by Sergey Lepilov
07/31/2007 | 05:43 PM
Heatpipes have become an irreplaceable part of overclockers’ lives. Being something exotic in the earlier days, now they are an integral part not only of every CPU cooling system, but are also actively used in cooling solutions for graphics cards, chipsets, mainboard power elements and even hard disk drives. Most CPU and GPU coolers are currently designed with heatpipes in them.
Today we are going to introduce to you two new graphics card coolers that are also no exception from this rule: their heatsinks are built on heatpipes, too. However, their design, application field, efficiency and pricing are very different from one another. Nevertheless, we decided to check out both of them in together in a single article.
Allow me to introduce: Zalman VNF100 and Gigabyte V-Power!
The Korean Zalman Company doesn’t need a special introduction, I assume. The cooling systems from this well-known manufacturer have been in the market for a long time already and have won the computer enthusiasts’ hearts. These days each new product from Zalman is being expected as something supernatural to say the least of it. It is sad however, that Zalman hasn’t really launched anything really impressive from the overclockers’ standpoint lately dealing mostly with the modifications of the already existing cooling solutions.
But today we are not going to discuss the efficiency of these modifications, but would like to introduce to you a true Zalman newcomer – VNF100 VGA cooling solution.
Flat carton box Zalman VNF100 comes in is pretty small and is designed in light color gamma. The front side of the box has the cooler model name written on it alongside with the 0dBA of generated noise and the list of graphics accelerators this solution is incompatible with:
On the back side of the back you can find a list of Zalman VNF100 key features and technical specifications.
Inside the box there is a transparent plastic casing that holds the main cooler component – the Zalman VNF100 heatsink. All accessories are packed in a separate box:
Among the accessories coming with the Zalman VNF100 there are the following things:
Now let’s take a look at the Zalman VNF100 cooler.
First of all I would like to point out that this cooling solution is very lightweight. It weighs only 180g with 166x95x38.5mm in size. They managed to make it so light due to the heatsink design using only three copper heatpipes and aluminum ribs only 0.4mm thin. Besides, Zalman VNF100 simply doesn’t have a base at all:
It actually forms when you put the system together. Zalman VNF100 uses patented VFP (Variable Fin Profile) Fin Design technology. It implies that the ribs in the array are of variable height, and the manufacturer claim that it allowed increasing the heat dissipating surface area and overall cooling efficiency without increasing the cooler weight:
Zalman VNF100 is a passive cooling system, no fan is intended to go on top of the heatsink.
The base is a very finely polished pretty thin aluminum plate:
There are three grooves on the reverse side of this plate, where the three heatpipes should go. And that is about all regarding the design of the new graphics card cooler from Zalman.
As for the actual cooler assembly and installation on the graphics card, I have to say right away that although it is very simple, it will still take you quite some time to complete. You can find the detailed description of the installation procedure on the company’s official web-site. Here I would only like to show you the schematics of the retention types and the list of supported graphics cards for Zalman VNF100 cooler:
As you can see, this list is quite long and it even includes such “hot” graphics cards as GeForce 7900/7950. But don’t get too excited about it just yet. Check out the test results first, before making any conclusions.
To cut the long story short, if you wish to install Zalman VNF100 onto a graphics card, you need to insert the base into the retention type that fits your graphics card (Socket A or Socket B). After that you need to crew it to the PCB with the spring screws, so that it gets pressed against the GPU heat-spreader:
Then you need to apply thermal grease to the base grooves and set Zalman VNF100 on top locking it tight in this position with the decorative heatsink included with the cooler. This is what Zalman VNF100 looks like on GeForce 7900 GS:
I would like to point out a few things here. First, as you can see, two memory chips on the front of the PCB will remain without heatsinks, because the heatpipes won’t let you put them on. Although on other graphics cards this may not be the problem. Secondly, since the heatsink is actually positioned behind the graphics card, then you need to make sure that there are no components in the area around the PCI Express graphics card slot on the mainboard that would rise above 40mm. In our case, we had to remove the tall Thermaltake Extreme Spirit II cooler from the mainboard chipset, but again, this is an individual case. I don’t think that many of you will have tall heatsinks/coolers in that area, but nevertheless, make sure you check it out in advance.
The recommended retail price of Zalman VNF100 is around $29-$30, which seems to be too much for a passive heatsink, in my opinion. For example, you can buy Arctic Cooling Accelero S1 for the same $29:
I don’t think any comments are necessary here.
Gigabyte Company doesn’t specialize in PC component cooling. However, they are offering a number of solutions for CPU and graphics card cooling at this time. Today we are going to talk about a new VGA cooler - Gigabyte V-Power.
The cooler is shipped in a small flat box with a transparent window in the front and a smaller window in the back of the box:
The box contains extensive details about the cooler features in 11 languages and the list of its specifications:
The accessories bundle is quite large, take a look:
It includes the following components:
And now it is time to meet Gigabyte V-Power in the flesh.
