board, but this puts the fan intake side at the back of any case, if you do this and you have a fan there you will have to remove or turn off that fan. I recommend installing with the lettering upright. This way any existing case fan can assist in air flow through the V8.

{mospagebreak title=Design}
Design:
As I mentioned earlier, the V8 gets its name from the fact that it uses eight heat-pipes to remove heat from the main CPU block. Theses pipe rise in pairs into four sets of cooling fins. These sets of fins crisscross each other and are located on opposite sides of a central 180mm high-air flow adjustable fan. This fan is setup to pull air in through one set of fins and blow out through the other. The fan is capped by a plastic topper with the CoolerMaster Logo and the V8 name. This cap also hides red LEDs that light up when the fan is in use. According to CoolerMaster this design will handle up to 180Watts of heat, this means (on paper at least) that it should be good enough to cool even the hottest of stock Intel and AMD CPUs.

After taking a close look at the V8, the design (to me) looks a little ungainly; the crossing of the cooling fins gives the impression  it will impede airflow in addition the central fan will be pulling in cool air but blowing hot air out. All of these factors would seem to make a very inefficient design, but as they say looks can be deceiving so we won’t know for sure until we put it on the grill.
Installation is exactly the same as the GeminII S

{mospagebreak title=Performance}
Performance:
For testing I used the same setup and CPU as in the GeminII S cooler evaluation.
Temperatures shown are the highs for idle, and each type of CPU load. I did not note the minimum or average temps as they are not what is going to kill your CPU it is that Maximum temp spike that will do the job. Ambient temperature was 27.5c.
For CPU load I used HyperPi 0.99b, AutoGK 2.45, Lightwave 9.3 x64 and Lost Planet DX10.
I also used multiple temperature applications to read the CPU temps:
RealTemp – Core0/Core1
CoreTemp – Core0/Core1
Asus AI Suite – Die Temp as a single reading.
Since the V8 is adjustable I also recorded the peak temps with fan set to high, medium and at its lowest setting.

The Test System
Intel C2D E8400 (WolfDale)
Asus P5Q Deluxe
Mushkin XP2-8500 4GB DDR2-1066 (2x2GB)
Asus EAH3850 Smart OC
Western Digital RaptorX 150GB HDD
CoolerMaster 850Watt PSU (RS-850-EMBA)
Microsoft Windows Vista Ultimate x64

Idle temps were logged after the system was powered on and sitting for 45 minutes.

As you can see the V8 handily out cools both the stock Intel cooler and the GeminII S.

Load Testing:
HyperPi 0.99b
HyperPi was used as it is able to push both CPU cores up to 100% and keep them there for extended periods. I ran dual 32M runs and recorded peak temps.

The V8 takes the prize here, granted not by much and only with the fan on high but it still runs better.

AutoGK
AutoGK was used to transcode a full 2 hour movie from DVD source to Xvid at 100% quality; again maximum temps were recorded during the Transcoding.

Again we see the V8 win, this time it manages to keep the Transcoding temps under the GeminII S and Stock Intel even with the fan on low.

Lightwave 9.3 x64
For my Lightwave load testing I rendered frame 32 of the moonbase scene (found on the Lightwave 8 content CD1) I used a resolution of 1920×1080 (1080i HD), a 7-Pass PLD AA filter and a Guassian Sharp reconstruction filter; as with all other tests maximum temperatures were recorded.

Again the V8 manages to outperform the other coolers.

Gaming:
For gaming I only used one game but it is one of the few that was developed from the ground up for multi-core CPUs and also have DX10 support. I ran through about 45 minutes of gaming in Lost planet at 1600×1050 resolution 2x AA and 16x AF. I played on the hardest difficulty level for maximum CPU usage.

As with the other stock tests the V8 manages to cool the E8400 more efficiently.

{mospagebreak title=Overclocking}
Overclocking:
As with the GeminII S I did not really overclock the CPU but bumped the voltage up to 1.4v And the FSB up to 400MHz I feel that with the temps I saw you would even be good with a dual core running at up to 1.45v

The V8 does not get saturated as easily as the GeminII S did, this enables better temps while overclocked in fact my temps were below that the GeminII S had at a lower voltage.

{mospagebreak title=Conclusion}
Conclusion:
When I first saw the V8 I had my doubts about its ability to cool a CPU at all, I was concerned that it would be another GeminII S and look good but not work as well as some of the claims on the box.
After working with it and running it through its paces though I found that it is more than capable of cooling your CPU. I also feel that you would not have any problems running this on a Quad at stock and with a minor OC. CoolerMaster has put out a great tower air cooler with the V8 and one I am sure will make it into many Air Cooling enthusiasts Systems.

Discussion Link

For our latest installment of CoolerMaster goodness we are taking a look at one of their CPU coolers.
Named the GeminII S it is reminiscent of the Silver Stone NT06, but with a few differences. The CoolerMaster offering has cooling fins for (potential) extra cooling. Will this be enough for today’s hot CPUs?  Read on to find out.

