ASUS Extreme N6600GT Silencer: Extremely Silent, Extremely Hot

NVIDIA’s GeForce 6600 GT is renowned for exceptional performance at affordable price-points. While this chip is one year old and seems like it is time for it to leave, the market is still looking forward the GeForce 6600 GT and considers it to be the best performance-mainstream solution. This is a reason why ASUS decided to introduce its new passive SilentCool cooler along with its GeForce 6600 GT-based graphics card, which is, by the way, one of a few GeForce 6600 GT products with 256MB of memory.

by Alexey Stepin , Anton Shilov
10/19/2005 | 10:40 AM

Nowadays personal computers and their main components consume rather huge amount of electricity, produce a lot of heat and cause a lot of noise with their coolers. This is a payback for unprecedented power at our fingertips: all the key components of modern personal computers are made using the leading edge technologies and are designed to deliver the absolute maximum they can.

 

Designers of microprocessors, graphics processors, hard disk drives and other products are struggling to make their devices fast and not requiring powerful cooling solutions. They do their best, but at the end of they day computers still produce a lot of annoying noises. What’s the solution? Passive cooling for everything and/or low-speed fans! But are they truly sufficient and cost effective?

At least, part of this question will be answered by today’s article about ASUSTeK’s Extreme N6600GT Silencer graphics card.


Package and Accessories

ASUS does not seem to position the Extreme N6600GT Silencer as a solution for enthusiast gamers: all the gamers who liked the GeForce 6600 GT have already acquired appropriate graphics cards and today the GeForce 6600 GT is likely to appeal mostly to gamers in budget, who are unlikely to pay attention to the passively-cooled product which costs higher compared to a typical board. As a result of that, the Extreme N6600GT Silencer seems to be primarily aimed at computer users who seek for balanced performance, silent operation and are not really in budget, but prefer not to spend much.

As a result of its positioning, ASUS Extreme N6600GT Silencer has received appropriate package: a truly large box with a handle for easier carrying, similar to those used with high-end ASUS graphics cards. Previously, mainstream graphics accelerators from this manufacturer were supplied in much smaller boxes (for details see our article called ASUS N6600 GT/TD Graphics Card Review ). The design of the package resembles design of other boxes from ASUSTeK graphics cards and has an aggressive computer gamer drawn on its front-side.

While the product is not based on the world’s highest-performing chip, its bundle is very impressive, at least, for this class of devices. Along with the board users get numerous games, necessary software as well as a couple of adapters.

Have a look at the listing of supplied programs:

The list is pretty impressive for a relatively inexpensive product. Previous performance-mainstream part from ASUSTeK did not include that rich set of additional software, which once again proves that the ASUS Extreme N6600GT Silencer is not an ordinary product.

The graphics card is equipped with D-Sub, DVI-I and HDTV outputs, which is a pretty ordinary set of connectors for mainstream graphics adapters. In addition, ASUS supplies a special dongle that has TV-outputs, including S-Video, YPbPr and RCA connectors. Unfortunately, the adapter has very short wire and it is unlikely to be very comfortable to use, as we explained in some of other reviews.

Since we are talking about a product which comes from ASUS and which cannot be called a budget offering, the manufacturer supplied its own software designed for gamers with the Extreme N6600 GT Silencer. The software bundle includes:

As we see, ASUSTeK’s Extreme N6600 GT Silencer comes with a fine set of games and additional software, especially keeping in mind that this product is not supposed to be very expensive. ASUS definitely deserves a credit for supplying the same set of software with the Extreme N6600 GT Silencer as it provides with the more advanced products, such as those based on the GeForce 7800-series graphics processors.

Closer Look: PCB Design

Since the Extreme N6600 GT Silencer is the only NVIDIA GeForce 6600 GT-based graphics card we have so far seen on the market that has 256MB of GDDR3 memory onboard, its print-circuit board had been seriously reworked.

When comparing to the original GeForce 6600 GT design, we can notice that the graphics card from ASUS has different power supply circuits as well as higher number of capacitors for additional stability and reliability. For extra lifetime of the product ASUSTeK uses Japanese electrolytic capacitors from Rubycon, well-known for their trustworthiness.

ASUS Extreme N6600 GT Silencer utilizes 2.0ns GDDR3 memory from Samsung Electronics.

