by Sergey Lepilov
05/17/2010 | 06:41 PM
We keep getting an endless flow of new cooling systems for central processors. Manufacturers announce and put into production new solutions on a regular basis. And since there are in fact no significant improvements in cooling efficiency or acoustic performance, then we dare assume that it is the purely marketing aspect that matters most at this point, because it allows the manufacturers to sell almost the same cooler models provided with new names and new packaging. Nevertheless, we sometimes come across really unique and technologically interesting coolers. And our upcoming articles are going to tell you all about them. In the meanwhile, we would like to introduce to you a couple of new coolers from Zalman and Zaward.
Almost a year has already passed since the first model from the new CNPS10X family from Zalman hit the streets: Zalman CNPS10X Extreme. Since then we saw quite a few additions to this family already, namely Quiet and Flex modifications. Now time has come to check out the fourth member of the lineup – the Performa model and see how it differs from the other CNPA10X coolers.
The front of the cooler box bears the cooler photo in a cloud of some blue smoke (which looks pretty pungent, judging by its color). On the back of the box you can find photographs of the cooler’s key features with a brief description:
Also they listed the cooler specifications together with the enclosed accessories. The latter remained almost the same compared with the accessories bundle that comes with the Flex model:
The only difference is an included short cable with a variable resistor and a small pack of Zalman ZM-STG2 thermal paste instead of a syringe.
Zalman CNPS10X Performa is very similar to the Flex version of this heatsink:
The heatsink plates are of a bit different shape, but of the same thickness of 0.5 mm. there are fewer plates in Performa model: 47 instead of 53 in Flex. They measure 130 x 71 mm each. Therefore the effective heatsink size has become smaller: 7,900 cm2 instead of 8,200 cm2. The heatsink plates are spaced out at a larger distance from one another that equals 1.9 mm instead of 1.7 mm in the Flex model:
As a result, the cooler has become 1 mm taller making 152 mm total height. Zalman CNPS10X Performa weighs 748 g.
In our opinion, the key feature distinguishing the Performa cooler from the predecessors is a set of 15 plates in the center of the heatsink:
These plates are of rippled shaped, although Zalman doesn’t really explain the purpose of this innovation claiming that this plate set “makes it sleek and cool aesthetic”. At the same time we know that heatsink fins of variable height not only reduce the airflow resistance but also lower the generated noise level. The only question is why Zalman didn’t make the entire heatsink rippled? This could be saved for the cooler version number five…
Five copper heatpipes 6 mm in diameter pierce the heatsink in staggered order distributing the heat more evenly over the heatsink surface:
The heatpipes lie in special grooves in the copper cooler base. They are soldered to the base, which thinnest part measures 2 mm:
The evenness and finish quality of the base surface remained as impeccable as before:
The changes also occurred in the 120 mm fan. Now Zalman CNPS10X Performa comes equipped with a seven-blade Zalman ZP1225ALM fan model:
This fan is not any different in constructive prospective from the fans used with Extreme or Quiet modifications, however, it works in a different speed interval. To be more exact, this fan can be PWM controlled in two different speed intervals: 900-1350 RPM (17-24 dBA) and 900-2000 RPM (17-36 dBA). For the fan to work in the first mode, you have to include RC24P resistor into the power circuit (it is included with the cooler accessories). The fan measures 113 mm in diameter, and its rotor – 43 mm. It comes with a 300 mm four-pin cable. Due to a double ball bearing the fan should last for at least 50,000 hours.
Zalman CNPS10X Performa can be installed onto any contemporary platform. It comes bundled with the appropriate retention kits, like the ones we saw with Zalman CNPS10X Flex, so we won’t discuss the installation procedure all over again at this time. I would only like to add that the distance from the cooler base to the lowest Performa’s heatsink plate is 37 mm, so it is highly unlikely to interfere with any tall heatsinks on the voltage regulator components. Zalman CNPS10X Performa looks like that inside the system case:
It is remarkable that the recommended retail price of this cooler is set at only $44.90, which is $5 less than the price on Flex that comes even without a fan.
The next cooler we are going to discuss today - Vapor120 - is made by a not very well-known Taiwanese manufacturer called Zaward and is shipped in a small box with a cut-out window in the front:
There are technical specifications on the back of the box. Inside the box you can find the actual cooler sealed inside a plastic blister and the following accessories:
We would like to point out Zaward TCG002 thermal paste with claimed thermal conductivity of over 6 W/m*°K.
The cooler is designed in a very simple manner: three copper heatpipes 8 mm in diameter that form part of the cooler base (H.D.T. technology) hold an array of aluminum heatsink plates, and there is a 120 mm fan attached to one of the heatsink sides:
Vapor120 measures 120 x 77 x 160 mm and weighs 684 g. there are 52 plates 120 x 50 mm that are spaced out at 1.8 mm from one another. Each plate is 0.45 mm thick. The plate edges are bent downwards on both sides of the heatsink covering them:
As you can see, everything is very simple. Although this cooler does in fact have a unique feature. The major peculiarity of this product is the heatsink plates that have stamped dimples in them:
However, these are not just little dimples in the aluminum surface, as there is a little hole in each. This is what the effect from this dimpled surface looks like schematically:
According to Zaward engineers, these dimples ensure more effective heat transfer in those parts of the heatsink that are closest to the heatpipes, which increases the overall cooling efficiency. I wonder if this solution really works they way they say it should and at what speeds.