Product: CoolerMaster |GeminII S CPU cooler
Author: Sean Kalinich           
Category: Enthusiast(entry)/High-End HTPC            
Reviewed on: August 26th 2008
Product cost: $35.99
Manufacturer: CoolerMaster
Spelling and Grammatical Editors: Paul Mercer

Discussion Link

{mospagebreak title=Packaging and Accessories}
Packaging and Accessories:
The GeminII S arrived in a small yet… I guess colorful is the word I am looking for. Every side had some sort of information or logo on it. The back of the box is tightly packed with images and text describing why the GeminII S is the best cooler(!).
Opening the box we find the GeminII S tucked securely away in a tight cardboard tray. Here we also find all of the things we will need to get the GeminII S installed and running.  One interesting addition is the rails for adding on two 80mm or 92mm fans. Although this is a nice touch and it should be no problem to cobble together a couple of extra fans I think that CoolerMaster would have done better to provide a method for adding a second 120mm fan for additional cooling. Either way the rails do provide you with options for cooling.
Other items in the box are mounts for AMD and Intel CPUs, all of the screws you will need and a small socket that can be used with either a flathead or Phillips screwdriver. You can use this to tighten the bolts for the base plate with. The included tube of thermal paste from CoolerMaster is not much more than standard thermal interface material, but if that is all you have it works.

{mospagebreak title=Design}
Design:
The GeminII S is a combination heat-pipe and traditional “finned” CPU cooler. There are five heat pipes that pull heat from the CPU up into a series of fins where this heat is removed by a 120mm fan.  But the fun does not stop there, CoolerMaster has also added in fins that should help remove heat from the CPU block.
To add to the cooling power of the GeminII S, you can also remove the 120mm fan and replace it with two 80mm or 92mm fans, this will allow the GeminII S to cool your CPU as well as your RAM and, to a limited degree, any passive chipset cooling you may have on your mainboard.
For mounting CoolerMaster came up with a novel design. Instead of mounting from the bottom up your screws actually mount though the top of the mount legs and extend down through the board.
From there you attach bolts to the underside of the mount bracket and use the included socket drive to tighten them up. It was a much easier job to install the GeminII S than many other coolers I have used.
On the downside you will have to remove your mainbaord to mount the GeminII S unless you pick it up with a new build.

{mospagebreak title=Performance}
Performance:
Here is where the rubber meets the road, a cooler can have all the nice features, the best looking design and even the quickest mounting, but if it cannot keep your CPU cool it is wasted money.
For testing I compared the GeminII S to the stock cooler that ships with the Core 2 Duo 8400 Wolfdale CPU.
Temperatures shown are the highs for idle, and each type of CPU load.

I did not note the minimum or average temps as they are not what is going to kill your CPU it is that Maximum temp spike that will do the job.
For CPU load I used HyperPi 0.99b, AutoGK 2.45, Lightwave 9.3 x64 and Lost Planet DX10.
I also used multiple temperature applications to read the CPU temps:
RealTemp – Core0/Core1
CoreTemp – Core0/Core1
Asus AI Suite – Die Temp as a single reading.
It is interesting to note the differences between these applications.

The Test System
Intel C2D 8400 (WolfDale)
Asus P5Q Deluxe
Mushkin XP2-8500 4GB DDR2-1066 (2x2GB)
Asus EAH3850 Smart OC
Western Digital RaptorX 150GB HDD
CoolerMaster 850Watt PSU (RS-850-EMBA)
Microsoft Windows Vista Ultimate x64

Idle:
My idle temps were taken after the system had been powered up and left to sit for 45 minutes.

Here the GeminII S did well but for some reason AI Suite showed it having a higher temp than the Stock Cooler.

Load testing:
HyperPi
For HyperPi I ran a full 32M run on both cores and checked the peak temps.

AutoGK
AutoGK was used to transcode a full 2 hour movie from DVD source to Xvid at 100% quality; again maximum temps were recorded during the Transcoding.

Lightwave 9.3 x64
For my Lightwave load testing I rendered frame 32 of the moonbase scene (found on the Lightwave 8 content CD1) I used a resolution of 1920×1080 (1080i HD), a 7-Pass PLD AA filter and a Guassian Sharp reconstruction filter; as with all other tests maximum temperatures were recorded.

Gaming:
For gaming I only used one game but it is one of the few that was developed from the ground up for multi-core CPUs and also have DX10 support. I ran through about 45 minutes of gaming in Lost planet at 1600×1050 resolution 2x AA and 16x AF. I played on the hardest difficulty level (which means I died a lot). To allow for the CPU to come into play more.

Performance conclusions:
In just about all of the testing the GeminII S was roughly 10 degrees cooler than the Stock cooler from Intel.  Not bad at all for the price.

{mospagebreak title=Overclocking}
Overclocking
Since I wanted this to be about cooling and not overclocking performance on the CPU I started off with voltage adjustments. I found that with the stock cooler even bumping up to 1.25 caused instabilities as the stock cooler just would not take the extra heat. I was very surprised by this but in the end I did not run any overclocking test with it. For the GeminII S I found that it was capable of handling up to 1.35v (I also bumped the FSB to 400MHz) on the E8400 but pushing the vCore much higher than that would probably cause you grief. I know I did not have to bump the voltage that much to get the extra 66MHz but since this is about cooling the CPU the extra voltage shows the cooling abilities of the GeminII S.

