Air / Water Cooing Performance Tests
Air / Water Cooling Performance Tests
Test Bed Hardware
System Hardware:
Gigabyte GA-P35-DS3R rev.2.0
Intel Core 2 Quad Q6600 @ 3.6GHz 1.5Vcore (in BIOS) 9x400
2x1GB Corsair Dominator PC2-10000
eVGA 8800 GTX Superclocked
OCZ GameXStream 600W power supply
Samsung 80GB SATA2 HDD
Lite-On IDE DVD-ROM
Software Used: Rivatuner GPU temperature monitoring
Water Cooling Hardware:
Swiftech MCP655 variable pump (set to 5 for all performance tests and 1 for bleeding)
Clearflex 1/2" ID 3/4" OD tubing
Thermochill PA120.2 radiator with 2x Yate Loon D12SL-12 @ 12V
Picnic-sized cooler reservoir
Arctic Silver Ceramique thermal compound used on STEALTH (included in the box)
OMEGA HH506RA dual input thermometer using Type T probes in the reservoir and in front of the radiator intake
Testing Methodology
Each cooler was measured at idle (windows desktop), then the video card was loaded using 3Dmark 2006 SM3.0 tests looped for 10 minutes running at 1280x1024 8xAA 16xAF. Idle temperatures did not fluctuate for any of the cards, and for load temps, the highest recorded temperature in Rivatuner was taken as a peak load temperature.
I'll be reporting on temperatures while using the reference cooler, the Hydra 8800 with only air, the Hydra 8800 with air and water, and the Swiftech STEALTH using water cooling. The Swiftech STEALTH was equipped with D-TEK 1/2" high flow barbs, and the Cooler Master used the included 1/2" compression fittings (but without the ends on them. Zip ties were used to secure the tubing due to the design of the fittings).
**Operating the Hydra 8800 with only water cooling is not included because it is not a recommended operating method for the cooler according to Cooler Master. The water contact area is only on one of the heatpipes in the fins, and thus would offer very little in terms of cooling to the IO chip (too far away and not directly connected to the heatpipe) or the MOSFETS on the card.**
Ambient temperatures at the radiator intake were measured to vary between 20.5C and 21.7C. Water temperatures were between 22.6C and 24.0C as measured by a Type-T thermistor throughout testing. Please account for 1-3 degree margin of error when interpreting the results.
Air Cooling Performance
First let's have a look at how well the Cooler Master Hydra 8800 performs with air cooling. I honestly didn't have very high hopes for this aspect of the performance testing, given that the Hydra is basically a stock cooler with some of the fins removed and a different (less efficient looking) fan.
Not bad actually. It is hard to say how much of that can be attributed to my excellent job of keeping the thermal compound thin (not sure if you've ever seen the job the factory does on the reference cooler....), and how much can be attributed to the fan the Hydra uses, but it manages the pull the core down a couple degrees lower under load, even with fewer fins.
Water Cooling Performance
Next we have the make-or-break part of a water block review: How well it performs when you're actually running water through it. Just as a reminder, the Hydra was tested with the fan running in addition to the water. This is the configuration we would recommend using for best performance with the Hydra.
This reviewer almost crapped his pants when he saw the numbers (that's me). I was very surprised to see the Hydra walk away from the reference cooler with just a little bit of water contact between a copper tube and one of the reference cooler's heatpipes. This thing really works. The STEALTH, a more conventional water block outperforms the Hydra by a few degrees, but it's also quite a bit more expensive and doesn't have the flexibility to be used without water. This result is very good.