Recent water block (GT/Fuzion/Storm) test results...

[lawnmowerman]

A couple of issues I would like to raise

Measuring relative performance of these blocks on the same rig, whatever its configuration, is entirely valid since you are comparing like with like - however the delta T figures being used ( core temp to ambient ) are meaningless as they are completely dependent on the performance of the whole cooling loop. The only valid comparison would be the delta T between core temp and water temp at a given flow rate but since each tester is using a different configuration that is not possible. The only useful conclusions that can be drawn are from the relative performance of each block in the tests to see what trends emerge which would seem to be that - on Kentsfield at least - Fuzion and Apogee GT (bowed) about same, better than Apogee and noticeably better than Storm and MP-05.

Sorry if this is a bit of a rant but I hate reading conclusions based on meaningless results. Some of the results listed would require a cooling loop that could take a load of 150w+ and yet have the water remain at ambient!!

Regarding the full load figures I should remind people that Prime95, Orthos etc do not heat up a Kentsfield as much as CPU Burn in and way less than TAT. Difference for me between Orthos & TAT is around 5C!

My rig QX6700(lapped)@3.75(266 x14) @ 1.5v(1.45v actual), 680i cooled with modded DDC ultra, PA120.3, 3 x Noctua 1200rpm fans, Apogee GT (bowed), DD880GTX block, DD680i N/B & Swiftech MCW30 on S/B - in parallel. Can run any bench/test 24/7 in closed case. I also have Storm, Apogee, MP-05 and ApogeeGT outperforms them all. I am waiting on a Fuzion so I can continue testing but I am not expecting any great performance difference from ApogeeGT.

Incidentally regarding the load on motherboard - especially with bowed blocks - I too am more than a little concerned and feel that a backplate would be a good idea. Swiftech? D-Tek?