An End to the Insanity.

[pHaestus]

It's frustrated trying to be the voice of technical expertise when it's attacked on all sides by legions of monkeys flinging poo here on the forums. Stay if you want. Do what you want. Have fun.

You're right Etacovda; I am also an unwelcome voice/opinion here on these forums as well. Take it easy.

[Cathar]

Originally Posted by pHaestus

And it is personal. This whole shit storm has become an us vs. them thing, and I am sorry but I am going to side with the people who have training, experience, and education in technical matters. We have a union, after all.

It never had to be personal. We were pursuing this in a scientific manner, until someone started throwing the insults around, gosh darn, for daring to ask questions.

The crap about the ttv has poisoned what is, in my mind, the only realistically attainable way of CPU-based testing. That being the Socket 775 with TC in top of IHS. It's also run off or silenced pretty much all of the people I cared about interacting with in this hobby. So yea I'm an angry bitter man. Enjoy.

I thought we were trying to figure out a way in which we could quantify the unknowns.

Why did that have to become personal?

When did the enthusiast user base lose faith that it has the capacity to cater for more variables, and out achieve something that comes in a box from some commercial entity, to the point that we all throw our hands up in the air and say it's all too hard?

Incoherent is keeping the faith alive, and is doing an excellent job. bobo is analysing it all too as well. You have your extensive experience too.

We have a mountain of experience here across many individuals.

Since when did we all decide that we couldn't do better? I mean, isn't that the attitude that started this whole web-site in the first place? That somewhere, somehow, a bunch of collective enthusiasts can come up with ways that equal or better what some commercial entity is pushing out?

When did that spark die? Why did it die? I've been arguing for that spark to come back, not against it. When did we all throw our hands up in the air and say "Nope. Can't do better. Sure there may be issues with the boxed thing, but screw it! Let's give up!"

[bobo5195]

Nice to have you back stew. Unfortunate that Procooling seems to be reverting to old ways. There is not even creative use of swearing anymore tsk tsk. I thought we were ending personal attacks. Everyone hear is coloured by experiences that is why we are hear we all know something different. Maybe we should pick a random newbie off the [H] forums to decide and argue? At least they won’t have bias.

First off I quite agree with the bit tech review intel cpus are heat load limited if intel was willing to produce TDP max 200w cpus I’m sure they could get to 5ghz quite easily (a real world tech article pointed out that 65nm chips have 500mhz chopped of the design case simply from heat), but users would hate them and most computers wouldn’t use them. It was an example case of what good cooling allows you to do.

This is no where near as complete as I wished but here goes. Sorry about my grammer and stuff but the red bull is wearing off.

My take on all of this is that the TTV is a very good setup, I do take mgfr data from the TTV as near perfect, jaydee. It is a factory made industrial design thermal testing rig after all. The results for all TTV setups must be identical if correct methods are used and I disagree with any talk of them being flawed as they are unknown. Intels only possible course of action is to use the best methods available so not knowing what the test does is less important. For intel there is too much riding on it for it not to be a god test. A few dodgy heatsinks get through it is not worth it for them as a company. The methodology in using it is perfect the results are repeatable and it is accurate. It is perfect for testing. If every website in the world used it for testing then all reviews should come up with the same numbers and there are smilly faces all round. It can also seems to be a true model of real life. Secondary loses through mobos are included TIM and heatspreader is included.

So why then if it such an uber well made test do I think that it is flawed? There are a few little things all related to the fact it is a verification not testing device as such it is a flawed platform for testing. In short it is not as accurate as other methods, less variability between results but less resolution 0.5 ± 0.1 degrees(TTV) compared to 0.1 ± 0.2c(DIY) say.
• Results only give as degrees. No reporting in Watts. Alright this may seem like a nitpick but it is an exceptionally important variable. It means that you can compare waterblock A to B but you can’t test them under different loads. You can’t tell performance on different chips as you can’t do a C/W conversion.
• I think that its accuracy is low for two reasons.
o No modelling of actual CPU. I don’t believe cpu die variability is included and a standard heater is used. This is a fine approach for validation and admittedly not many other testing rigs have this but it is still a flaw.
o Inclusion of IHS skews results. The use of something in the way is going to modify results. Accuracy is going to be lower as heat has to travel through the IHS and its TIM, this can be seen in the reduction of gap between the apogee and storm in tests carried out on real cpus. Like working out engine power measured at the wheel compared to crankshaft.
• We don’t know how the TTV compares to real case CPUs. I presume it’s a worse case value. Ie this is simulation of the worst CPU from a heat and IHS interface point of view that intel would release into the wild. As such it does not measure the case of real cpus well and we do not know what correction factor to use to convert up to the “average” cpu in the wild.
• I am worried about the thermocouple and groove effecting results. It is not a true model of a cpu and as such prone to error.