Given that the Extreme N6600 GT Silencer does not support VIVO and thus does not require any additional controllers, the backside of the product is empty: nothing is mounted there except memory chips.

It is important to point out that ASUS decided to cool-down memory chips on both sides of the graphics board, which may potentially improve overclockability of the card as well as enhance its lifetime.

Generally speaking, the PCB of the Extreme N6600 GT Silencer leaves a feeling of a fine and well-engineered product that will serve its owner very well for a long period of time. Let us now take a look at the main part of the design: the SilentCool cooler.

Closer Look: Cooling System

ASUS’ cooling system used with the Extreme N6600 GT Silencer graphics card is a relatively unique device. We have never seen such a construction, in fact.

The main radiator of the device is a large heatsink made of aluminum that cools-down both GeForce 6600 GT graphics processing unit and memory on the front-side of the board. The heatsink is connected to a relatively large copper radiator mounted above the graphics card using a heatpipe and a special mounting mechanism.

The mechanism allows to turn the large radiator located on top of the card by 90 degrees in order to cool it down using the flow from the central processing unit’s cooling system. Unfortunately, in case you are using a tall CPU cooler, you will not be able to turn the radiator to the position above that fan and consequently the flow from the CPU fan will not be able to cool-down the GPU. ASUS recommends that height of the CPU cooler should not be more than 11 centimeters so that the large radiator of the Extreme N6600 GT Silencer could be cooled down using the CPU fan.

Unfortunately, we have a large Gigabyte cooler used in our testbed, which is why we could not take advantage of ASUSTeK’s mechanism. Still, we believe that the device will show itself really well in case of a slow CPU fan due to natural reasons.

While the above mentioned positioning of the large copper radiator has its advantages and disadvantages, it still should be kept in mind that in case a user has a fan located on the left side of ATX computer chassis, the board will be cooled-down well even without turning the heatsink by 90 degrees. The problem is that few people seeking for silence are likely to install additional fans on sides of their computer cases.

Noise, Overclocking, 2D Quality

Obviously, ASUS Extreme N6600 GT Silencer graphics card produces no noise: there is no fan and it is even impossible to mount it onboard, like, for example in the case of the PowerColor Bravo X700 (for details please see our article called PowerColor Bravo X700 Graphics Card Review: High-End Features at Affordable Price? ). However, there are some points we should note about the product.

By actively cooling the copper heatsink located on top of the graphics card we have managed to overclock the Extreme N6600 GT Silencer to 580MHz for the GPU and 1200MHz for memory, which is one of the best results we have ever achieved on a GeForce 6600 GT graphics card.

The graphics card demonstrated amazing 2D quality on our Dell P1110 and Dell P1130 monitors: in all resolutions, including 1800x1440@75Hz we had no issues with the board.

Testbed and Methods

We used the following test system for our today’s test session with the new graphics accelerator family from ATI:

ATI and NVIDIA drivers were configured as follows:

ATI CATALYST:

NVIDIA ForceWare 78.01:

Each game was set to provide maximum image quality, similar for ATI and NVIDIA solutions. If the game offered any pre-integrated options for testing, such as demo recording and fast playback with the fps scores monitoring, we used them. In all other cases we resorted to FRAPS utility. WE tried to measure not only the average fps rate but also the minimum framerate achieved by the testing participants, which allows us to better evaluate the performance.

If the game allowed adjusting the full-screen anti-aliasing and anisotropic filtering settings, we enabled FSAA 4x + AF 16x by appropriate settings. Otherwise we forced the necessary mode from the driver of the graphics accelerator. If the gaming engine didn’t support FSAA, we didn’t run the tests in eye candy mode at all.

Besides the ASUS Extreme N6600GT Silencer/HTD, we have also tested the following graphics cards:

The following games and applications were our benchmarks:

First-Person 3D Shooters

Third-Person 3D Shooters

Simulators

Real-Time Strategies

Semi-Synthetic Benchmarks

Synthetic Benchmarks

Performance in First-Person 3D Shooters

Battlefield 2

As we have already mentioned several times, Battlefield 2 is based on a gaming engine using DirectX 9.0 features and offering very realistic physical model. However, the most important thing in this case is the extreme scalability of the scenes, which has definitely affected the requirements to the graphics memory subsystem taking into account the high image quality we are aiming at. We can clearly see that ASUS Extreme N6600GT Silencer is about 20%-28% faster than the regular GeForce 6600 GT depending on the resolution. As a result, this solution working at its nominal frequencies can compete successfully with the RADEON X800 GT in pure speed mode. The only competitor it yields to is the new RADEON X1600 XT featuring more advanced architecture and more pixel processors onboard.