The cooler base is finished typically for a product employing heatpipe direct touch technology. However, the gaps between the heatpipes in the base are extremely big – 3.5 mm:
Unfortunately, we can’t hope to get even heat transfer from the processor to the heatsink with gaps that huge. As a rule, coolers like that are not too effective.
The fan is also quite unique – with dimples on its blades:
We have already tested a fan like that before in our fan roundup and as we remember it didn’t work wonders. Zaward Vapor120 has a PWM controlled fan working in 300-2000 RPM range. It creates maximum airflow of 93.96 CFM and generates between 10 and 34.3 dBA of noise. The fan measures 113 mm in diameter and its rotor – 45 mm in diameter. It comes with a 240 mm long cable.
The fan is attached to the heatsink with two wire clips:
The photo above also shows one of the four LEDs that add blue glow to the working fan. The dimples on the fan blades add a striped effect that is particularly noticeable in the dark:
Zaward Vapor120 is a universal cooler. Due to a universal backplate and two retention kits it can be installed onto any contemporary platform including LGA1156:
The cooler is installed in the same manner onto AMD and Intel platforms. At first, we install a backplate with retention mounts held with screw-nuts and insulating washers beneath them:
Then you use the screws to attach the corresponding retention plates to the base. They are universal for LGA 775/1156/1366 sockets:
After that you simply install the cooler onto the CPU and tighten large screw-nuts:
The retention ensures very secure hold with high pressure which is important for coolers using heatpipe direct touch technology. The distance between the lowest heatsink plate and the contact surface is 38 mm. Therefore, since the heatsink is pretty narrow (50 mm), the cooler will fit even onto those platforms that use memory modules with tall heat-spreaders.
There is no mention anywhere in the manual about the preferable cooler positioning, so we had to figure it out on our own. We discovered that it doesn’t matter for Zaward Vapor120, which way it is installed onto the processor, because the results were the same in both cases. It must be the 3.5 mm distance between the heatpipes in the base of the cooler that determined equal outcome in case the heatpipes were turned along or across the heat-spreader of the LGA1366 CPU.
The recommended retail price for Zaward Vapor120 is $39.
We are going to test the cooling efficiency of the new coolers and their today’s competitors in the following testbed:
Processor overclocking was limited by the least efficient cooler of our today’s testing participants in its quiet mode. As a result, we managed to overclock our quad-core processor (with polished heat-spreader) with the multiplier set at 21x and “Load-Line Calibration” enabled to 3.84 GHz. The nominal processor Vcore was increased to 1.325 V in the mainboard BIOS:
The memory voltage was at 1.64 V and its frequency was around 1.45 GHz (7-7-7-14_1T timings). All other parameters available in the mainboard BIOS and connected with CPU or memory overclocking remained unchanged.
All tests were performed under Windows 7 Ultimate RTM x64 operating system. We used the following software during our test session:
So, the complete screenshot during the test session looks as follows:
The CPU was loaded with two consecutive Linpack test runs with the settings as indicated above. The stabilization period for the CPU temperature between the two test cycles was about 8-10 minutes. We took the maximum temperature of the hottest CPU core for the results charts. Moreover, we will also provide a table with the temperature readings for all cores including their average values. The ambient temperature was checked next to the system case with an electronic thermometer with 0.1 °C precision that allows monitoring the temperature changes over the past 6 hours. It was pretty warm already, but the AC wasn’t working yet, so the room temperature during our test session varied between 25.9-26.1 °C.
The noise level of each cooler was measured between 1:00 and 3:00 AM in a closed room about 20 m2 big using CENTER-321 electronic noise meter. The noise level for each cooler was tested outside the system case when the only noise sources in the lab were the cooler and its fan. The noise meter was installed on a tripod and was always at a 200 mm distance from the cooler fan rotor. The tested cooling systems were placed at the edge of the desk on a sheet of polyurethane foam. The lowest noise reading our noise meter device can register is 29.8 dBA and the subjectively comfortable noise level in these testing conditions was around 34 dBA (do not mix it up with low noise level). The fan(s) rotation speed was adjusted in the entire supported range using the new controller revision by changing the voltage with 0.5 V increment.
Besides the default fans, Zalman and Zaward coolers were also tested with one and two Thermalright TR-FDB fans at 1100 and 2050 RPM installed onto the heatsink for air intake and exhaust. Moreover, we will also include the results of the ultimate winner and our today’s performance leader – Noctua NH-D14 equipped with two 140 mm Noctua NF-P14 fans:
In this case the super-cooler costs almost twice as much as Zalman CNPS 10X Performa or Zaward Vapor120, so we also added the results for Zalman CNPS10X Flex, Performa’s predecessor, and Alpenföhn Nordwand, which is practically the today’s best cooler with heatpipe direct touch technology.