As such DIY made rigs are better because we know more. The best rig for CPU testing is a bare die precision rig. Heat in is accurate, no other effects going on and as such because we have followed KISS less potentially to go wrong.
• We can model other effects and as such vary them. Switch between bare die and IHS easily. New IHS comes out we can correct and compare to previous data. New testing rig results can be directly compared as they are both accurate heaters their results should be identical. New cpus, new mounting methods all correctable.
• Heat input can be varied and quantified. We can dial up the heat and test high thermal loads. This is very important as while non-dimensional analysis means we should be able to scale result this is often not the case and in any case there aren’t the people around to do the conversion formula accurately.
• Accuracy is greater as we know exact numbers and there is greater resolution of the testing device. The real world is getting in our way.

In short we are trading accuracy and ability to examine different cases for repeatability. For testing this is a good compromise as accuracy is the most important but for a verification rig that intel has produced in the TTV it is not an acceptable trade off as repeatability is key.

What we need to do is make sure that other variables are accounted for. Data on secondary losses should be quantified. TIM / IHS data be put back in and corrected for.

[jaydee]

Good Lord I guess only parts and pieces of my posts are read.

Originally Posted by jaydee

I am not sure why we are even concerned with TTV results. None of use have access to one and the results of it are only used in Swiftechs specs. How many of you take manufacture data 100% seriously? I sure don't. Not saying the data is bad or wrong just saying PR could possibly contaminate such data or maybe they are out right lying. Who knows?

What pH, Robotech and any other reviewers role IMO was is verifying that the manufacture data is reasonable. Not necessarily accurate but reasonable to think it is.

As pH said any way you test there are flaws. Just have to decide if those flaws are acceptable or not.

Am I wrong here? Does pH know all about the TTV to make his claims that it is the best thing out there to test with? If none of us have the info then how do we know? I am saying it doesn't matter one bit about the TTV because no one here will be able to use one or share info about it so it is not worth bickering over.

Originally Posted by jaydee

Knowing those variables is nice to. The TTV is pretty unknown to most of us. That is why manufacture data should be taken with a grain of salt. We don't know if they used variables that are positive for their PR. Many do many don't. We just have no real way to know until it is independently verified that their data has the possibility of being correct.

Some here (will not mention names) would seem to have us think that manufacture data should be taken as fact until proven wrong. I see it the other way around. It is wrong until proven right. Especially in industry such as this that has no real regulations on it.

Pretty much what I said in the last quote. Added that a few people here would like us to believe their results should not be questioned because they are elite and/or business men trying to push into the market or have in the past.

Originally Posted by jaydee

What does the TTV have to do with all manufactures on the planet. Which I was referring to, not just heat sink outfits. Swiftech used the TTV but we have no clue what version of the TTV was used or how up to date it was. Bill claims he knows all yet was a praising and USING and CHARGING people using this now acclaimed chunk of copper made in a garage by us dumb ass enthusiests..... And he wants us to take his testing superiority seriously now and believe the Swiftechs TTV results because of why again?? Oh yeah because HE did the testing? I would be a lot more convinced of the results if it was done at Intel in their controlled lab with their engineers. But it was not.

Can't see how I discredit Intel here.

Originally Posted by jaydee

I am not even sure why you keep backing the TTV. It doesn't matter to any of us. We will never use it. We know nothing about it so we cannot even begin to understand how right or wrong it is. Also was it the same engineers that designed the original P4 that sucked ass? And you want us to put faith in them? The original P4 was nothing but PR hype. Even with all their fancy specs and bull shit in the end it was crap and they knew it yet they hyped it up as the best CPU ever. Now you are telling me I should just believe everything they say because they should know what they are doing and blindly buy the CPU?

