by Sergey Lepilov
07/25/2007 | 04:38 PM
We are very happy to introduce to you a new player in the cooling solutions market – the XIGMATEK Company. Founded in 2005, this company is still quite young, but already has a number of impressive products to offer. Among them are such solutions as AIO-S80DP compact CPU liquid-cooling system; a processor cooler called HDT-D1264, very similar to Thermaltake Big Typhoon but with the heatpipe direct touch technology; AIO PI-UN7 graphics card cooling solution.
However, although today we are not going to focus on any of the above listed products, you won’t be disappointed. The solution we will introduce to you today, the new HDT-S963 cooler (CAC-S9HH3-U01) with H.D.T. technology (Heatpipe Direct Touch), is an extremely interesting product as well, and quite efficient, as you will see from this article. But let’s start from the very beginning.
Please, welcome XIGMATEK HDT-S963!
The cooler arrived in a small white carton box without any design or inscriptions. It was evidently not the retail package, although it doesn’t really matter that much:
There is a transparent plastic window in the middle of the box front. You can see the cooler fan through it. At the bottom of the box there were two small plastic bags with all the accessories nicely sorted out. Namely there were the following things:
And here comes the XIGMATEK HDT-S963 cooler itself:
The overall design shape and looks may seem pretty common for our regular readers, we have already come across tower design like that before.
The cooler features a heatsink of 42 aluminum fins, each about 0.3mm thick. The fin array sits on three copper heatpipes with 6mm diameter:
There is a small aluminum pad at the bottom of the cooler that hold all three heatpipes together and serves as a retention base. The heatsink is very lightweight, as the cooler dimensions are only 92x50x134mm and its total weight is only 410g.
The cooler width may suggest that it is cooled with a 92mm fan, and it would be absolutely true and correct:
The heatsink design doesn’t allow installing a second fan for air outtake, because the fan retention groove is available only on one side of the heatsink:
By the way, the fan is fastened with four rubber pins that are inserted in-between the heatsink fins:
This type of retention allows not only to eliminate all vibrations from the rotating fan, but also to reduce the level of generated noise. It is also very easy to replace the fan with a new one in case it breaks down.
The fan used with XIGMATEK HDT-S963 has no sticker on the inside, so I can’t really tell you what company made it:
According to its technical specifications, the fan rotation speed is PWM controlled and varies between 1200RPM and 2800RPM with the noise level of 23-35dBA. The fan is connected with a four-pin connector.
There is a little over 4cm between the cooler base and the lower heatsink fin that is why I cannot think of an electronic component in the area around the processor socket that may prevent you from successfully installing XIGMATEK HDT-S963 cooler.
Each heatsink plate has four slits and two stamped bumps on the side opposite to the fan:
XIGMATEK engineers applied this original approach to ensure proper cooling of the mainboard MOSFET (or MOSFET heatsinks). These slits and bumps allow installing the spoiler that comes with the cooler at any height within the fin array:
XIGMATEK engineers believe it will be the primary tool for directing the airflow in the desired way.
The cooler base is covered with protective film with a warning message that it needs to be removed before cooler installation:
The principally new thing that we haven’t yet revealed in any of our previous articles is the H.D.T. – heatpipe direct touch technology. One glance at the photo of the cooler base is enough to understand what it actually means.
The heatpipes are not just passing through the cooler base: they are an indisputable part of it and are in direct contact with the processor heat-spreader cap. I can totally understand the engineers’ desire to get rid of any additional material between the CPU and the heatpipes, because the shorter is the distance between the processor core and the heatpipes, the lower is the heat flow resistance and hence the higher is the heat transfer rate.
In my opinion, it is not a very simple task to fulfill. The heatpipes should be very tightly mounted into the cooler base and should be bent properly to level out ideally with the base. My only question is: how thick are the heatpipes in the base segment?
Although there is no mirror-shine finish on the heatpipes and base segments, the surface quality is quite fine. I believe that by polishing off the surface you can theoretically win a few additional degrees.
Installation of XIGMATEK HDT-S963 is as simple as the cooler itself. To mount the cooler on Socket 754/939/940 and Socket AM2 mainboards you should use a provided retention bracket with the clip lock that catches to the hooks on the standard plastic socket bracket:
If you have an LGA775 mainboard, you have to screw two retention brackets to the cooler base. These brackets have typical plastic pin-clips already installed:
You won’t need to remove the mainboard from the system case in any situation. The cooler cannot be installed onto Socket 478 or Socket A (462) platforms (note that more and more manufacturers give up the support of these pretty outdated socket types, which is quite logical, actually).
XIGMATEK HDT-S963 cooler looks like this inside of the system case:
The cooler will not hinder installation of any other components thanks to its very small dimensions.
In conclusion I would like to list the technical specs of the XIGMATEK HDT-S963 for your reference:
This solution is currently selling for around $30.
The coolers were tested on an open testbed as well as in a system case with the following configuration:
Note that compared with the previous test sessions performed in our lab, the system configuration is very different this time and it certainly affects the results. Namely, we have intake and outtake fans with lower rotation speeds and hence lower level of generated noise and the side panel fan is connected via a resistor lowering its rotation speed. Moreover, we modified the PSU cooling system that used to be pretty loud. As a result, the background noise of our system case without the processor cooler fan has now dropped to ~34.1dBA (it used to be ~36dBA).
