Bill Adams waterblock testing article

[bigben2k]

Click me!

Nice one. Here's hoping that review sites are going to start applying some logic to their testing!

Bill: out of curiosity, where are you going with this? I mean, you could market your services, either for a website, or to the WB manufacturers. You mention "avoiding WB testing like the plague" but you've obviously invested a lot of time and resources into creating this workbench. What's up?

[Skulemate]

Thanks Ben, though pHaestus already posted this in another thread. A good read...

[BillA]

why Ben ?

Joe Kelly made me do it ! (its all his fault !)

after giving him the redas* for BSing w/o facts (or research, ahem)
I thought to test rads (and later wbs) in my garage no less
well, it was much much harder to do (well) than I had thought, MUCH harder
but I know a bit about testing so I persevered (over 2 years now, and thousands of $)

I had naively thought that the mfgrs would have an interest, and they do - in their profits
the link between factual product performance data and sales is unknown
-> because no mfgr uses any real data at all, and they sell their wbs just fine, thank you

one business plan bites the dust

and a website can only pay if its a cash cow, and we know which site that is
I won't play in a sand box that has turds in it, so that's out too

hi ho

[airspirit]

They call that a litter box, Bill.

[myv65]

Originally posted by unregistered
the link between factual product performance data and sales is unknown

. . .and will remain so until the masses start buying purely on performance (real measured performance). That's not liable to happen any time soon in a world where many would rather spend money on window mods and cold cathodes instead of faster/better equipment. Not that's there anything wrong with that. . .

It's a sad world for engineers when folks often buy purely on looks rather than features/performance (cars, computers, etc.). Now, if everyone only had the same tastes as engineers. Hmm, yeah, that'd be a weird world.

Good job Bill.

[BillA]

should I then say:

oohh, what muscular pussies ?

EDIT: posts crossed

re the 'weird world': indeed, I can see it now - SQUARE CORNERS

[Blackeagle]

Thanks for another good article.

[bigben2k]

Here's the follow up article:
http://www.overclockers.com/articles655/

[g.l.amour]

thk u BillA

[murray13]

THANK YOU!!!


BillA is THE man!!!!!

[bigdawginva]

Both are excellent reads. Well done and thank you.

That said, would it be possible for some to offer some WCing "Cliff Notes" for the following...

...every effort should be expended to identify and select components having low flow resistance.

The present emphasis in pump selection on 'flow rate' must shift to an increased awareness of 'head capability'.

There is little point in running a pump at 1/5th its free flow rating because it does not have sufficient pressure capability. Those WCers interested in performance must accept the need to understand - and use - pump P-Q curves to evaluate what will be most effective.

Performance costs money, why would WCing be any different than other 'go fast' activities?

Where the purpose of WCing is to decrease noise, then the criteria for pump selection shifts slightly, but the very same factors will determine the system's performance.

It must be stated again that this kind of testing does not identify a 'winner', although clearly some wbs are inferior in some (or even all) respects as compared to others. It is up to the WCing system designer to identify those wb characteristics of significance, establish their relative importance and to then optimize the tradeoffs and compromises necessary to select complementary components of the WCing system.

such as, based on the results of the two articles and, the types of components that are widely available to the average enthusiasts, what would be some of the combinations that would adhere to the above guidelines for optimal performance? I've grown tired of reading posts and reviews from those who have no more knowledge than I do but recommend this WB or that pump or this kit or that rad simply because that's what they use or like or are paid to promote. I enjoy reading articles such as the above but still leave with the question "so what specific components should I put together?"

I would be interested in some unbiased suggestions from some of the experts like BillA, myv65, bigben2k, and others for optimizing a WCing system. Such as, if 1/2" system use WB "A" + Pump "B" + rad/hc "C" or if WB "B" is used then Pump "C" + rad/hc "A" should also be desireable. Or, if using a 3/8" system, then WB "C" + Pump "A" + rad/hc "B". These recommendations may also require assumptions such as all components self-contained in a server tower case where the rad/hc is no more than 18" above/below the pump, etc. I realize making these type of recommendations is not just black and white or easy but certainly some of you have opinions on what some of the available components are the most complimentary. Has anyone ever thought of designing some sort of WCing matrix or table that could identify some solid combinations?