You can check out the test results in the following table and diagram:
Click to enlarge
First of all, let’s check out the coolers’ performance with the default fans. In this case the least efficient one is Zaward Vapor120. Even at maximum rotation speed of its default fan it cools the CPU 2°C less than Zalman CNPS10X Performa in moderate fan speed mode at 1100 RPM. I have to say that both new coolers proved to depend a lot on the rotation speed of their fans. The cooling efficiency of Zaward Vapor120 increases by 7°C when we raise its fan speed from 1100 to 2050 RPM. At the same time, even at maximum speed of a single default fan and at pretty high level of noise Zalman CNPS10X Performa can at best catch up with Noctua super-cooler in quiet mode.
Now let’s compare two new coolers against their direct competitors all tested with identical Thermalright fans. The first thing you should pay attention to, is the fact that the cooling efficiency of Zaward Vapor120 increases once you simply replace the default dimpled fan with a Thermalright TR-FDB one. At the same speed Vapor120 drops the peak processor temperature by 4°C, which allows us to conclude that its original fan is not effective enough. The interesting thing is that Zalman CNPS10X Performa with an alternative fan in quiet mode at 1100 RPM outperforms itself with the default fan by 1°C, while at maximum rotation speed the results are equal. With a second fan installed for air exhaust improves the cooling efficiency of both Z-coolers by 5°C except the quiet mode on Vapor120.
As for the comparison between the new coolers and their competitors, as we have expected, Vapor120 was defeated by Alpenföhn Nordwand. The advantage of the cooler with a cool Scandinavian name varies between 3 and 7 °C. On the contrary, Zalman CNPS10X Performa, it outperforms the Flex model by 2-4 °C in all modes. Taking into account all results obtained previously on Extreme and Quiet Zalman CNPS10X models, we can conclude that Performa solution is the best cooler of the four. We wish we could also compare the Extreme and Performa versions directly against one another, but unfortunately, we no longer had the former at the time of tests.
After that both today’s heroes were tested during maximum CPU overclocking. Both: Zaward and Zalman coolers, were tested inside a closed system case with two Thermalright TR-FDB fans at their maximum speed. As a result, Vapor120 could have Core i7-920 run stably at 3950 MHz at 1.35 V Vcore and peak temperature of 84°C, while CNPS10X Performa allowed us to push our CPU almost to the very ultimate top: it hit 4030 MHz at 1.375 V core voltage and 78°C temperature.
In moderate acoustic mode when both fans run at 1100 RPM, Zaward Vapor120 allows the processor to remain stable at 3900 MHz frequency and 1.3375 V core voltage when the peak temperature reaches 85°C. as for Zalman CNPS10X Performa coped with a CPU overclocked to 3990 MHz that was working at 1.36875 V core voltage and maximum temperature of the hottest core at 83°C:
I would like to add that in the same conditions Noctua NH-D14 with two 140 mm fans at 900 RPM cools the CPU 5°C better:
The graph below shows the results of our noise tests:
I can’t say that both coolers are equipped with very quiet fans. Nevertheless, they are acoustically comfortable up until 1100 RPM speed. At the same time, the noise from Zaward fan increases more predictably and even as the rotation speed increases, while Zalman fan changes its acoustic very abruptly. For example, in the interval between 1000 and 1100 RPM, the fam generates almost the same noise all the way, while in the interval between 1100 and 1200 RPM the noise jumps up by about 3 dBA! We can find more of intervals like that higher up the graph, too, which indicates that the fan of CNS10X Performa cooler is not balanced enough, and so is the fan bearing.
Although Zaward Vapor120 turned out the least efficient cooler in our today’s test session, even in quiet mode this product can ensure that the overclocked quad-core processor will run stably at 3.84 GHz under peak load. If you decide to buy Vapor120 intending to overclock your processor to its maximum, then it will definitely be a good thing to replace the default fan with an alternative one (or even two) with high static pressure. The cooler is completely universal, has relatively simple and at the same time reliable retention with high-pressure hold, beautiful LED lighting of the PWM controlled fan. At the same time, despite the average price point of $39, we think Zaward Vapor120 should cost even less, because you can currently purchase an even more efficient and quiet Scythe Mugen 2 for practically the same amount of money.
As for Zalman CNPS10X Performa, we totally admire it. In fact, this product has only one drawback: unbalanced fan with “floating” noise level. It is up to you to decide how significant this issue is for your particular case, but it won’t be a problem to pair a 44.90-dollar cooler (which is an unusually low price for a Zalman product) with a couple of quality fans for $5-$7 each. Otherwise, CNPS10X Performa has every right to be called one of the best air coolers with reliable retention, support of all contemporary platforms without any exceptions and exceptionally high cooling efficiency. There is only one thing we can wish to Zalman at this point: to stop perfecting the product that was launched about a year ago and design something really new and super-effective, which will undoubtedly please Zalman fans and stumble Zalman competitors.