The memory controller of the RADEON X800 GT is more efficient than that of the GeForce 6600 GT therefore once FSAA and anisotropic filtering are enabled, ASUS Extreme N6600GT Silencer cannot show the same excellent results any more. However, even in this case our hero outperforms GeForce 6600 GT and GeForce 6800, which feature half as much onboard graphics memory. The memory of RADEON X700 PRO works at slightly lower frequency, however, its memory controller is more efficient that is why its performance in eye candy mode is the same as that of ASUS Extreme N6600GT Silencer. If we overclock our hero, we will be able to achieve great results: in pure speed mode, i.e. without FSAA and anisotropic filtering, it will perform at the level of RADEON X1600 XT, and in eye candy mode – it will outperform RADEON X800 GT.

Chronicles of Riddick

The amount of onboard graphics memory doesn’t really matter in this game. At least the performance difference between ASUS Extreme N6600GT Silencer and GeForce 6600 GT can easily fit into the measuring error. This solution yields only to GeForce 6800 because it has more pixel processors: 12 against 8. When we overclock our ASUS card the gap shrinks. Moreover, the pure speed performance of the ASUS Extreme N6600GT Silencer overclocked to 50MHz/1200MHz frequencies in low resolutions is even higher than that of the GeForce 6800. ATI based graphics cards can’t compete on equal terms with the NVIDIA based solutions in the Chronicles of Riddick , because this game works ion OpenGL mode.

Doom III

id Software claims that the performance in Doom III benefits from large amount of onboard graphics memory, however, our benchmark results do not prove this statement, even with the Ultra High Quality game setting. Only in eye candy mode there is a certain performance difference between the two GeForce 6600 GT modifications we look at today: this performance difference doesn’t exceed 1-2fps. The performance improves much more after we overclock our ASUS Extreme N6600GT Silencer graphics card: in this case it wins the first prize having beaten even the GeForce 6800. I have to stress that all NVIDIA based graphics accelerators we tested today delivered acceptable performance in Doom III in all resolutions in pure speed mode, and only in 1024x768 in eye candy mode. Just like in the previous case when we discussed the Chronicles of Riddick, ATI RADEON X700, X800 and X1600 couldn’t compete with the rivals from NVIDIA because of the inefficient OpenGL driver.

The similar situation occurred in d3dm4 game level. The only difference is that the performance level is overall higher, because the only character participating in the gameplay here was the actual player, and there were no monsters.

Far Cry

Despite huge scene areas created in the Pier demo, Far Cry also doesn’t suffer that much from having only 128MB of graphics memory onboard that is why the results of our ASUS Extreme N6600GT Silencer are exactly the same as those of the regular GeForce 6600 GT, at least until we enable FSAA. But even with FSAA 4x and highest quality anisotropic filtering the performance gain resulting from the additional graphics memory is no higher than 10% in the maximum resolution. The overclocking to 580/1200MHz appears much more efficient for performance improvement: we get about 20%-25% performance growth depending on the resolution. In this case ASUS Extreme N6600GT Silencer catches up with the solutions featuring 12 pixel processors, such as RADEON X1600 XT and GeForce 6800.

Although the gameplay in Research demo takes place mostly indoors, a lot more depends on the graphics memory subsystem performance here, as you can see from the results obtained by our ASUS Extreme N6600GT Silencer card in 1600x1200 with enabled FSAA and anisotropic filtering. The difference with the GeForce 6600 GT reaches the impressive 30%, other than that everything is very similar to what we have just seen in the Pier demo. The overclocked ASUS card wins the race leaving behind even RADEON X800 GT and RADEON X1600 XT solutions.

F.E.A.R. Multiplayer Demo

ASUS Extreme N6600GT Silencer is only 1fps away from the reference GeForce 6600 GT, which cannot be regarded as significant difference even in eye candy mode, where the average performance lies between 11-16fps. F.E.A.R. game is extremely rich in pixel shaders that is why RADEON X1600 XT wins in this benchmark due to its 12 pixel processors and advanced architecture.