{mospagebreak title=Price}
Price
The GeminII S sells for $35.99 at NewEgg and is available from most Internet e-Stores. This is not a bad price at all considering many other third-party coolers are over $50 (and many of those do not come with fans). Also for the cost of a couple of 92mm fans you have the additional capability to cool your RAM and chipset more efficiently, in all not a bad deal for the price.

{mospagebreak title=Conclusion}
Conclusion:
I was very happy with the GeminII S although it is not the Cooler of choice if you are looking to do major overclocking it will keep things cool if you just want to bump your system up a little. In terms of noise I was actually able to hear my PSU fan over the GeminII S so you are not looking at having to deal with a screamer to keep your system cool. I would put this cooler in the Entry Level Enthusiast and High End HTPC category as far as who this will benefit the most. It is light years ahead of any stock cooler but not at the top of the heap quite yet.

Discussion Link
 

Product: Sunbeamtech Core-Contact Freezer
Author: Jason Hambly
Reviewed on: August 12, 2008
Product Cost: $34.99
Manufacturer: Sunbeamtech
Spelling and Grammatical editor: PlanetX64 Staff

{mospagebreak title=Package/Specs} 

Package:

Sunbeamtech ships the C-CF in a box with lots of pictures and information about the product.  My box arrived with a corner smashed in, but the product survived intact.

Contents:

-Core-Contact Freezer
-120mm MFDB fan
-Fan mounting clips
-Fan controller
-Intel motherboard mount
-TX-2 Thermal Grease
-Instructions

Specifications:

Dimensions (mm) – 125(L) x 104(W) x 155(H)
Weight – 590g without fan
Bearing Type – MFDB
Air Flow – 90.65 CFM (Max)
Noise – 16~20db +/- 10%
Thermal Resistance – 0.092 (ºC/W)
Fan size – 120mm

Compatible CPUs

AMD – Socket AM2, Socket 754/939/940
Intel – LGA775

The C-CF looks like your standard tower
type cooler.  An aluminum base with four copper heat pipes work together to wisk the heat up into the massive array of aluminum fins.  The fins are then kept cool by the 120mm fan.

{mospagebreak title=Installation}

Installation:

Sunbeamtech was very nice to include a full size tube of Tuniq TX-2 thermal grease with this cooler. They also suggested that twice the normal amount of grease be used. This is due to there being small grooves between the base of the cooler and the heat pipes.  Hardcore types may be inclined to lap the base to make it smoother and remove these ridges.

Installation on an AMD motherboard required no additional work be done. You just attach the clamp to the standard AMD bracket.  However that was easier said than done.  With a motherboard already installed in a mid-tower case there is little room to maneuver your hands around the enormous cooler. Getting the first side clipped was easy but the second side required a ridiculous amount of force in a very tight space.  In the end I needed to use a screwdriver and my free hand to get the clip attached. I still ended up spilling some blood when the smashed my thumb into the PSU. I contemplated a new reviewing system based on the amount of blood lost during installations but felt some readers may not enjoy the visuals.

Intel users will need to install the included bracket onto the motherboard first.  This bracket doesn’t require the motherboard to be removed, it just installs like a standard Intel cooler.  However after that you have to go through the same process as the AMD install.

Once the cooler was in place the rest was easy, the fan controller attaches to an open expansion slot in the back while the fan easily attaches to the cooler with two metal clips. I had to reverse the fan and put it on the opposite side of the cooler in my case.  On the side it was setup to go on the fan came in contact with my RAM.
 

{mospagebreak title=Performance} 

Test System:

AMD Athlon64 X2 5600+
Gigabyte GA-MA69GM-S2H mATX motherboard
2GB OCX Platinum XTC PC2 6400 RAM
Asus X1950Pro
Sunbeamtech 3DStorm case

Performance:

To test the cooling prowess of the C-CF I ran HyperPi on each core using the 16M option and monitored CPU temps with CoreTemp.  Max temps were generally reached within a minute or so but the test was allowed to run for the full duration and the max reached temp for each core was recorded.  For comparison I used the stock AMD cooler.

When the fan controller was set to Low, Idle temperatures on the C-CF and the stock cooler were equal but the Load temperatures with the C-CF were 7ºC lower.

Adjusting the fan controller to the Mid setting (This setting is marked on the dial with a notch so you know when you are half way) the C-CF ran 4ºC cooler at idle and 9ºC under full load then the stock cooler.

Bumping the fan controller up to full speed continued to lower the temperatures another couple degrees but not by as much as you may expect at full speed.  Noise was not an issue since the fan was very quiet.  Idle temps were 6ºC cooler then stock while the load temps ran 10ºC cooler.

Only having a 10ºC delta in temperatures at load compared to the stock cooler I was a bit disappointed in the performance of the C-CF.  

{mospagebreak title=Conclusion} 

Conclusion:

Overall the Sunbeamtech Core-Contact Freezer falls into the pile of coolers that work well but have some major flaws.  The ridiculously hard installation is this coolers biggest downfall. It is so hard to install that damage is a very likely outcome to either your motherboard/CPU or even yourself. Performance was not bad but not the greatest, however at only $35 US it is half the price of some of the better known air coolers (cough* Zalman).  I have in the past liked all of the products that Sunbeamtech have sent us but unfortunately this time around I have to give the thumbs down to the Core-Contact Cooler until they make it a little more user friendly.