I'm still learning all of this and I'm not an engineer. I don't have the resources to effectively determine what tradeoffs are necessary and to determine which components are the most complimentary.

Having said all that, and referencing some of the widely available WC products, what are some complimentary combinations? I realize I'm cheating but I'd like your advice.

[Joe]

Originally posted by unregistered
why Ben ?

Joe Kelly made me do it ! (its all his fault !)

after giving him the redas* for BSing w/o facts (or research, ahem)
I thought to test rads (and later wbs) in my garage no less
well, it was much much harder to do (well) than I had thought, MUCH harder
but I know a bit about testing so I persevered (over 2 years now, and thousands of $)

I had naively thought that the mfgrs would have an interest, and they do - in their profits
the link between factual product performance data and sales is unknown
-> because no mfgr uses any real data at all, and they sell their wbs just fine, thank you

one business plan bites the dust

and a website can only pay if its a cash cow, and we know which site that is
I won't play in a sand box that has turds in it, so that's out too

hi ho

I can respect your candor, but I also have to point out you are not all that and a bag of chips either. Everyone has ****ed up at some point. ( and yes Bill even you are fallible.. as much as that may on its own cause all of existence to disappear.. its true.) As time goes on methodologies evolve and accuracies change. Being a young geek and not an old fossil, I have the ability to think with an open mind and not assume that just because something isn’t to full NIST standards and to .001 accuracy that it’s "BS". I don’t have an engineering degree, or anything spiffy like that ( I also don’t have thousands to spend to test 50$ water blocks) What I do have is a true love for cooling, I learn more every day, and coming from the outside I have few preconceptions that limit what I will try.

There is value in data if many types not just one set or type of data. I understand that you wear your many years of NIST committee and college stuff on your shoulder with pride and use that to validate anything you say and I can respect that. But you gotta come to grips with the fact not everyone thinks like you do (Thank god, imagine all the shootings that would happen!), and everyone sees testing a slightly different way and has different goals for what data they want to get out of the tests.

Even while you may call the testing I did "BS" its still light years ahead of anything your favorite [S]ite (or its wizard of a reviewer) has done. There is a big difference between testing where NO thought is put into it ( ie: the [H]ype zone) or where profits are the only motivation, and testing that while maybe primitive or not to what you would do, still was thought out with goals of making it the best test possible with the time/resources available. I think its more about having the drive and intent to do the best testing you can with what you have, than it is about setting goals that are unrealistic or unachievable with what you have.

I respect your work a lot Bill, back nearly a year ago I got over the fact that you were just an ass in the forums, and saw your testing was real well done. But I think your 16 year old hormonal teen type posts really do a discredit to your articles/testing, and overall devalue your talents as a technical reviewer/tester. Your flak to me is fine, I learn from my mistakes and grow ( yet another advantage of not being a fossil). Also I run a cooling website, I have no illusions that the stuff I post isn’t open for critique, and ridicule by the readers (that’s also how I learn more and grow).

One thing I do find interesting is the recent revelation that you have come appon about mfg's not wanting to pay to have their product "really" tested. I have understood that there was a separation coming for some time. Where the more radical and different testing methodologies would be cast aside no matter how exacting or accurate they are in favor of a "common" or [H]ype driven review that contains really no info or data . I mean you have to figure... for a company like CPUfx, do they want their products tested by a "real" testing setup that could put contrasting data up compared to what the other testing sites saw? Or see that their block really performs like ass?. It’s about like a fat person avoiding a scale, he knows he’s fat, but avoiding the scale at least leaves the possibility that he’s realty not fat but just has a “bad perspective” on things .

