by Sergey Lepilov
04/26/2007 | 05:54 PM
We have recently tested a new CPU cooler from Thermaltake aka MaxOrb. Before this cooler came into the world, Thermaltake was quiet for almost three years having released nothing very exciting in this product family. Gigabyte Company seems to have done practically the same thing, when back in 2004 they introduced a not very successful 3D Rocket cooler and then stopped working on this product family for almost three years also (for details on the first 3D Rocket cooler from Gigabyte see our article called High-End CPU Coolers Roundup: The Battle for Silence and Efficiency). In 2005 the company announced a very efficient, but pretty loud G-Power Pro cooling solution, though it is based on a completely different concept from what 3D Rocket series is employing.
And now, in the end of 2006, Gigabyte proudly announced improved Rocket – Gigabyte 3D Rocket II (GH-PCU23-VE) model that has borrowed only the heatsink shape from the predecessor. Everything else has been modified and improved dramatically. So, let’s find out what has been changed and how it affects the efficiency of the new cooling solution today.
Let’s get started now.
The cooler package is made from transparent plastic:
The stickers at the upper side of the front of the box read that this cooler is capable of handling CPUs with up to 160W heat dissipation and that it produces only 16dBA of noise during operation. In addition there is also a sticker with the list of all supported processors and with information on Socket AM2 support. If you look at the other sides of this box you will get acquainted with the detailed technical specifications of the device and its key features.
In the bottom part of the package there is a small section containing a plastic bag with accessories:
You can see that the package contains everything necessary, even multicolor rubber rings (blue, green, white, orange) that will definitely please modding-fans.
Gigabyte 3D Rocket II corresponds very well to its name, looking somewhat like a real rocket:
The cooler design is tower-shaped, but at the same time is highly original and unique. Four copper nickel-plated heatpipes rise from the copper base. They go through numerous aluminum plates and the whole thing is topped with two plastic casings, the lower one hosting a 92-mm fan:
The fan blades face up, i.e. the air is pushed away from the top of the cooler heatsink and is sucked in through its side ribs, and not from the bottom of the heatsink as you may have assumed at first. It happens to be this way because there is another small fan at the very bottom of the aluminum heatsink and it directs the airflow right to the mainboard.
It is interesting but there are no special wide slits for air intake between the heatsink plates. So, both fans suck the air in only through the available gaps between the plates and therefore there is a plastic “skirt” with four guides looking very much like rocket wings. In fact, it is hard to believe that these plastic guides really improve air intake, but let’s believe the manufacturer’s word here.
The tower fan has 7 arched blades and features four blue LEDs:
The variable resistor included with the cooler allows changing the fan rotation speed from 1500rpm to 3000rpm. The noise level in this case will be 16dBA and 33.5dBA respectively.
Unfortunately, the manufacturer didn’t provide any information on the technical specifications of the fan at the bottom of the heatsink:
Of course, this fan serves to cool down the area around processor socket.
The cooler base is covered with protective paper sticker warning you that it needs to be removed before installation:
As we saw when we removed the sticker the finish quality of the base surface is far from ideal:
So, if you are longing to get that nice-looking mirror shine on your cooler base, you will have to do some work on your own. Despite the non-shiny finish of the cooler base surface, it is very even, no complaints here: we checked it by placing a thermal grease print onto a test glass.
Right above the heatpipes there is a small heatsink with deep slits:
They serve to catch the clip when the cooler is installed onto Socket AM2 or Socket 754/939/940.
The heatpipes sit in the grooves in the copper base of the cooler and are soldered to the base:
There is a small PCB with connectors and simple regulator circuitry in the lower part of one of the heatsink sides:
There are two connectors for the cooler fans, one power supply connector for the mainboard and one more connector for the fan rotation speed regulator.
The cooler installation technique is described in great detail in the user’s manual, as I have already mentioned above. For Socket AM2 and Socket 754/939/950 mainboards the entire installation procedure consists of pushing the retention clip into the slits in the cooler base and locking it to the plastic bracket:
As you see, the cooler, namely the heatpipes in the base of its heatsink, to be more exact, are quite compact and do not get outside the bracket. Therefore, this cooler will not interfere with any electronic components on the mainboard or tall heatsinks of the other cooling systems employed on the board.