The CPU was overclocked to its maximum with the weakest of the testing participants. As a result, the reference point was set at 3204MHz with 1.45V Vcore:
All tests were performed in Windows XP Professional Edition Service Pack 2. SpeedFan 4.32 was used to monitor the CPU temperature, reading it from the CPU sensor. The CPU was heated up with OverClock Checking Tool version 1.1.0 in a 25-minute test when the system remained idle during the first and last 4 minutes of the test.
The mainboard’s automatic fan speed management was disabled for the time of the tests in the mainboard BIOS. The thermal throttling of the Intel Core 2 Duo processor was controlled with RightMark CPU Clock Utility version 2.25. We determined the thermal throttling rate of our processor experimentally: it began to skip clock cycles at 82°C and higher temperatures.
I performed at least two cycles of tests in each mode and waited for 25-30 minutes for the temperature to stabilize during the tests in closed system case. During the tests on an open testbed we took half the time for temperature stabilization. The maximum temperature of one of the four processor cores 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 at least once again). Despite the stabilization period, the result of the second test cycle was usually 0.5-1°C higher.
The ambient temperature during the test session was controlled with an electric thermometer that could monitor temperature changes over the period of up to 6 hours. The room temperature remained at 26°C during the tests (as marked on the charts with a vertical red line). I would like to add that the cooler fan rotation speeds on the diagrams are provided according to the monitoring readings and not by the fan specifications.
Taking into account he price segment of the XIGMATEK HDT-S963 cooler, we selected a well-known Cooler Master Hyper TX solution as its primary opponent for this test session. Hyper TX was taken as is, without any modifications or additions to it. This is an efficient and pretty inexpensive cooler from the mainstream price range (the recommended price sits around $22). Besides, we have also included the results of one of the super-coolers we tested before – Scythe Infinity – that will serve a reference point for our analysis.
So, let’s check out the obtained results.
First of all let’s take a look at the thermal performance of our cooling solutions with an overclocked quad-core CPU:
Well, XIGMATEK HDT-S963 killed all its competitors in a closed system case, including the Scythe Infinity. Of course, in an open testbed Scythe Infinity took back the leading position, especially since it can easily accommodate a second 120-mm fan, while XIGMATEK HDT-S963 doesn’t allow it. Although at the maximum fan rotation speed, XIGMATEK HDT-S963 didn’t fall too far behind its today’s competitors.
However, now we should better check out the noise level diagram (the measurements were taken according to our methodology described in the previous articles; the subjectively comfortable noise level of 36dBA is marked with a dotted line):
As you can see, XIGMATEK HDT-S963 is too noisy with the fan working at its maximum rotation speed, which is also proven subjectively. The lowest rotation speed of 1650rpm was also selected for a reason (we performed the management with SpeedFan tool): if you raise its rotation speed above this level, the cooler will immediately stand out against the background of the considerably quiet system case. Moreover, it is a pretty dramatic change in the acoustic performance, that indicates a not very high quality fan motor, because it is the fan motor and not the airflow or the bearing that generates most of the noise.
In addition to the thermal measurements described above, we would like to say that the heatsink of XIGMATEK HDT-S963 cooler heats up almost immediately when the processor workload increases. Heatpipe direct touch technology works exactly as it is supposed to, however, the heatsink heat dissipating surface area is not big enough for the overclocked quad-core processor that is why the results are pretty low in the quiet mode and improve significantly (by 8?C in an open testbed) once the fan rotation speed hits its maximum. So, looks like they will need to increase the heatsink surface area to improve the cooler efficiency.
Now let’s check out the maximum processor frequency and its peak temperature in burn mode for each of the testing participants. All the tests were performed in an open testbed:
For a more adequate analysis of our testing participants’ cooling abilities we also provide the noise measurements (taken from a 1m distance). We would also like to add some processor voltage data at maximum frequency, which varied for different coolers:
The results demonstrated by our today’s main hero, XIGMATEK HDT-S963, are quite logical: high processor temperature and the lower overclocking result in quiet mode and at the same time excellent result at maximum fan rotation speed. Cooler Master Hyper TX looks even better: although it loses to XIGMATEK HDT-S963 under peak workload, it gets the CPU to run stably at 3304MHz with lower noise level (37.5dBA against 39.4dBA by XIGMATEK HDT-S963). As for Scythe Infinity, it is beyond all competition this time.
I have to admit that XIGMATEK managed to design a great cooling solution: efficient and quiet, and VERY efficient (but quite noisy) when necessary. All in all, it will be a good choice for quiet system lovers as well as for benchers with limited budget. The most important thing for this company, in my opinion, is to continue working on improving their solution by adding the fourth heatpipe, for example. At this time I have every reason to state that Cooler Master Hyper TX has finally got a very worthy competitor.
We will continue telling you about XIGMATEK products on our site, because as I have already mentioned in the beginning of this article there are some pretty interesting products in their range. So, stay tuned for more!