Pariah

The fantastic Pariah shooter, on the contrary, doesn’t care that much about the amount of the available graphics memory, so both GeForce 6600 GT modifications show identical results (with up to 1fps precision). Maybe the eye candy mode could have revealed certain differences, however this game doesn’t support FSAA, because of the Bloom effect. As for the results of the overclocked ASUS Extreme N6600GT Silencer, it manages to get ahead of all the competitors here.

Half-Life 2

Half-Life 2 does benefit in a certain way from the 256MB of GDDR3 graphics memory by ASUS Extreme N6600GT Silencer. Although you can really feel the benefit of having so much memory only in high resolutions when full-screen anti-aliasing and anisotropic filtering are enabled. As for the results of the overclocked card I can say that it boosts the gaming performance of our hero up to the level of 12-pipeline solutions.

We can see practically the same things in d3_c17_02 demo, however, here ASUS Extreme N6600GT Silencer is a little bit behind GeForce 6600 GT in pure speed mode and in 1024x768 resolution. The performance difference is hardly noticeable, and can be regarded as a measuring error or as a result of different memory timings. In all test modes the ASUS Extreme N6600GT Silencer working at its nominal frequencies is as fast as RADEON X700 PRO or slightly faster, and the overclocked Silencer is almost as fast as RADEON X800 GT.

Painkiller: Battle out of Hell

In low resolutions all testing participants perform almost equally fast, however, when we switch to eye candy mode with FSAA and anisotropic filtering the game turns out more sensitive to the efficiency of the onboard graphics memory subsystem, especially in 1600x1200 and up. ASUS Extreme N6600GT Silencer leaves the reference GeForce 6600 GT behind in no time showing up to 25% performance advantage in the heaviest game mode and running neck and neck with the RADEON X800 GT. The overclocked Silencer gets beyond any competition. Note that the game is not very resource hungry and even the minimum results we obtained during this test session doesn’t fall below the 65fps limit, which is more than enough for comfortable gameplay.

Project: Snowblind

The workload laid on the memory subsystem in Project: Snowblind is not very heavy, either: even enabled FSAA and anisotropic filtering do not affect the testing participants’ performance - ASUS Extreme N6600GT Silencer and NVIDIA GeForce 6600 GT perform equally fast here. Since the game uses a lot of pixel shaders 2.0, overclocking of our ASUS Extreme N6600GT Silencer appears very efficient: it provides the good 20% performance advantage over the reference GeForce 6600 in pure speed mode. In eye candy mode the positive effect of overclocking reduces with the growth of the gaming resolution, so that by 1600x1200 there is hardly any left despite the increase of the memory bus frequency by 100 (200) MHz. It looks like this game prefers high fillrates and the 4 ROP of the NV43 appear a definite limitation here. The results of RADEON X1600 XT are another indirect proof to this point: this guy is in even worse situation, because it features only 4 TMUs for its 12 pixel processors. Therefore, in 1600x1200 it rolls down to the last but one position in front of RADEON X700 PRO.

Unreal Tournament 2004

In low resolution, all our today’s testing participants run neck and neck, while in higher resolutions there I no visible difference in performance of the reference GeForce 6600 GT and the new ASUS Extreme N6600GT Silencer. Only when we enable FSAA 4s and anisotropic filtering, we can see the difference, although it hardly goes beyond 5%. Overclocking again proves a more efficient tool for achieving better performance rates: the overclocked ASUS Extreme N6600GT Silencer shows its real best in eye candy mode staring with the 1280x1024 resolution and thus wins the race.

Metallurgy level is less CPU dependent, so we can draw more definite conclusions about the performance of our mainstream testing participants in pure speed more. Both GeForce 6600 GT modifications prove worthy competitors to RADEON X800 GT despite the 128-bit memory bus of the latter. And the overclocked Silencer conquers the second best position after GeForce 6800. Even though the latter works at lower clock rates, it features 12 pixel and 5 vertex processors, which determined its victory.