Introduction:

Silverstone are quite a large company founded in 2003. Better known for
their PC enclosures and power supplies, today I?ll be reviewing a different
product from them in the form of a CPU cooler.

The Silverstone Nitrogon NT06 is billed as ?The ultimate in forced-air cooling
solution? on the Silverstone website, over the next few pages I?ll be seeing just
how well it can improve the operating temperatures of a Core2Duo CPU over the
stock Intel cooling solution.

{mospagebreak title= Packaging}

The Silverstone Nitrogon NT06 arrived in perfect condition, in it’s retail packaging. Encased securely in fairly stiff foam, the heatsink is unlikely to be damaged in transit.

    

 

   

Included with the NT06 are the various fixings and mounting hardware needed to fasten the cooler to either AMD or Intel CPU sockets and a tube of (dried out) generic thermal paste. The NT06 is compatible with Intel P4 Socket 478, LGA775 and AMD Socket 754, 939, 940.

 

There’s also a handy fan speed controller and of course a useful set of instructions which also has the warranty information.

{mospagebreak title= Design and Features}

According to the Silverstone Website, the base, heat pipes and
upper fins are all copper. I believe this is slight mistake, the upper fins are in fact
aluminium, which helps to keep the weight down.

Looking at the rather large heatsink, you could be forgiven for
thinking that it looks rather like a Thermaltake SI120 (minus two heat pipes)
or very similar to an AeroCool Dominator.

The size gives expectations that it should be a good performer. Short but stocky, the Silverstone NT06 is designed with a solid
copper base, with copper heat pipes to transfer the heat from the base to a
large array of aluminium fins at the top. The fins are then in turn, cooled by
the airflow from a high-pressure fan. The fans RPM (Revolutions Per Minute) can be tuned using
the separate fan speed controller, allowing you to vary the amount of airflow
(and noise) needed depending on the situation at hand.

 

As I took a close look at the base of the cooler for the first time, it appeared to be
fairly smooth with a few machining marks, this was no mirror finish, I have
seen worse bases on a few other coolers and of course the base
could be lapped if desired, but I wasn’t too impressed with the finish.

Examining the heatpipes, we can see that to maximize surface area
contact, Silverstone has opted to squash the heatpipes and solder them rather
than machining channels into the copper base which would have been the better method.

{mospagebreak title= Installation}

Installing the Silverstone NT06 cooler was fairly straightforward.
The motherboard must be removed from the case and the supplied back plate fitted to support
the weight of the Silverstone NT06 (570g). The back plate is held in place
using four screws, with plastic washers on the motherboard side for insulation
and then a hex nut on each screw. This leaves the top of the screws exposed and
forms the supports for the heatsink. 

     

Securing the cooler is made a lot easier
by the use of knurled thumbscrews as opposed to some clip mounting mechanisms
on certain other heat sinks I could mention.

{mospagebreak title= Testing and Performance}

Test System:
1x Intel Core 2 Duo E6700
1x Asus P5B Deluxe WiFi/AP
2x 1024MB Crucial 10th Anniversary PC5400 3-3-3-8
1x XFX Nvidia 7900GT
1x 80GB Hitachi SATA2 HDD
1x Stock Intel Boxed Heatsink

Testing was done initially with the standard boxed cooler to get a
base reading for comparison and throttling functions were disabled in the BIOS throughout all tests.
I had planned to run the computer in two different
scenarios; at stock speeds and overclocked… I’d never actually used the boxed Intel heatsink before and unfortunately, after testing at
stock speeds, it quickly became apparent that
overclocking with this heatsink would be out of the question.
For maximum heat
generation I used Orthos and the ‘Small FFTs’ test to stress the CPU. Temperature
readings were taken after half an hour, the computer was then left to idle for
10 minutes until the temperatures stopped falling and the idle levels recorded.
The process was repeated twice with the Silverstone NT06 cooler, once with the
fan running at the slowest speed and once again at full speed.

With a standard speed of 2667MHz and using only 1.3V, I didn’t expect the Intel boxed cooler to do so badly.

Clearly, the Silverstone Nitrogon NT06 cooler does a much better
job at cooling than the standard heatsink. Even with the fan turned down to 800 RPM it still kept up with the Intel cooler running at full speed.
When turned right down, the fan spins at approximately 800 RPM and is very
quiet, but when running at full speed, although rated 2400 RPM, the motherboard
monitor registered around 2750 RPM. At this speed, it is certainly audible to
say the least. During the overclocking tests the NT06 was left at high fan
speed, personally I could not run this fan at full speed in a 24/7 system, the
noise level is just too distracting. Although for benching, the extra pressure
and airflow can help to cool the components surrounding the CPU socket such as
the chipset and/or memory.

*caution*
when turned all the way down the fan speed controller will actually stop the
fan from spinning. This is most likely to do with the type of fan used; high
pressure/high flow fans such as the FM122 usually need approximately 5-7V
before starting.

Using the Silverstone NT06 heatsink, overclocking was definitely
possible, I increased the CPU speed to 3450 MHz with 1.3375 V (set in the BIOS)
before reaching the same 65C under load as the stock cooler at stock speeds.

{mospagebreak
title= Price and Warranty}

I found the Silverstone "Nitrogon" NT06 using froogle for less than $53
or just over ?35 and is about average for a half decent replacement cooler these days. Of course,
there are dirt-cheap water-cooling kits available for a similar price too, but
these are not worth bothering with if you care about your PC at all.