Looks very beautiful, doesn’t it? The cooler is built on a copper base with four copper heatpipes coming out of it in pairs in two opposite directions to the aluminum heatsink halves. In the center of the heatsink there is a 90mm fan, according to the specification (the fan blades from a 92mm fan, actually). This fan features read LED lighting.
The system is covered with a meshed plastic casing:
Gigabyte V-Power is 186x105x34mm in size. It is designed to cool down AMD (ATI) Radeon X1800, X1900, X1950 series and Nvidia GeForce 6800, 7800, 7900 (7950) series graphics cards including CrossFire and SLI configurations. Theoretically, Gigabyte V-Power can fit onto any graphics card with four mounting holes and diagonal distance of 75mm between them. For example, I could easily install this cooler onto GeForce 8600 GTS based on the reference design that wasn’t even mentioned in the compatibility list.
If we remove the plastic casing and take down the fan, the system heatsink will look as follows:
The gaps in the side and top parts of the heatsink let the airflow cool the PCB and the memory chips of the graphics card freely. Note that aluminum heatsinks sitting on each pair of copper heatpipes are pretty small, although there is enough room to actually make them bigger. The weight of this cooling system is not mentioned in the specs, but subjectively it weighs around 300g.
The fan on Gigabyte V-Power is made by Everflow:
The rotation speed of this fan may be adjusted in the interval from 0 to 2000rpm with the help of the controller that comes with the cooler. The maximum level of generated noise is claimed to be 25dBA. Thanks to the frictionless bearing, Gigabyte can claim the MTBF of 40,000 hours (which is over 4.5 years of continuous operation).
The copper base is covered with protective paper sticker warning you that it needs to be removed before installation:
The base finish quality leaves much to be desired, as you can see:
However, its evenness is beyond any criticism.
The installation procedure of Gigabyte V-Power cooler is extremely simple (you can download the detailed installation instructions from the company web-site, a 1.94MB file in PDF format). First you have to stick aluminum heatsinks to the graphics card memory chips (making sure that no heatsink will be in the way of the cooler heatpipes once installed). Then you insert the pins in the retention holes and tighten the spring screws at the reverse side of the PCB. This is what Gigabyte V-Power looks like on Radeon HD 2900 XT:
Unfortunately, you will not see the performance results for Gigabyte V-Power on a hot graphics card like that, because the cooler base was practically “hanging” on the metal frame around the GPU without even touching its heat-spreader.
However, it fit perfectly onto GeForce 7900 GS:
The same was true for Radeon X1950 GT from Palit and for the reference GeForce 8600 GTS.
The only disappointment is the recommended retail price of $50, which seems to be a little too much for a cooling system like that, in my opinion.
Let’s take a closer look at the side by side specifications of the two new solutions and compare them with the Arctic Cooling Accelero S1:
* - checked out by XbitLabs
The tests of all cooling solutions were performed in a closed system case only in identical testing conditions. The testbed was built with the following hardware:
The tests were performed in Windows XP Professional Edition SP2 operating system. We used DirectX 9.0c (released July 2007), ForceWare 93.71 and Catalyst 7.6 graphics card drivers. The graphics cards were warmed up by running the Firefly Forest test from the synthetic 3DMark 2006 benchmarking suite 10 times with disabled full-screen anti-aliasing but with enabled 16x anisotropic filtering.
The graphics cards temperatures were monitored with RivaTuner version 2.02 utility. Each cooler was tested at least two times with a stabilization period of 13-15 minutes between test cycles. If the results obtained in a repeated test were more than 2o C different from the previous ones, the tests were performed again until we got fine precision results. The room temperature during this tests session was stable at 26o C (marked with vertical red line on the diagrams).
The noise level of the tested cooling systems was measured at 3cm, 1m and 3m distance according to our methodology described in the previous articles. The background noise of the system case measured at 1m distance during our test session equaled 34.1dBA.
The average fan rotation speed of all testing participants in different modes is listed according to the monitoring data, and not according to their default specs. Reference graphics card cooling solutions are marked as “Stock cooler” on the diagrams below. We used GeForce 7900 GS 256MB with the standard cooler and overclocked to 576/1720MHz chip and memory frequency respectively. It has already been shown on the photo above. Another graphics card we used was Palit Radeon X1950 GT 512MB with a non-standard default cooling system onboard:
This graphics card’s cooler is designed with an aluminum blade heatsink cooled down by a 92-mm fan covered with a plastic casing:
In Auto mode the fan runs at 40~45% of its maximum speed. The maximum frequencies when Palit Radeon X1950 GT 512MB remained stable were 587/1382MHz for the GPU and memory respectively (500/1200MHz nominal speeds).