The results obtained in eye candy mode are almost the same as we have just seen in Torlan benchmark. The only difference exists in higher resolutions: ASUS Extreme N6600GT Silencer doesn’t outperform GeForce 6800 any more, but runs neck and neck with it. Since the graphics complexity of Unreal Tournament 2004 is not that tremendous, all testing participants provide sufficient gaming performance even in high-resolutions and with enabled full-screen anti-aliasing.

Performance in Third-Person 3D Shooters

Prince of Persia: Warrior Within

Prince of Persia: Warrior Within doesn’t support full-screen anti-aliasing, just like the Pariah game, because it uses some specific shader effects (Bloom). Although the game is pretty rich in textures, they are not that huge for the 256MB of onboard graphics memory to provide any significant performance advantages. Besides the textures, the game also uses a lot of not very complex pixel shaders that is why the overclocked ASUS Extreme N6600GT Silencer performs pretty well, although it doesn’t outpace the GeForce 6800. RADEON X1600 XT again suffers from some texturing problems (because of only 4 TMUs onboard) and from simple shader performance issues. Relatively raw drivers as well as inappropriate architectural optimizations are the ones to blame for shader performance issues.

Splinter Cell: Chaos Theory

I can’t say that Splinter Cell: Chaos Theory can boast a lot of large textures, because the performance in this game depends mostly on the ability of the graphics accelerator to process Shader Model 3.0 and normal maps. We may disregard the results of RADEON X800 GT and RADEPN X700 PRO here, because both of them support only Shader Model 2.0 and hence work in lighter conditions than the rest of the solutions supporting Shader Model 3.0. So, in this respect our ASUS Extreme N6600GT Silencer will be competing with GeForce 6800 and RADEON X1600 XT in the first place. The first one features 121 pixel processors, and the second one has its architecture optimized for Shader Model 3.0, so our today’s hero working at its nominal frequencies can hardly hope for success. When we overclocked it, the results turned out much better. In pure speed mode ASUS Extreme N6600GT Silencer takes the lead and later on stays at least as fast as GeForce 6800 even with enabled FSAA.

Performance in Simulators

Colin McRae Rally 2005

Well, 128MB of graphics memory seems to be quite enough for Colin McRae Rally 2005 : ASUS Extreme N6600GT Silencer is not any faster than the reference GeForce 6600 GT even with enabled full-screen anti-aliasing and anisotropic filtering. Overclocking does make a lot of sense, as it provides 10%-18% performance boost compared with what we can achieve at the nominal frequencies.

Pacific Fighters

The increase in the amount of onboard graphics memory from 128MB to 256MB doesn’t provide any performance advantages in Pacific Fighters , too. ASUS Extreme N6600GT Silencer and the standard GeForce 6600 GT are equally fast in all resolutions. Both cards are just a tiny bit behind GeForce 6800, but when we overclock our ASUS card it catches up with the rival in pure speed more and outperforms it in eye candy mode. NVIDIA based solutions are undefeated in this game, as usual, thanks to the efficient OpenGL driver.

Lock On: Modern Air Combat

It could be the multi-layer clouds or big earth surfaces that determined the victory of ASUS Extreme N6600GT Silencer over the reference GeForce 6600 GT in Lock On game. The advantages of the 256MB of onboard graphics memory are evident, although only in eye candy mode: in 1600x1200 the performance grows 30% higher. However, this success will hardly be of any practical value to us: 14-15fps performance rate is a way too low for comfortable gaming experience. Overclocking helps improve the situation and the framerate increases to 18-24fps, i.e. almost to the level of the pure speed performance. Unfortunately, this is the maximum we managed to achieve in this game with the highest image quality settings and level of detail.

Performance in Strategies

Perimeter

In 1280x1024 ASUS Extreme N6600GT Silencer outpaces the reference GeForce 6600 GT by a pretty impressive value of 20%, and in 1600x1200 this gap increases to 30%. No wonder, since the gaming engine creating real-time changing vegetation and landscape (which is still a unique feature for today’s gaming industry) works with a lot of data and hence can benefit a lot from the 256MB of graphics memory. Unlike GeForce 6600 GT, ASUS Extreme N6600GT Silencer can compete successfully with RADEON X1600 XT, although only in high resolutions.

