[Player0]

Player, I am curious where you get that assessment that doing testing with less than a controlled environment or less than enough test sets that you get better data? You mean data that has numbers that are less exact but closer to what the mfg or the people get with mobo probes is better in some way?. I personally like to know how good something is in an absolute fashion, not just "well it runs about xx Deg"

I never said my data was more accurate or better than someone who invests in a very serious test of the waterblock in a much more controlled manner. You misread my statement. I beleive that in my 'rougher' testing, i can recommend one waterblock over another to someone with a good deal of accuracy. It may not be AS accurate as someone who spent more time and effort testing. But, in my experience, my performance results jive up consistantly to what many other people experience with the same blocks, be it in other reviews, or personal experiences. So if someone had a Maze3, and upgraded to a MCW462, which tested better for me, most of the time they would agree that the Swifty worked better. My contention is that yes, you need rigid testing standards to get accurate numbers. But good testing standards (which I have) is generally good enough to show people which blocks are working better than others. There is nothing in my upcoming review that looks out of the ordinary in terms of dT (dont go quoting my Maze4 numbers here now as being out of the ordinary because no one here has seen my final results yet with water temp taken in to consideration). The only block that surprises me is the SlitEdge, which has been doing poorly for me. I honestly cannot get better temps with it. But I will continue testing it to make sure. It is when things look odd that I spend the extra effort to make sure.

The argument about "average joe" stuff has come and gone and has been used for close to the last 2 years of testing evolution. The simple fact is, the "average joe" is wrong. Data that’s more error ridden, uncontrolled, and un validated is never right.

My data is not error-ridden and uncontrolled. It's simply not fine tuned to the level that some other reviewers go to. Ive never dissed anyone for putting that much time in. I simply justify my not doing that because again, in my experience in testing blocks, my results always resemble what other people get, including the guys who spend much more time reviewing each block.

You should also take in to consideration that this is a round-up. Not a review dedicated to one particular block. In the effort to actually get the review finished, yes, I cant be remounting the block 10 different times. I consider 4-6 times to be enough (3 on the cascade) because I havent seen any bad deviations. I do a very good and consistant job of ensuring a proper mount each time, and I know that its again, not perfect. But, I havent seen any large errors in my data that Im worrying about.

Yes testing takes insane time, insane commitments, and insane costs, and really is the reason I don’t do it much anymore. pH has taken the torch and is producing data that is second to only a few other setups around. But testing takes time, shitloads of time, and as you said, no matter what you do people will bitch.

No matter if you spend $500 or $5000 in testing a block, there will be people who wont beleive it. No matter how good your testing equipment is, nothing stops ANYONE from skewing results. The review is only as good as how much you trust the person. That is why a lot of people dont bother to spend the time, ESPECIALLY manufacturers of the blocks, on doing large reviews. Because no matter how good you do it, people will think you did it wrong, or were swayed by a manufacturer. You guys could spend a year reviewing a block, but does it really do you any good if no one beleives you were honest in doing it? It makes you question the validity of ANY result, be it mine or someone who does it for a living.

Now I have done a good deal of testing, got burned at the stake for some of it... most of it. While there were lessons to learn with every test I did, they all pointed at the same goal - repeatable and controlled results (which when I got deeper and saw how expensive it was turning, I couldn’t continue). Call us demanding, or "ego driven" folks, but the simple fact is, a good few of us have been though the review ringer, and most of this is talk from experience and previous grillings that we got. Your 'posse" over there will back you too no end, I think that’s very clear. Over here, most back good solid testing, and no bullshit reviews, not a just a person for the sake of it.

I have been through this same ringer multiple times. I have gained a lot of experience from those who question my methods. I beleive I have a solid testing foundation and this point, and feel that although my results may not be fine-tuned, that doesnt make them sloppy, or innaccurate, or worthless. I can prove that only by showing in my past revies (published, or on BBSes) that my numbers jive with those who spent much more time reviewing the product. I feel that how I do things is 'good enough' in the sense for people who aren't as expert about this stuff, who dont want to know any numbers other than which block is best for them, and who certainly dont care about design.

Now with that said, how can you tell me which block is the best in the market? just by how hot it runs your CPU on a given day, with a certain coolant, at a certain flow, which are mostly uncontrolled or undocumented errors in the equation.

Because all those factors HAVE been controlled to some level, even on my 'lowly' test rig. My houses temperature remains constant as does humidity. The pump certainly should remain constent, so should flow. I cant picture any factor in my test bed that flucuations too rapidly, and im testnig each block multiple times, and getting REPEATABLE results. I take many other factors in to consideration, other than just CPU temp. As far as Im concerned, if my results can be reproduced, thats really the most important thing.

Living in the world of make believe may be nice for some things, "reviews" of cooling products is not one of them. I think its an insult to the general cooling community to say that they don’t want accurate numbers.

You are the only ones claiming my numbers aren't accurate, and I dont by it. Ive been accused of that for every review, and so havent most people here. Yuo dont know me, you dont trust me, so you wont respect my review. Thats fine, there are others who will. I beleive my numbers are as accurate as they are going to get given my resources, and that my testing methodoly is solid. Theres nothing sloppy about how I did things. It may not be up to some peoples standards, but I dont think thats a difference between right and wrong numbers. I think thats a difference between coarse tuning and fine tuning.