If the cooler is installed onto an LGA 775 mainboard, you will have to take it out of the system case first. You need to remove it from the case in order to fasten the retention bracket before installation:
Then you can install the cooler and fasten it to the mainboard with other clip-brackets included into the accessories bundle:
To install the cooler onto Socket 478 mainboards you will use the same clips as in the previous case, but they will hold on to the standard preinstalled plastic bracket on the board, so you won’t need to remove the mainboard from the system case. By the way, the retention clips of Gigabyte 3D Rocket II press the cooler very tightly to the CPU heatspreader, so the cooler won’t move or rotate when installed.
In addition I have to point out that Gigabyte’s own thermal grease is very thick and hence is very hard to spread evenly over the heatspreader surface. However, we have compared its conductivity with one of the leaders – Arctic Silver 5, and we can admit that Gigabyte’s thermal grease is absolutely as good. It is definitely another point in favor of 3D Rocket II.
Let’s take a look at the installed Gigabyte 3D Rocket II:
This solution turned out to be one of the two tallest coolers we have tested in our lab so far. With the top plastic ring installed as well, it is as tall as another record-breaker – Scythe Infinity. Its height is 160mm.
In the dark the cooler looks very impressive:
However, some of you may think that this bright light is too annoying, but it is a matter of taste.
Well, now that we have discussed the installation procedure there is one more thing we should dwell on before we proceed to the actual thermal tests.
I would like to pay special attention to the orientation of the top plastic ring and direction of the airflow. Take a closer look at the ring:
It is sealed on one side and has air outlets for warm air on the remaining three sides. The manufacturer specifically stresses this peculiarity in the user’s guide:
Besides, there is also a schematic of the correct airflow direction when the cooler is installed on the mainboard that can be set with this ring:
In this case the warm air should be ousted towards the case rear panel and case fan (if there is one available, of course):
Moreover, if there are ventilation holes in the case side panel, then you can direct airflow towards this panel by removing the Air Duct ring, as you can see from the illustration above. In reality, however, the air is still ousted mostly away from the cooler top. Even with the air duct ring installed, most air still goes away from the cooler top, and not sideways away from the sealed side of the ring. And this also was proven during our practical experiments, which you will get to in just a moment.
I would like to share my subjective estimate of the level of noise the new Gigabyte 3D Rocket II generates during work. At the maximum fan rotation speed, the cooler is actually pretty loud, I would even compare it to a medium-sized vacuum cleaner. At this rotation speed the airflow is so powerful that the sheet of paper placed above the cooler gets lifted vertically or horizontally (if you do the trick with the side panel of the case open). I doubt that the maximum 3000rpm rotation speed will be of any good use at such high noise level.
In “Normal” mode according to the regulator markings (2000rpm) the manufacturer declares 25dBA noise, which may seem as not that much. However, in reality, the cooler is still too loud. Even if you reduce the fan rotation speed down to the minimal 1400rpm (according to the monitoring report), the cooler won’t be a noiseless one, despite the claimed 16dBA level of noise. You can still hear the fans working quite distinctly. However, this is the mode when you can call it quiet, as this work mode would be considered the most optimal for everyday work.
The table below sums up all the technical specifications of the new Gigabyte 3D Rocket II cooler:
During our test session we will compare the thermal performance of Gigabyte 3D Rocket II cooler against that of a well-known winner Thermaltake Big Typhoon in its standard modification without any modding done to it. Why we chose this particular rival? Well, it is widely available in the market these days, it is very inexpensive for its class and it provides excellent cooling efficiency.
Both coolers were tested in a closed system case and in an open testbed with the following configuration:
All tests were performed in Windows XP Professional Edition Service Pack 2. SpeedFan 4.32 was used to monitor the temperature of the CPU, reading it from the CPU Sensor. The AMD CPU is heated by S&M version 1.9.0b for 15 minutes at 100% FPU load. The Intel CPU was heated up with Intel Thermal Analysis Tool for 25 minutes (according to the method described in our article called Originality or Efficiency? Cooler Master Mars, Eclipse and Hyper TX Cooling Solutions Reviewed).
The mainboards’ automatic fan speed management in the BIOS was disabled for the time of the tests. The thermal throttling of the Intel Core 2 Duo processor was controlled with RightMark CPU Clock Utility version 2.25 (our processor began to skip clock cycles on reaching a temperature of 81.5°C).
I performed at least two cycles of tests in each mode (TAT and S&M) and wait for 25-30 minutes for the temperature to stabilize during each test cycle. During the tests on an open testbed we took half the time for temperature stabilization. The maximum temperature in the two test cycles was considered as the final result (if the difference was not bigger than 1°C – otherwise the test was performed once again). Despite the stabilization period, the result of the second cycle was usually 0.5-1°C higher.