In eye candy mode with enabled FSAA and anisotropic filtering the performance advantage may reach up to 50%. However, we can actually see it in 1600x1200 resolution only where the performance drops below the acceptable level. The overclocked ASUS Extreme N6600GT Silencer working at 580MHz chip and 1200MHz memory frequencies (compare with the nominal 500MHz chip and 1000MHz memory) can run from 11% up to 40%-50% faster depending on the work mode.

Warhammer 40.000: Dawn of War

Another synthetic game, Dawn of War , benefits even more from the increased amount of onboard graphics memory. ASUS Extreme N6600GT Silencer outperforms GeForce 6600 GT even in the lowest resolution and gets farther and farther ahead reaching 70% better performance than the rival in 1600x1200 with enabled FSAA 4x and AF 16x.

Performance in Semi-Synthetic Benchmarks

Aquamark3

The amount of available graphics memory doesn’t affect the Aquamark3 results. Despite the extensiveness of the gaming scenes, this benchmark is not that new any more so even 128MB of graphics memory can be quite enough for a graphics card here. ASUS Extreme N6600GT Silencer demonstrates results typical of any GeForce 6600 GT based solution.

Final Fantasy XI Official Benchmark 3

This test works only in 1024x768 resolution, so I wouldn’t expect ASUS Extreme N6600GT Silencer to show any better results than any GeForce 6600 GT would do. The 47 point difference between the two GeForce 6600 GT based cards can be written off as a measuring error.

Performance in Synthetic Benchmarks

Futuremark 3DMark03 build 360

Nothing changes in 3DMark03 , because all the benchmarks included into this suite require 113MB of graphics memory at most. Nevertheless, it makes sense to take a closer look at each gaming benchmark individually, because 3DMark03 total score test uses 1024x768 resolution by default and we know that in higher resolutions as well as with enabled FSAA and anisotropic filtering the tests start demanding more from the memory subsystem. It will also be very interesting to see when overclocked ASUS Extreme N6600GT Silencer shows the best results.

As we have expected, the doubling of the graphics memory amount showed some benefits only in 1600x1200 with enabled FSAA and anisotropic filtering in Game 1 test. In all other test modes ASUS Extreme N6600GT Silencer performed just as a regular GeForce 6600 GT, because the test uses only 23MB of memory: 16MB for textures, 6MB for vertex buffers and 1MB for index buffers. Overclocking improved the performance of our hero by 10%-15%, which is a pretty good result.

In Game 2 test the initial requirements to the graphics memory subsystem are higher: textures eat up 80MB of memory. However, ASUS Extreme N6600GT Silencer performed with even lower efficiency than in Game 1 test. As for overclocking, it not only speeds the card up by the good 20%-30%, but also makes it an indisputable leader.

There are smaller textures in Game 3 test: only 64MB is required. However, the vertex and index buffers increase to 19MB and 2MB respectively. Anyway, this doesn’t affect the final results: GeForce 6600 GT and ASUS Extreme N6600GT Silencer run equally fast here. Just as in Game 2 benchmark, overclocking allows ASUS’ solution to win the leadership in all resolutions and in all test modes. As we can see, Game 2 and Game 3 tests feel quite comfortable with 128MB of graphics memory available even with FSAA and anisotropic filtering enabled in high screen resolutions.

Game 4 test sets more sever requirements for the graphics accelerator: the overall amount of graphics memory allocated for textures, vertex and index buffers reaches 113MB, and theoretically, the test may require over 128MB of memory in high resolutions. This theory comes partially true, because ASUS Extreme N6600GT Silencer starts to outperform GeForce 6600 GT in eye candy mode in all resolutions above 1024x768. the performance gain we observe is about 10% in 1280x1024 and grows up to 20% in 1600x1200. In all other cases the card performs just like any other GeForce 6600 GT. Overclocking makes our hero an absolute leader only in eye candy mode, while in pure speed mode the overclocked ASUS Extreme N6600GT Silencer runs almost as fast as RADEON X1600 XT.

Well, as we have just seen, our suppositions about the performance improvements of the ASUS Extreme N6600GT Silencer have just been proven true: only in two tests 256MB of graphics memory appeared really demanded, and that was only in heavy test modes. In all other cases the ASUS solution performed just as any other GeForce 6600 GT would do.