And yes review sites cant make money… just doesn’t happen unless you are being paid off or have some big MFG’s hand up your ass (someone say [i]ntel?) But that doesn’t matter to me, I don’t run anything for a profit, I do it for fun. I just find it funny to see what [W]ebsites do for a quick buck. ( was that an 8 or a 3?)

I did like your write up, was a good read. I learned something from it, which is all I ask from anything I read on the web! Your follow up was equally good to read.

Hey but on the lighter side, I was just given credit for helping improve the lil cooling niche, guess that counts for something… think I am going to go and buy me a hot fudge sundae to celebrate this

[BillA]

mercy Joe, cut me some slack

indeed we all evolve, you have heard me say (I'm sure) 'every day is a new lesson'
- this week for me is all about 'resonant reservoirs' causing pressure pulsations, bleh !

and I quite agree that there are, rightfully, different types of testing useful for different purposes
I just get really difficult about unreal data used to 'support' bogus conclusions

I was not giving you sh*t, Joe - just another of my backhanded complements
for you did indeed get me into all this

hey Joe, are you fat too ?
<jk, eh, jk>

EDIT: and thanks for the nice words
those 2 articles = 8 months

[Joe]

hehe fat? Nah, I think I just have a bad perspective on the situation I fear the scale

hehehehehe

well I guess I owe you an appology for getting you into this backwards and screwed up industry hehehe

Edit - See, I can totally relate to that 8 month thing for a quality test. I really dont grasp how places like (dare I use the brackets again?) think its SOOO simple to do. or the kids who say it takes them 15 min to test a WB fully...

OH and the fact your testing setup must look good in order to be functional hehehe cause you know.. ghetto looking stuff doesnt test good at all... atleast thats what mr. muscles sez

[bigdawginva]

Okay, now that we've had our group hug, any perspective on my request?

[MadDogMe]

Good luck on getting that kinda answer out of BillA ...

PS, has anyone worked out the question to '42' yet?...

[Stepback]

A very good read Bill !!

Joe and Bill: This is the way two people should handle this kind of disagreement. If we could just get other people in this little nich world we play in to think like this .....

bigdawginva: I would like to see someone do this as well, but the cost and time to do it are just not very realistic. As Bill has said the mfgs don't want that kind of info out here for all to see.

[bigdawginva]

Good luck on getting that kinda answer out of BillA

Perhaps, but there are others I'd be interested in their opinions as well.

[murray13]

bigdawginva: Personally I like the fact that they don't pick or endorse a specific product. It lends credibility to testing and or their view about a cooling subject, (wb's, pumps, rads, etc). Maybe it's just my engineering side coming out wanting to find my own solution to this very complex problem. BillA's articles help out a great deal, but to do what your asking he would have to test every wb known. And even then he could only give you which wb did what during the test. WC is such a complex issue it would be VERY, VERY difficult and very time consuming to create a matrix type chart. The real benefit I get out of his articles is he puts some numbers on parts that the manuf. don't.

If I was able to get detailed specs on wb's, and rad's, things like pressure drop at a given flow rate, C/W info, etc. It would be relatively easy to pick out complementary pieces to fit into a system. It's the lack of information given and the interdependency that makes the question so complex.

The only thing WC'ing gets like this is pumps. The manuf's give you the detailed specs (for the most part). But without the rest this info only helps in relation to other pumps, and really won't tell you if a certain pump will actually perform in your 'system' any better or not.

I don't want to trash the IDEA, that's great, I'd love to see someone do a matrix. I just don't think it will happen any time soon the way you would like. And if some one does do one the first thing I'll do is try to corroborate the info somehow, but with very few doing really subjective testing that will be hard to do.

OK, OK, I guess I woke up on the wrong side of the bed this morning. I normally just keep my opinions to myself, but there's mine.