Here I am not specifically jabbing their engineers but pointing out that they are not nearly as perfect as pH would like us to think. It is well known the original P4 was junk and the P3 was better. How am I supposed to know they didn't make the same mistake on the TTV? I don't. Do I think they did? No, I think they did better than anyone else but I have no way to PROVE it. Which is my point. Who is to say the failed P4 was even the engineers fault and not some pencil pushing fool wanting a new product out fast? I don't. There for I have to no believe any of there info UNTIL IT IS PROVEN. I am just asking for proof.

I still don't understand why you keep on with what Intel povided.... I am not arguing it maybe the best test bench ever. I am simply saying it doesn't matter. What does it have to do with you, me or anyone else here? No one here is going to be able to make such a bench or even use one for that matter. Even if they did use it they can't disclose anything of it. It is irrelevant and that is what I am saying. We should completely ignore the fact that Swiftech used it IMO and just take the data provided and try to prove it reasonable or not. That is all we can really do isn't it?

Yeah really bitching about Intel here.... I don't see how I am wrong here either. Who has a TTV at their disposal for all us to learn about?

Originally Posted by jaydee

I think we can all agree Intel would not make a high tech test bench to skew results. They have no reason to. We are saying other companies that use the device may not use the most valid results or simply may not know how to work it properly. The thing must have some sort of learning curve to it. Also there are several different models from what I understand. Which one is being used for testing and how is it different than the other? Does the manufacture even know if they are screwing up or not?

To think we can just take manufacture data as perfect is not a good thing.

Still fail to see were I discredit Intel.

Originally Posted by jaydee

Ummmm ok, then explain all the bull shit products claimed to be the holy grail of cooling by the manufacture....

Oh and let's not forget this: http://www.bit-tech.net/news/2006/03...liquid_cooler/ Are you telling me this is all it is hyped up to be by Intel?

Read the article. He gather info from the event he was at from the display from Intel. I don't think he made all that much of it up. Still not discrediting Intel's engineers just Intel's PR as a point companies bloat their products up no matter how big they are.

I thought we were here do decide FOR OUR SELFS with OUR OWN testing procedures and equipment how well something works. Not blindly take manufacture data at face value. If that is the case why bother reviewing anything at all?

Ok, that will be my last post in this thread. I don't even see a argument really. Just a disagreement.

[Joe]

pH, myself, and many other readers have every right to be bitter that this BITCH FEST has many people who just keep throwing extremes and absolutes out instead of talking about reality. I was personally against the idea of the TTV initially, but after doing my own research I came up to the conclusion that it is a viable test platform.


I don’t need someone to read from their own personal pulpit of narrow minded testing, some song about how big companies cant be trusted, but some error ridden garage setup can. I also dont see why everyone needs to agree... Make up your own mind and go that route... we arent lemmings after all.

The thing I do find funny is if you didnt know which "side" of this cathar is on, much of his post could be in a Pro-TTV camp

Cathar, your the one who did the dramatic "Im gone" Pm's in a way to pressure us to ban Bill... so yes this is the shit you get when you try to pressure someone with idle threats. We didnt tell you to leave, were just encouraging you to be true to your word.

[Joe]

Unfortunate that Procooling seems to be reverting to old ways.

you mean like before 4 months ago? it was nice back then... that would be old ways. This... is just following the path that cathar and bill and others have set us on. As pH said, this whole topic has been poisoned by emotion now.

[Cathar]

Originally Posted by Joe

The thing I do find funny is if you didnt know which "side" of this cathar is on, much of his post could be in a Pro-TTV camp

That's 'cos the only side I'm on, is the side of removing the unknowns. The TTV does about 90% of that. IMO, it would be an excellent testbed and solution, if one glaring unknown was solved. I believe that I've been saying that for, oh, about 3-4 months now.

It's not "funny" at all. It's not about sides. It's about answering the questions. Others have done their absolute best to paint me as some unreasonable individual motivated by Apogee vs Storm, and wants to see an end to all testing that I don't approve of.

Far from it. I'm pushing for the exact same principles as everyone else here is - answer the questions - quantify the unknowns - and then everyone's happy and there are no doubts - just error bars. That's what got Procooling respected in the wider community in the first place - by doing it one step better than everyone else.