Besides the default graphics card cooling solutions we have also included the results of a pretty popular Zalman VF900-Cu LED cooler on copper heatpipes (in two operational modes: at 1800rpm and at maximum 2650rpm). We also included the results for Arctic Cooling Accelero S1 (in passive mode and with one 120-mm fan from Scythe running at its nominal speed of 1200rpm). Unfortunately, we couldn’t test Arctic Cooling Accelero S1 on Palit Radeon X1950 GT, because the cooler legs were hitting against the PCB textolite way before the base touched the GPU heat-spreader, so there was no contact at all. It is a similar situation as the one we have just described above with the Radeon HD 2600 XT and Gigabyte V-Power cooler, although in that case it was the GPU frame that prevented the cooler from fitting properly.
First of all let’s see how our testing participants coped with the graphics processors cooling:
I would like to start with the overclocked GeForce 7900 GS. Although the new Zalman solution coped with the graphics card cooling, the temperature was still too high. Moreover, it was comparable to the results of the noisy default reference cooler as well as the results of Gigabyte V-Power in the passive mode. So, Zalman VNF100 is evidently intended for cooling mainstream and budget graphics cards with not very high heat dissipation. And as for Gigabyte V-Power, its 90-mm fan increases its cooling efficiency a lot: in quiet mode at 1200rpm fan rotation speed it catches up with Zalman VF900-Cu LED and at maximum fan rotation speed it becomes the second best of all the testing participants. So, everything looks pretty good, only the Arctic Cooling Accelero S1 solution left all the rivals very far behind. I wish we could check it out on the hotter Palit Radeon X1950 GT, which is the next topic of our analysis.
The stricken-through temperature values at the maximum Radeon X1950 GT load indicate that Zalman VNF100 and Gigabyte V-Power failed the test and 97o C is the last temperature reading from the log before the graphics card shut down. The preliminary conclusion therefore would be that Zalman VNF100 doesn’t suit for cooling of graphics cards like that and hence will hardly be of any interest to overclocking fans. As for Gigabyte V-Power, the cooler becomes highly efficient with the fan turned on and wins the race at maximum rotation speed on Radeon X1950 GT.
In addition I would like to offer you the temperature readings from the graphics card PCB for Radeon X1950 GT:
The overall picture is the same as in case of GPU temperature test. The difference may be more noticeable on the warmer Radeon X1950 XT or Radeon HD 2600 XT, however, we didn’t have these graphics cards at our disposal at the time of tests.
In conclusion here are the results of our acoustic measurements (the subjective comfortable noise level is marked with a dotted blue line):
The only system that can be called really noisy is the reference GeForce 7900 GS cooler and default Palit Radeon X1950 GT cooler at maximum fan rotation speed. Zalman VF900-Cu LED at 2650rpm, Gigabyte V-Power at maximum 2000rpm and the 120-mm fan of the Accelero S1 (with some allowances) can be regarded as not really importunate, but noticeable against the background system case noise. At medium fan rotation speed these coolers are very quiet and all the others are simply completely noiseless (passive mode).
Our today’s test session demonstrated that both new cooling solutions turned out pretty successful, each in its own way. Zalman VNF100 will cope perfectly with overclocked mainstream and budget graphics cards. As you could see, this cooler installed on an overclocked GeForce 7900 GS proved as efficient as the reference cooling system of this graphics card, only Zalman newcomer was completely noiseless.
Gigabyte V-Power is a much more efficient cooler although it is indeed positioned as a solution for the top price segment. I wish we could check out this cooler on Radeon HD 2600 XT: hopefully the manufacturer will take into account this frustrating issue and make sure that this cooler can be installed on the graphics cards that really need it.
Of course, we cannot disregard the price of the new cooling solutions. At this time we can only base our conclusions on the manufacturer’s recommended price, which will hardly make them competitive in the market. As we see, the same Arctic Cooling Accelero S1 even with an additional 120-mm fan will cost less than Gigabyte V-Power and their efficiency is simply incomparable.
In conclusion I would like to draw your attention to one more important thing. In fact, I believe you may have already noticed yourselves that there are fewer and fewer universal cooling systems coming out lately. For example, although the recently launched Arctic Cooling Accelero S2 and S2 are considered universal solutions, they have different positioning for two different graphics card types. The today’s Zalman VNF100 and especially Gigabyte V-Power also cannot fit on all graphics cards out there. We had to completely exclude from our today’s tests the third cooler, Thermaltake TMG ND2, because it wasn’t compatible with any of the graphics cards participating in our today’s test session. The diversity of old and new generation graphics cards with different PCB layouts do not allow the cooler manufacturers to design a universal air-cooler for them. those available in the market today (such as Zalman VF900-Cu LED) are not powerful enough to cope with the top graphics accelerators and sometimes lose even to the reference cooling solutions. Therefore, overclockers who are about to upgrade their graphics card need to purchase at least two coolers or be prepared to modify the cooling system in order to meet the thermal requirements of their new accelerators, if that is possible at all…