[gmat]

A matrix would require centuries of work.
But what BillA showed - one more time - is that 'top dog' waterblocks (Innovatek, TC4 and others) just perform about the same. No one will really notice a difference in 'cooling performance', out of measurement imprecision.
So which one to choose ? There are more important criteria now:
- mounting system
- fitment on a given mobo (does it touch capacitors...?)
- finition. Does it require lapping ? Surface geometry correction ?
- ease of maintenance. O-rings, plexi vs alu tops etc..
- price, shipping, availability
- barbs (nylon ? chrome ? brass ? other ?)
- sturdiness
- and for teenagers, eye candy.
I think a good WB review (without the multi-million-dollar HW that BillA exhibits) would compare those points, setting totally aside the 'cooling performance' of any given WB.

[bigben2k]

BigDawg: what's your request?


I think that there's a big gap between what Bill presents, and what the perspective watercooler is facing. It's a given that watercooling is recent, but I think that Bill Adams has pioneered watercooling better and by far, than anyone else.

A major part of the gap is in heatercore performance. What we need is Bill type testing for all Fedco cores, and to build a website that would allow someone to select all necessary components, so that they know exactly what to expect whjen they receive the parts.

We're VERY far from that.

Another point: I think that Bill has pushed watercooling theory right at the limit, if not a bit beyond, the cost/benefit aspect. EVERYONE owes Bill Adams a big "thank you"!

[bigdawginva]

BigDawg: what's your request?

My apologies...I assumed you read my post...

Having said all that, and referencing some of the widely available WC products, what are some complimentary combinations? I realize I'm cheating but I'd like your advice.

EVERYONE owes Bill Adams a big "thank you"!

Clearly, I'm not in a position to speak for "EVERYONE", but I seem to recall at least 4 "thanks" or "Thank you" above.

[g.l.amour]

heh. indeed; lots of things have started dawning on me. but ofcourse there are alot of ppl to thank. billA's top point is that he is educating us. normally teachers get good pay, he seems to do it out of ... dunno ; crazyness?

lets say. all billA's ranting about tim joints. crap, i used to think, what difference will that make. i mount the damn WB and thats that.

so yesterday i mount my new WB, and started getting temps almost the same as my best maze2 mount. u can figure that i was not all too happy about that. so , some good ideas come at night. the next morning i remeber billa's deflection of temps through different mounts graph. so i said to myself: u would be one lazy bum not trying to find out what the deflection range of my specific block is.

so after redoing the mount, temps have dropped (as far as i can accurately notice) +/- 2°C.

[BillA]

good to see you critiquing your own results g.l.amour
while the kiddies and egomaniacs seem to get upset with this,
(self)criticism is the path to improvement

looking even harder at the TIM joint parameters;
the range you will experience is the sum of the independent variables
- goop application
- mounting pressure and uniformity
- wb bp surface flatness (more so than finish)

some vary with each mounting, others may be wb specific
'control' is achieved of each independently

bigdawginva
thinking about the 'answer' your question merits

[MadDogMe]

What's your take on the best setup bigdog?, how would you go about working out the best block for the best rad for the best pump for the best tubing (assuming they are all 'best' for each other as well once joined as a system) without specific information that just is'nt available for anything except pumps?...

BillA has gone to alot of time and trouble to 'start' to give us the information we need to work this all out. short of going out and buying one or two of each and every worthwhile component, popping some amphet' and staying up for a few weeks it's going to take some more time yet...

Also given the amount of time and trouble BillAs gone to, I think the least we could do is go and look at what info is available, come to our own conclusion, then ask BillA for his take on it. don't forget he's a teacher at heart, not a walking encyclopedia. if you're prepared to work and learn for yourself (from what I've seen) BillA will help/steer you along in his own way. ask him for a short answer and you'll problly get a shorter reply than you bargained for . I think he has no interest at all in helping people who are'nt prepared to help themselves, or at least try(asking nicely can help as well)...

we'd all like the short answer, but then knowledge is useless without understanding, and anyway where's the fun in cheating? , you never prosper(or so it's said )...

Failing that: Swiftech UH with 3/4' outlet 3/8'~1/2'inlet or OCPC Atlantis with same barbs, Iwaki MD20/30RZT, 3/4' hose all through. Serk rad or twin parrallel OCPC 'Abyss' rads with 3/4' barbs ...