Cathar, your the one who did the dramatic "Im gone" Pm's in a way to pressure us to ban Bill... so yes this is the shit you get when you try to pressure someone with idle threats. We didnt tell you to leave, were just encouraging you to be true to your word.

I said to you that Bill's continual insults and slings against the enthusiast user base that makes up the majority of this forum's base, and the forum moderators continual unwillingness to act, was causing damage. I pointed out that it had driven many away, including myself taking a two month leave of absence to see if it cleared anything up. I pointed out that the attitude was killing participation in the forums. I said that I was certainly getting sick of it, and it was killing my motivation too, hence my comments about leaving.

Hey - I'm just a messenger. Ready. Aim. Shoot!

Or maybe I'm just a stubborn ass who after cooling down from a heated emotional moment decided that he loves the hobby, more than he dislikes some individual.

[Etacovda]

See, theres your problem from the start joe, you assume people have a 'side'. The people that have a 'side' are clearly siding on the 'Intel did it!' team.

The rest of us just want quantification, and when we ask for it, we're faced with 'intel did it, it must be right'.

Its sickening. If the testbed was made by someone other than intel, I guarrantee questions would have been asked about without all this hoopla and bollocks.

Once again, what the HELL is wrong with asking questions based on fundamental scientific/eng procedures?

Maybe we should pick a random newbie off the [H] forums to decide and argue? At least they won’t have bias.

nah, they'd either be bitching because its not pretty enough, or that its worthless because its not made by aqua computer

[Joe]

I agree that people should want quantification of all aspects of a test setup. But we havent had it yet in any setup ever... so why the big concern about it now with this option? I mean I understand a concern, but I dont remember 3 months of ranting and raving threads about die sims, or Bare die AMD cpu's...

[Etacovda]

because this uses an IHS, with a possibly variable tim, with results in the past that have been suggested to be suspect. If you want further, ask Bill or Cathar, but they're probably both under NDA.

Do we know what the heater element is? do we know what wattage it outputs?

The difference is, that when someone had a question with say, pH's testbed, pH and others were WILLING TO TALK about it and figure it out, not just stick their damn heads in the sand and claim that it was right, which is what happened with the TTV.

The die sims are extremely simple, and easy to understand from an eng POV. The variables can be understood and accounted for in one way or another, and its a solid bit of copper...

Cpu's are always suspect, pH is the only person whos done a decent job on a CPU, as is most peoples opinion.

I dont see the hard thing to understand here....

[Cathar]

Originally Posted by Joe

I agree that people should want quantification of all aspects of a test setup. But we havent had it yet in any setup ever... so why the big concern about it now with this option? I mean I understand a concern, but I dont remember 3 months of ranting and raving threads about die sims, or Bare die AMD cpu's...

Actually I recall about 4 years of ongoing investigations, qualifications and attempted quantifications on the bare-die->heatsink TIM layer. For the large part of it, it was all very mature, organised, and amicable. People asked the hard questions, people with experience and open minds and skills came up with the experiments, the testbeds, the tools and the answers, and after about 4 years we started to develop a pretty fair understanding of the issue, and quantify the TIM layer within a certain set of assumed circumstances, that being flat and even pressure between the CPU and the heatsink. That non-flatness and uneveness caused variations is, to most everyone, an automatic given. The work done here in the community is excellent, and sets Procooling above and beyond most any other enthusiast forum out there.

Along comes the unavoidable march of progress, along with IHS capped CPUs which simply cannot be ignored any longer, and testing devices such as TTVs. Naturally, as one would expect, we attempt to start the process of exploring the variables, but this time everyone who doesn't believe that TIM1 is (effectively) invariant, even when faced with non-flat or uneven and/or flexing base-plated heastinks as possible problem scenarios, is being shouted down, subjected to derision, and being treated like they're mindless bigoted idiots simply for asking the same questions that formed the heart and soul of this community for the last 4 years.

Don't know about everyone else, but I'm still asking myself: "What changed all of a sudden?".

[Angry_Steel]

Does anyone here even have a TTV? Is anyone planning on getting one soon? If not, then I really dont see how any of this is relevant to anything. If Im understanding right, these are not available to the average joe and lots of information is not available either under NDA's or some crap. So what is the purpose of arguing over something that cannot be proved or disproved because of

A. A lack of Equipment in the hands of the independent tester.
or
B. A lack of Information in the hands of the independent tester.