Unfortunately for Silverstone, there are a lot of other coolers out there at a similar price point, and to make an impact into the market you need to have something really special.

Silverstone offer a limited 1-year warranty from the date of
purchase.
Quote from the Silverstone NT06 manual:
"During the warranty period, assistance for replacement or exchange of
defective components is available at the place of purchase."

{mospagebreak title= Conclusion}

The Silverstone Nitrogon NT06 certainly offers a marked
improvement over a boxed Intel cooler and is worth considering as a replacement if your case isn’t large enough for one of the mammoth tower coolers out there.
When running at full speed, the fan can be rather noisy, but
luckily this can be adjusted if not too heavily overclocked. You can be sure
that the Silverstone Nitrogon NT06 will keep your system much cooler than a
stock heatsink ever could and at the same time allow for a fairly decent
overclock into the bargain. While only slightly overshadowed by much bigger
coolers, the NT06 is pretty compact and would happily fit into most cases while
still maintaining good cooling performance.

Overall it’s not bad considering everything
that you get, a decent heat pipe cooler, a powerful fan and a fan speed
controller. Couple this with the fact that it is compatible with most CPU
sockets out there and you have a pretty good package.

Pros:

• Powerful high pressure Fan
• Supplied with a fan controller
• Cools components surrounding the cpu socket
• Fits in most cases

Cons:

• Loud fan at full speed
• Requires
  the removal of motherboard to fit
• Fan
  controller sits at the rear of the PC

Specifications/Features: 4./5
Appearance: 4.5/5
Performance: 3.75/5
Price/Warranty: 4/5
Overall: 16.25/20

We would like to thank the folks over at Silverstone for sending us our sample for review.

Once again, Danger Den has graced our test bench with another one of their water-cooling products. This time around we are looking at their newest pump model.  The Mag II-LE is designed to save space and compete against the highly recommended D4 and D5 models that have been used by many over the past few years.  Danger Den has always shown high quality and high performance to us in the past, so I expect the same this time around.  But can a pump this small really hold up in a fully water?cooled system with four high performance components being cooled on the same loop? Let?s try it out!

{mospagebreak title=Specifications and Features}
Specifications:

-    Nominal voltage: 12 VDC
-    Motor type: Brushless DC*
-    Nominal power: 8 W
-    Pump bearing type: Polished Sapphire
-    Maximum head: 2.5 M (3.5 PSI, 97" H2O)
-    Size: 2.1?W x 2.1?H x 1.74?D
-    Maximum flow: 567 l/hr (150 GPH)
-    Weight: 241 g (8.5 oz)
-    Connection size: G 1/4 BSPP
-    Warranty: 30 Months (2-1/2 Years)
-    Power connector: 3-pin (fan header style)
-    50,000 Service Hours MTBF

Features:
?    Extreme Service Life : This pump is designed with extreme service life (50,000 hours*)
?    Bearings : The MAG features dual in fluid bearings, using the best material available: gemstone. These bearings are made of highly polished Sapphire. This is an extremely hard and long lasting material, often found in expensive watches, high precision instruments, and gyroscopes used in the aerospace and military industries.
?    Leak Proof : The MAG has no shaft seal, eliminating the possibility of leaking occurring due to long term seal wear. The stator assembly has a plastic barrier between the rotor and the fluid operating on magnetic coupling. The drive and impeller housing is plastic, to reduce the chance of long term fouling.
?    Brushless DC motor : The MAG features a brushless DC motor designed with an operational component life in excess of 50,000 hours. All heat generating components are thermal conductive bonded to the thermoplastic case. This provides a large heat sink area allowing the motor to run under constant high efficiency.
?    Small Size : The dimensions are 2.1? x 2.1? x 1.74". This is the smallest most relieable pump available to customers.
?    Direct RPM : The pump features a hall sensor that directly reads the rotation of the impeller.The impeller RPM can be monitored using your standard fan motherboard utilities via the fan port.
?    Power Suppy : This unit requires 12V DC and is equipped with a Tx3 connector.
?    Ultra low noise : The pump operates at a very low noise level due to the sealed motor compartment and improved pump housing design.
?    Easy Installation : Purchase the industrial velcro to mount anywhere in case.
?    Flow Direction : The MAG features a side inlet and top outlet for flow direction. The pump is now available with 3/8" OD or 1/2" OD High Flow or Perfect Seal straight fitting.
?    Ports : The MAG features a top outlet and side. The pump is now available with G 1/4 BSPP straight fitting ports.
?    Finish : The MAG features a new thermal plastic housing to ensure long term corrosion protection and motor heat transfer properties.
?    Multi-Unit : The small size and low power consumption of the MAG allows multi-unit installs. Though it will not increase the flow beyond a single pump unit (150 GPH) but it will help obtain the maximum flow rate possible in the system.

{mospagebreak title=Appearance}

Appearance

The Mag II-LE is a sharp looking little pump. When the box arrived I thought Danger Den sent the wrong item because the box was so small. I opened it up and sure enough, there was this tiny little pump. Compared to my previous pump, the D4, the Mag is about 1/3 the size. The pump comes all dressed in black with a touch of silver and the Danger Den logo on the end.  There is only one short wire that comes out of the pump and a small bracket extends beyond the bottom for easy mounting.