Futuremark 3DMark05 build 120

The almost complete absence of any performance growth by ASUS Extreme N6600GT Silencer repeats in 3DMark05 : 74 points can hardly be considered a serious performance improvement. Unlike 3DMark03, overclocking doesn’t help our hero to win the race. It looks like the ASUS baby has a very dangerous competitor providing better performance during complex pixel shaders processing: ATI RADEON X1600 XT. Nevertheless, it makes perfect sense to take a closer look at the results of each test individually.

In Game 1 test we still see a bit of a performance gain, even in low resolutions. This gain doesn’t exceed 5%-10%, but it is undeniable. However, despite the action taking place in a space station, the game scenes are pretty broad and rich in textures and light sources. The use of 2048x2048 normal maps is also quite important. Even the overclocked ASUS Extreme N6600GT Silencer cannot catch up with RADEON X1600 XT, however it does have a chance to do it, when we switch to the eye candy mode. In this mode RADEON X1600 XT loses a lot of speed because of the driver issues and only 4 TMUs onboard. For the known reasons we couldn’t compare the results of ASUS Extreme N6600GT Silencer against those of GeForce 6800 and GeForce 6600 GT in this mode.

The scenes in Game 2 test are less overwhelming than those of the first one so the texturing workload here is lower. As a result, ASUS Extreme N6600GT Silencer outperforms GeForce 6600 GT by just a tiny value, if outperforms at all, and this value can be actually regarded as a measuring error. Overclocking doesn’t help our hero to catch up with RADEON X1600 XCT even in eye candy mode, which is quite logical: the geometrical workload is quite heavy and the new ATI solution features only 5 vertex processors working at 590MHz speed. Of course, the driver issues prevent it from revealing its full potential, but it is still fast enough not to let ASUS Extreme N6600GT Silencer take the leadership here.

The situation in Game 3 test is quite curious. As the resolution growsб the efficiency of ASUS Extreme N6600GT Silencer overclocking reduces, so that in 1600x1200 it gets even below 10%. It is quite strange keeping in mind a pretty significant memory frequency increase and high texturing workload. We also see the about 10% advantage of the twice as much onboard graphics memory here for the first time. The situation is almost the same in eye candy mode, although overclocking turns 20% more efficient here. ASUS Extreme N6600GT Silencer competes quite successfully with RADEON X 1600 XT throughout all the tests in eye candy mode with enabled FSAA 4x and AF 16x, just like in Game 1 test.

All in all, the total score result we saw in the beginning is pretty logical. Just like in 3DMark03, ASUS Extreme N6600GT Silencer outperforms the reference GeForce 6600 GT equipped with 128MB of graphics memory only in a few rare cases, such as high resolutions and really heavy performance modes with enabled FSAA and anisotropic filtering. As for the overclocking efficiency, it didn’t help ASUS Extreme N6600GT Silencer to catch up with ATI RADEON X1600 XT in any of the pure speed tests from the 3DMark05 suite. In eye candy mode, however, this task turned out not that impossible to complete.

Conclusion

The Extreme N6600 GT Silencer graphics card from ASUSTeK Computer should definitely attract attention of enthusiasts of silent PCs as well as those who want to have the best performing NVIDIA GeForce 6600 GT-based graphics card. Not only the graphics card provides incredible overclocking potential, its 256MB of onboard memory definitely offer noticable performance advantages in such games as Battlefield 2, Painkiller, Perimeter as well as Warhammer 40 000.

While the Extreme N6600 GT Silencer is totally quiet itself, it is strongly recommended that its additional radiator was cooled-down by either CPU fan or any other fan. Otherwise, the graphics card will be too hot.

A yet another advantage of the Extreme N6600 GT Silencer from comparable graphics cards from other makers is definitely product bundle that is extremely good for a product of this class: it is definitely nice to receive a truly unique graphics card with passive cooling system along with massive amount of games, which are not the newest and are not bestsellers, but which are exciting.

To sum everything up, we can recommend ASUS Extreme N6600 GT Silencer to enthusiasts of 3D gaming who do not want to spend huge sums of money on graphics cards, those who assemble noiseless computer systems (make sure that the big radiator located on the top of the board is compatible with your setup) as well as for users looking for a top-quality feature-rich product. Unfortunately, there are some compatibility issues with this product’s cooling system and some CPU coolers, which is something that potential buyers should take into account.

Highs:

Lows: