Toys for testing are fun...

[Petra]

Well, I've finally gotten the most recent iteration of my CPU waterblock test system together... fun stuff. Photos, a few details, and a somewhat incoherent video tour are below.










Sleep deprived video tour: http://www.youtube.com/watch?v=8OwoygkGKNs

I plan on using the system to run through and test most of the blocks in that drawer, with my goal being to put together a reasonably comprehensive library of CPU block performance data (ranging from old blocks like the WhiteWater to the latest stuff on the market) all gleaned from the same test system and, therefore, directly comparable (also somewhat more relevant, due to the presence of a soldered IHS). I'm not going to go too far into the details right now since I'm pretty tired and will likely botch any sort of explanation of what I've been up to for the past five years... lol

With that said, here's what's on that table covered in stuff:

- Omega FL-1504A-B flowmeter
- Sper Scientific 840081 manometer
- Extech EA15 data logging thermometer
- Volt-modded Asus P5K Deluxe
- Instrumented Intel E6400 (details on the process can be found in an early test run that I did with some heatsinks back in 2008 here...I've learned even more since then and have some new tricks up my sleeve)
- Iwaki RD-30 pump
- HW Labs BlackIce GTX480 radiator
- (8) Delta EFB1212HHF fans
...and some other junk you probably don't care about.

Anyway, this is still going to be the first time I've released this sort of waterblock data to the public... I'll be whipping up my usual battery of pressure drop curves, deltaT (and, perhaps, thermal resistance) curves, as well as maybe gauging overall performance as a function of thermal resistance vs. pressure drop...or something like that.

Time for sleep....

[Conumdrum]

So very cool. The table must go from Cali to NMexico. Looking forward to your input. Should be enlightening and fun. Make sure a PT pen is in each pic for that warm fuzzy feeling.

Miss you guys................

[Groveling_Wyrm]

Ok....you have me jealous now, for having the time to do that sort of testing, and for being able to play with those toys. Nice stuff!

[astrodanco]

It's too clean, much too clean.

[defect9]

I don't see nearly enough discarded pizza boxes lying around propping up furniture. You must be doing it wrong

[Petra]

Ok....you have me jealous now, for having the time to do that sort of testing, and for being able to play with those toys. Nice stuff!

I have more time for it now than ever before, but I'm pretty sure you wouldn't be jealous of the circumstances.

Anyway, this setup sure beats the hell outta the first two or three generations of my test system

[nikhsub1]

It's about time Can't wait to see some data!

[shazza]

Looking forward to seeing your results. Thanks for sharing.

[DarthBeavis]

I have more time for it now than ever before, but I'm pretty sure you wouldn't be jealous of the circumstances.

Anyway, this setup sure beats the hell outta the first two or three generations of my test system

whatever the circumstances, I hope all is well and that you excel in your endeavors as you really deserve it (based on all that you have given to our community).

[Boulard83]

Great stuff !

[Petra]

It's about time Can't wait to see some data!

Well, first I have to take an educated guess as to which of these is going to be the worst performer and use that block to determine what my test heat load is going to be:



From left to right:

- Koolance CPU-330
- AquaXtreme WhiteWater LE
- DD TDX
- DD RBX
- Silverprop Cyclone Evolution SE
- Original Swiftech Apogee
- Swiftech MCW6002
- AquaXtreme MP-1 CPU
- Swiftech Storm G4 (Rev.1)
- Swiftech Storm G4 (Rev.2)
- AquaXtreme MP-05 SP LE
- Swiftech Apogee GT
- XSPC Delta v3
- D-Tek FuZion v1 (stock)
- Enzotech Sapphire (Final--not the early, thick-based one that they sent me for evaluation prior to launch)
- Swiftech Apogee GTX
- D-Tek FuZion v1 (w/mods... nozzles, washer, etc.)
- D-Tek FuZion v2
- EK Supreme
- Swiftech Apogee GTZ (not pictured)

I have a couple other blocks that I'm not going to bother testing and I'm still on the fence about the Koolance CPU-330. The trick here is going to be balancing my desired level of comprehensiveness with usefulness... because the worst block that I choose to test is likely going to be my limiting factor as far as heat is concerned. Obviously, the higher the heat load I test at, the larger the data spread is going to be, and the better the overall accuracy (since it'll help push the differences between blocks out of the margin of error).

I could break things into two heat load categories so that I can gather the best data possible for the more modern blocks... but I'd prefer to keep things as directly comparable as possible (then again, if I do a final comparison chart using estimated thermal resistance, it may be okay... hmm).

At some point, I'm also going to have to beg for/borrow the latest blocks for testing (Heatkillers, new Koolance stuff, Apogee XT, etc.)...

whatever the circumstances, I hope all is well and that you excel in your endeavors as you really deserve it (based on all that you have given to our community).

Thanks, I really do appreciate that

[Bail_w]

Seems like some one is having a lot of fun...