Appearance: 4.5/5

{mospagebreak title=Installation}
Installation:

Like any (almost any) water-cooling installation, there is a bit of work involved.  The first thing that had to be done after taking the pump out of the box was to insert the two ?? connectors to the pump, for the tubing to attach to.  Each connector came with a rubber o-ring to make a tight seal on the pump. Also in the box were two pieces of double sided tape for attaching the pump to the case.  The two pieces both fit on the mounting bracket side by side but I am pretty sure that only one would be needed to hold this little guy in place, you can save the second for another install, later. Next up comes the decision to power the pump via a motherboard three pin fan connector or using a 3 pin to 4 pin molex adapter (not included). Either method works as well as the other, but the second method allows you to power the pump without the PC being turned on, during the initial setup and filling of the system. It would have been great for DD to include this adapter even if it was just meant for this reason. All that is left for installation after that is connecting the rest of the system and filling it up.

Installation: 4.25/5
{mospagebreak title=Performance}
Performance:

Performance: 5/5
{mospagebreak title=Price/Warranty}
Price/Warranty:

Price/Warranty: 4.5/5
{mospagebreak title=Conclusion}
Conclusion:

The Danger Den Mag II-LE pump receives our Best on the Planet award.

Acknowledgements:
We would like to thanks Dan at Danger Den for sending us this water block.

Danger Den is known for their high performance water-cooling
solutions.  Over the years, they have
expanded their line of cooling products to cover just about everything that is
attached to your motherboard. In the past we have tried out their CPU and GPU
blocks, along with full kits. This time around we are trying out the Maze4
chipset block designed specifically for the DFI LanParty UT NF4 SLI-DR Expert
motherboard.

{mospagebreak title=Specifications/Appearance}
Specifications:

- Designed specifically for the DFI LanParty UT NF4 SLI-DR
Expert motherboard.
- 100% Copper base
- Clear Lucite top
- High flow 1/2? or 3/8? OD fittings
- Stainless steel hold down kit
- Machine lapped and polished beyond 1200grit
- Pressure tested before shipment to 85PSI

Appearance: 

When you are used to looking at larger water blocks, such as the TDX or NV-78,
the Maze4 chipset block is pretty darn small. However, once you get the stock
fan off of the chipset and see how small the chip really is, you understand why it
is so small.  The amount of space
available in that area of the motherboard is also limited, but we will get into that in
a bit.
 

The copper base is machine lapped to over 1200 grit and
offers a nice reflection when looked at. With some good thermal compound you will have no worries about bad
contact.
 

Like many of Danger Den?s products, the Maze4 chipset block
comes with a Lucite top that enables you to see into the block.  This doesn?t help performance any, but if you
run green dye and a UV light you can see the water throughout your loop.
 

My test unit came with ?? OD fittings to fit into my
existing ?? ID tubing system. The stainless steel fittings are the same ones
Danger Den uses on the rest of the water blocks that I have tested.

The DFI Expert Maze4 chipset block looks as well as you
would want a water block to look. With a clear top to see through and an all
copper base, the Maze4 chipset block is a great looking water block.

Appearance: 4.75/5

{mospagebreak title=Installation}

Installation: 

If there is a down side to water cooling it has to be the
installation.  Even all-in-one water kits
can be tedious and cause more headaches than necessary.  Multiply those headaches when you need to
route the tubing and find room for pumps and reservoirs for a custom kit. I am
not going to get into all of that in this review; I will simply stick to the
installation of the Maze4 block on the chipset.
 

To get things started, let’s say that the motherboard is not
in the case already.  If it is, back the bus up and pull it back out. You are
going to need to remove the stock heatsink fan by popping the two plastic clips
up through the motherboard.  The thermal
compound was so dry on my board that I don?t think it even needed the clips to
hold it on.  After a little bit of work
the fan was off and I was able to get a look at the chipset. 

After cleaning the chipset off and getting
rid of any old thermal compound, you need to insert the mounting hardware on the
board.  This involves placing the rods
through the board and putting a nut on each side of the motherboard to secure
it in place.

From here the Maze4 block slides down over these rods and is then
secured in place by a spring and a nut on each rod. Of course you remembered to
apply a small amount of your favorite thermal compound on the chipset before
securing it down.
 

If you haven’t done so already, you should have the motherboard back
in the case and be getting everything else ready to install.  In my test system, the Maze4 block will be an
addition to an existing water-cooling loop. Measure out the required tubing you
will need and attach it to the chipset water block now.  You won?t be able to do
it once you have your video card or cards installed.  In my situation I have NV-78 blocks on each
of my 7800GTs and the clearance between the Maze4 and the NV-78 on the top card
is very slim. 

I had to ensure that I
placed the hose clamps in a certain position so that I had room for the video card
to seat properly in place. Once the tubing is clamped onto the Maze4 I dropped
the two video cards in and finished connecting all the tubing. Leak test the
system and you are ready to go.  Overall
the installation of the DFI Expert Maze4 chipset block is not very hard,
but mixed with a pair of video cards around it, the installation can become a
little more difficult. It would have been great to not need to remove the
motherboard, but since the stock chipset fan is only held in place by push
clips, there is nothing on the back side to hold a water block in place.