The ambient temperature during the test session was monitored with an electric thermometer that could monitor temperature changes over the period of up to 6 hours. The room temperature remained at 23.5°C during the tests (as stated on the charts).
During our overclocking test session we had Gigabyte 3D Rocket II working in quiet mode inside a system case. In this work mode we could overclock our Intel Core 2 Duo E6400 (2133MHz nominal speed) to 3256MHz without increasing the processor Vcore:
Considering that the airflow from 3D Rocket II is being directed primarily upwards, and in this case the case fan blowing the air onto the processor socket may be disturbing, we decided to perform two test sessions in a closed system case. The first test session was performed with the sealed side window for the case fan (assuming that not all system cases have vent holes or fans in this part of the case). And the second test session was performed with the 120mm fan (1200rpm) installed into the side panel of the case, but instead of sucking the air into the case and directing it towards the processor socket, this fan was ousting the air outside the case. These results on the charts are marked as “+fan out”. When we tested Thermaltake Typhoon in the second test mode, the case fan wasn’t turned the other way around.
I would like to add that the fan rotation speeds on the diagrams are stated not according to their technical specifications, but according to the monitoring reports.
So, let’s now discuss the actual results:
As you can see, during the tests in a closed system with the sealed side panel opening new Gigabyte 3D Rocket II didn’t prove that efficient. In quiet mode at 1400rpm the CPU got close to the throttling mode. And that was during relatively modest overclocking without even increasing the processor Vcore! The loud work mode when the fan is working at its maximum rotation speed improves the cooling efficiency of the new 3D Rocket II by 8°C, which proves our above expressed concern that the slits between the heatsink plates may be too narrow for sufficient air intake.
The situation turns out much better if there is a fan installed onto the case side panel that ousts warm air outside the system case. In the same quiet mode Gigabyte 3D Rocket II works as efficiently as it performed at its maximum fan rotation speed in the previous test with sealed side panel. Moreover, when tested in an open testbed at the same 1400rpm the results of our hero are even worse than in the closed case with the side fan installed. Maximum fan rotation speed improves the efficiency by another 4.5°C. Nevertheless, these attempts of the new Gigabyte 3D Rocket II cooler to compete with Thermaltake Big Typhoon do not succeed.
Why do we claim the defeat? Although according to the results we have just discussed Gigabyte 3D Rocket II yielded to Thermaltake Big Typhoon only 3°C in a closed system and 5.7°C in an open testbed, the difference in the level of generated noise is dramatic. Besides, Thermaltake’s solution allows much better CPU overclocking. Therefore, we decided to check out the maximum CPU overclocking result that could be achieved with Gigabyte 3D Rocket II in the most optimal mode for this cooler and for the overclocker who cares about the noise issue. The experiment took place in a closed testbed with a 120mm fan installed into the side panel for air oust and Gigabyte 3D Rocket II was working at 1400rpm. Intel Core 2 Duo E6400 processor proved stable at 3.423MHz with 1.4V Vcore:
And again here come the results:
Although Gigabyte 3D Rocket II coped well with an even higher overclocked CPU, it still lost over 10°C to Thermaltake Big Typhoon.
The results of our test session indicate that Gigabyte didn’t manage to create a winner to beat the current cooling efficiency leader. But this solution will still find its customer, no doubt about that. Stylish looks, universal design, relatively simple installation procedure and low level of generated noise at minimum fan rotation speed, these are indisputable advantages. Fan highlighting alongside with the replaceable fluorescent rings will definitely win the hearts of modding fans. We also shouldn’t forget that this solution offers excellent cooling for the area around the CPU socket and comes with high-quality thermal grease. Looks like there are quite a few advantages about this product, however, for most users they might be secondary.
It is sad that despite a pretty high price of $60 Gigabyte 3D Rocket II loses over 10°C to Thermaltake Big Typhoon selling currently for $40. And that is despite the fact that we created the most favorable working conditions for this newcomer. If you install this cooler into regular system cases with solid side panel, it will be able to cope only with moderately overclocked CPUs. Of course, we can increase the fan rotation speed to the maximum and win additional 8°C, but you will hardly be eager to accept the high level of generated noise.
In conclusion I would like to once again mention all advantages and drawbacks of the new Gigabyte 3D Rocket cooling solution: