The Arctic Cooling S1 will be temporarily replacing my watercooling block on my graphics card, so it has a lot to live up to as I’m used to such low temperatures.
Once the card was removed from the case and the thermal paste cleaned off the core with Akasa TIM remover (If you ever have to do a lot of changing of heatsinks then this comes in very handy), the ramsink’s were attached using the pre-applied thermal tape. I was expecting that the most difficult part of this installation was going to be the removal of the backing on the thermal tape, but fortunately I was wrong…
The mounting clips for the heatsink require them to clip onto the edge of the PCB at the top of the card, so I had to remove the rather swish bar from the top of my graphics card, fair enough I thought to not many cards have this bar, so after that was done I tried installing the heatsink, only to find that the part of the PCI bracket with the barcode on was stopping the heatsink from sitting low enough on the card to install. The more observant of you may have already noticed how I got around this problem. ;-)
Determined to review this cooler, I decided to take off the PCI bracket for the purpose of this test. I think that this extra metal is something that only appears on XFX cards, so if you don’t have this tab then you will not have any problems. But it was disappointing to find that it did not install as easy as I had hoped for. (Maybe ArcticCooling could supply an actual replacement plate for this in box?)
The next problem came in the form of the supports for the heatsink that clip onto the PCB; they were fine clipping onto the PCB and slotting between the fins on the heatsink, but attaching the second half of the retainer clip on the top part of the cooler required to push the bottom half up with something like a screwdriver (as fingers wouldn’t fit into the narrow space) until they clicked together.
Another thing to note was that there is very little clearance between the cooler and the PCIe power connector, so to get that out I had to use a screwdriver to press down the release latch while pulling the plug out.
The final issue is that with the Turbo module attached the cooling and graphics card take up 3 slots(!), this meant that I had to remove my PCIe Wifi card (even though it is a half height card). The setup had now blocked access to both of my PCIe x1 slots. So this is a consideration for people considering using both the S1 and turbo module.
Testing Some interesting things to note were that it is recommended in passive mode that all your case fans are set to blow out. This is to create negative pressure inside your case so air is drawn in through the open graphic cooler slot at the back of your case. In my case I checked to see if there was enough air being drawn through the back of my case.
As you can see, the negative pressure was enough to suspend this receipt to the case, so there should be plenty of airflow for the cooler.
The S1 should not be used in a case with no airflow, or a case with an open side, as there will be nothing to dissipate the heat away, and as a result the temperatures of the card will just keep rising and rising until the card has to thermal throttle or worse the card is damaged (although these rules might not apply with the Turbo module attached).
Intel E6600 @ 272FSB.
XFX 7950GT XT
Realtek PCIe x1
Swiftech Watercooling loop: - Storm v2 CPU Block - MCW60 GPU Block - MCR220-QP Dual Rad - MCP655 Pump - Tygon 1/2" ID tubing - T-line
Crucial Ballistix PC2-6400
Non removable disks
2x160GB WD caviar
Antec NeoHE 500W
On review: Arctic Cooling Accelero S1 (and Turbo Module)
To test the cooling capabilities of the S1 I will run AT Tool’s 3D mode for 30 minutes to record load temperatures and then leave the system idle at the desktop for 30 minutes to record idle temperatures.
Well this cooler really puts out some impressive results. With the turbo module it is able to beat my watercooling at idle temps!
Things worth considering in this review are the fact that I have four 120mm case fans at the top of my case blowing air out and only one 120mm intake fan (all set at 7v during this test). This creates a lot of negative pressure inside the case which means that a considerable amount of air is drawn in through the back, creating an ideal air flow across the S1. In other peoples cases with only one or two exhaust fans there will not be as much air flow through the S1 and so will probably not give as spectacular results.
It is also worth pointing out that the watercooling loop also contained the processor before the graphics card which would have added heat to the loop and raised the GPU’s temperatures. Lastly, the fans on the radiator were set to 7v which decreased the watercooling performance, but made less noise.
Wow, I really am trying to defend and justify my expensive watercooling setup aren’t I? (Editors note: ermm it sounds like it)
Next Page - Conclusion