Installation: 4.25/5
 

{mospagebreak title=Performance} 
Performance: 

How well does this thing cool?  That is the question I was wondering from the
start.  After finally getting it
installed, I just wanted to get right into loading up the system and seeing how
much cooler the chipset ran with a water block. Add to that, how much warmer the CPU and GPUs would be with added heat
to the loop.  To test the cooling
performance of the water block, I used MBM5 to record all temperatures during
idle, CPU load, and CPU + GPU load. To load the CPU I ran two instances of
Folding@Home with affinity set for one on each core. Temperatures were recorded
after a half hour of running. To load the CPU and GPU I ran the same two
instances of Folding@Home as well as running the RTHDRIBL video demo at the same
time. Again temperatures were recorded after a half hour of running. The system
was then allowed to idle for a half hour and the temperatures were recorded for
the final time. Each of these temperatures was recorded with the stock chipset
fan.  The tests were then rerun with the
chipset block placed in the loop before the CPU block and then placed in the
loop after the CPU block.  This was done
to see which way was the most effective for a system with multiple water blocks
running.

DFI LanParty UT SLI-DR Expert
AMD Opteron 165 @ 2.5 with Danger Den TDX water block
Corsair TwinX PT 2x512MB
2x Leadtek 7800GT TDH Extreme in SLI with Danger Den NV-78 SLI water blocks
2x Seagate 7200.10 SATA 3G in RAID 0
2x120mm Black Ice ExtremeII Radiator
2x 120mm Papst 4312L fans
Danger DD12V D4 pump
5 1/4" Dual slot reservoir
1/2" ID Clearflex and Tygon tubing

The following graphs show the temperatures of the CPU, GPU1, GPU2, and chipset.  The first measurement in each graph is with the chipset running with air cooling. The second has the chipset mounted before the CPU in the water loop and the third has the CPU running before the chipset in the loop. The later two measurements were taken to decide whether or not there is a performance change when the blocks are placed in different orders. Each test was run with the CPU at stock speeds and then again with it overclocked and at a higher voltage.

To my pleasant surprise, adding an additional water block to
the loop did not adversely affect the temperatures of the existing water
cooling system.  I had initially expected
temperatures to be a few degrees higher across the board with the extra water
block absorbing heat.

After finding out that the extra block would not increase
the system temperatures, I proceeded to test the ordering of the water blocks
in the system. After recording the temperatures of the system with the chipset
cooled with air I installed the chipset water block first in the loop and then
to the CPU and GPUs. Temperatures remained almost identical except for the
chipset that dropped an impressive 12-14?C in each test.

Pulling the entire system apart and reassembling in a new
order with the CPU first and then the chipset and GPUs I ran through the same
tests and recorded a new set of temperatures. To my surprise again,
temperatures dropped by a degree or two in almost every test.

Overall the performance of the Maze4 chipset water block is
going to come down to your individual water-cooling setup. However the cooling
performance it there if your system can handle the extra heat absorbed by the
water block.

Performance: 4.5/5

{mospagebreak title=Price/Warranty}
Price/Warranty: 

The DFI Expert Maze4 chipset water block retails from Danger Den for
$38.50.  There are not very many water
blocks that are specifically designed to fit current SLI chipsets so this price
is more than reasonable.

Danger Den offers a one year warranty on their products form
the date of purchase. Covering manufacturing and material defects provided the
products were used in normal fashion.

Price/Warranty: 4.5/5

{mospagebreak title=Conclusion}

Conclusion: 

Cons:
- Limited performance increase
- Water-cooling installation
- Only fits one motherboard

Appearance: 4.75/5
Installation: 4/5
Performance: 4.5/5
Price/Warranty: 4.5/5

Overall: 18.25/20

The Danger Den DFI Expert Maze4 chipset water block receives our Best on the
Planet award.

Acknowledgements:
We would like to thank Dan at Danger Den for sending us this water block for review.

Arctic Cooling, as their name would suggest, specialise in computer based cooling products. Going back a few years, to when both ATI Technologies and nVidia released their 95xx and 4xxx series of cards, overclocking started to become a viable way for home users to squeeze every last drop out of their systems. It was during this time that Arctic Cooling released their ATI & nVidia Silencer series, for which they are best known. This is not all that the cooling company offer though, as they have expanded their entourage to include CPU coolers, case fans, thermal paste and even a line of desktop chassis, which are engineered from the ground up to improve on the cooling methods of the ATX design. Due to this legacy of products, Arctic Cooling has earned a reputation among system builders as being the Swiss army knife of the cooling industry.

Today we will be looking at revision 3 of the NV silencer 5, which has been updated to include the new 7800 series of cards. As we all know, the nVidia 7800 series are among the fastest graphics cards available; with this speed the cards also have a reputation of being rather loud, to say the least. Let’s see how the NV Silencer 5 Rev3 copes when pitted against one of these graphical behemoths.

{mospagebreak title=Features}

{mospagebreak title=Packing} 

Packing:

As with the entire NV silencer range, all components have been packed inside of a transparent plastic casing. The advantage of clear packing is that all components are on show, so you really know what you are buying. I searched for a way of opening the packing by hand; but ended up resorting to a pair of scissors. On opening, I discovered a slight variation to the web store photographs, which advertised this model as coming with a tube of thermal paste. Instead of thermal paste, I found a thermal pad for cooling the core.

This was not the only change in design from some of the earlier revisions, such as the underside of the heat sink; the part which actually makes contact with the RAM/Core, was more of a matt finish compared to the elegant mirrored shine of many high performance heat sinks. Arctic Cooling’s reason for this, I have to assume, is because of the number of complaints from customers concerning conductivity issues, many of whom had taken it on themselves to use a file in an attempt to improve thermal contact. Arctic Cooling’s answer has been simply to do away with the old design and use pads for the memory, much like the ones attached to 7800GT stock cooler only giving the appearance of being more hardwearing.

One feature, which was quite a pleasant surprise to find included, was the addition of a memory plate. Although not advertised as a 512MB card compatible cooling solution, the back plate unlike the front memory pads is made entirely from aluminium, a metal well known for its thermal conductivity properties. Also included in the package are two front port plates, used for connecting to the front of the card and the cooler to a chassis. In addition to this is an Arctic Cooling logo sticker and strip of white stickers, which are used when you decide to remove the cooler and want to keep the memory pads clean.

Features / Packing 4.7/5

 {mospagebreak title=Installation}

 Installation:

It is hard to explain in words just how simple installing this cooler actually is (but I will try anyway.) Once the stock cooler has been removed and all residue has been cleaned up from the chips, simply remove the stickers, which are attached to the memory pads. With the pads removed, thread the screws through the inner most set of screw holes located closest to the core. Once done, turn the card with silencer attached on its front (fan side down) and place the protective thermal ‘X’ shaped pad with metal attachment through the screws. With the card now in place, all that is needed is to tighten on the thumbscrews, I found that using a very small screwdriver to give them an extra turn helped keep everything in place, as the thumbscrews turned by hand don’t provide the solid connection needed to hold the silencer firmly in place.

Finally plug in the fan power cable and you are ready to plug this baby into the motherboard, additionally the option to install the PCI faceplates for both the card and cooler is available. I found that the test card already fit perfectly into the motherboard, so only the cooler plate was needed in order to stop any air rebounding back inside the case.

In all, I found installing the silencer a far simpler affair than removing, the actual stock cooler. Although the instruction leaflet is sparse to say the least, it provides all the necessary information that you would need to install this cooler (after all it’s not rocket science).

The only drawback with this cooler is that of its size. A stock 7800GT cooler comes in at just under 2cms where as the NV Silencer at its largest point measured approximately 2 inches. This puts any question of running an SLI setup with one of these cards an impossibility as the second PCI-E slot on the test motherboard had been completely covered over after fitting the cooler.

Installation: 4.2/5

{mospagebreak title=Performance} 

Performance:

To test performance of the NV Silencer I have decided to use 3DMark06, each test will run using a loop five times; I decided to exclude all CPU tests so as not to give the card a chance to cool off. While running the tests the nVidia control panel’s temperature display screen will be in view on the desktop, to provide an accurate reading as soon as the tests are completed. I have also decided to include an overclocked set of results, using coolbits 2.0. For the overclocked set of results I have set the core to 497 MHz and memory to 1.10 GHz , the card of choice for all tests is a PNY 7800GT.

As you can see, there is very little difference between either of the cooling options when running on the desktop. Once 3DMark06 had been fired up, though we see a slightly disappointing two-degree difference between cards. From the over clocked results we see a more encouraging eleven-degree decrease in heat from the NV silencer.

After running each test, I also made note of how fast the temperatures returned to their ambient readings. While the stock cooler maintained a high temperature during each test and in fact for quite a while after the tests had completed, the NV silencer dropped almost instantly down to forty degrees Celsius; not only after the loop had been completed but also during each two-second break as 3DMark prepared to launch the next graphical test.

The NV silencer is also rated as running at an audible level of 0.9 Sones, while no audio reading equipment was available at the time of the review; I found the noise level to be inaudible compared to before fitting the NV silencer. Where previously, the loudest component in the test system up to that point had been the 7800GT, now it was the motherboards chipset fan causing the most irritation.

Performance 4.3/5

{mospagebreak title= Price / Warranty} 

Price / Warranty:

Pricing, as with every component, varies store to store but I was able to pick up the NV silencer for around twenty pounds (Incl. P&P) I found this price to be very reasonable. Arctic Cooling offer a Six-year warranty with all of their graphics coolers, since the average graphics card becomes obsolete after 3 years, to most enthusiasts, the cooler will at least be covered from all defects past the ownership of the card.

Price / Warranty 4.5/5

{mospagebreak title=Conclusion} 

 Conclusion:

In a nutshell, Arctic Cooling have released another great product in the form of the NV Silencer 5 (Rev.3) and one which excels in both silence and thermal conductivity. Although it has to be said, die hard overclockers may want to look at other options such as water cooling for their needs, since the results of our test indicated this cooler would be better suited to prolonging the life of your card rather than achieving a maximum overclock at rock bottom temps. We were tempted to run each test again in retrospect using Arctic Silver 5 thermal compound, but thought it would be unfair to both Arctic Cooling and our reader base as the coolant is not included in the NV silencer’s retail package.

Pros:

- Low price

- Inaudible when running

- Good conductivity

- Easy installation

Cons:

- No SLI capability