Is it really neccessary to fight over this? Would it not be better to quit talking about things you have no control over and actually do something positive with what you do have? There comes a time where beating a dead horse ends up in a broken stick and a sore arm.

[ricecrispi]

I don't think Intel would build a plaform that skews data in there favor. The TTV was designed for them to perform their test to met their goals. It MIGHT work for Intel but might be out of spec for different purposes.

TTV might have limitations and "exploits" that other companies have used or didn't know about and adequately address the problem. TTV might still be viable if we have a clear understanding of whats going on but we don't and make wild acquisations.

We can't test the TTV. We don't have a setup to reproduces results or the test the ability of TTV. How can we perform test and reproduce results and quantify variables on TTV when its not available. We should considered TTV to be an anomaly. It might be useful but is out of reach for usand we should move on.


We already have equipment and test that are adequate enough for WB testing. We should just continue on and find this TIM variablity. IHS will be around for a while so its worth the trouble. It will help improve WB performance if they offer this dulling effect like robotech hinted in his test.

Im going to enjoy some SAKE now.

[Voodoo Rufus]

Originally Posted by Angry_Steel

Is it really neccessary to fight over this? Would it not be better to quit talking about things you have no control over and actually do something positive with what you do have? There comes a time where beating a dead horse ends up in a broken stick and a sore arm.

Wouldn't this be nice. Something so obvious yet ignored so much.

[bobo5195]

Mmm sake...

Can we take the dulling effect as fact. It is obvious the IHS has its own thermal resistance and as such the thermal resistances are additive (im not going to go further an work out the effect as my brain isnt working at the moment and i dont want to look like and idiot)
R(measured) = R(ihs) + R(Block)

This as the blocks get better and better this effect gets more and more.

[Incoherent]

I think that if we can establish the TTV as a viable method, we can emulate it.

I think that soldering an IHS to a heat die (the Mushroom solution suggested earlier) would yield useful data. It might be close to a TTV in behavior with the added benefit of an inline sensor. It would simplify the problem of calibrating the heatshadowing effects by dampening the variations caused by different blocks.
But I still am leery of the groove for the thermocouple. I am pretty sure that something designed for testing thick based heatsinks behaves unexpectedly with thin based waterblocks.
Haven't got a handle on this yet.

[ricecrispi]

Originally Posted by Joe

I agree that people should want quantification of all aspects of a test setup. But we havent had it yet in any setup ever... so why the big concern about it now with this option? I mean I understand a concern, but I dont remember 3 months of ranting and raving threads about die sims, or Bare die AMD cpu's...

Well there was a little hoopla over the test performed by joe Citarella on his prototype thermoshypons. Their was also some confusion on his test beds with a change in die sims size and different cpu testbeds etc. His ethics were also questioned since he was assisting in designing for some company.

http://forums.procooling.com/vbb/sho...=Thermosyphons
http://overclockers.com/articles1246/index02.asp
http://overclockers.com/articles1248/

[Etacovda]

aaaanyway, it clear that the sides are FINALLY understood now.

I think Incoherents idea of the emulated set up would be useful, incidentially its what i pushed from the start.

Would it be useful to have a mushroom shaped solid die, and a soldered variety within the same tolerances to see the TIM effect of the solder?

With this emulated set up, it would be possible to do the groove and the die temp, obviously allowing better understanding of the bond.

[ricecrispi]

This dulling effect might not appear in intel IHS as they seem to be better than AMD's and actually do their job. Still can vary from cpu to cpu.

Robotech claims to experience this dulling effect on AMD socket 939 cpu test. There is a good possibility and I decided to donate one of my 3 939 cpus to Robotech to see if its true. Should've done it 3 months ago. .... a well

I was going to sell my one 1 of the 2 cpus but ended up having 3. I can keep one, sell one, and donate one. Realized if I didn't miss the $100 then, I won't now.


====
As for the IHS on heat die, cant emulate TDP patterns of actual CPU but does account for IHS influences and can test TIM layers by varying them. A step in the right direction.

[pHaestus]

I think it's a lot more complicated than that etacovda. What did someone say the other day? Everything affects everything? It's a 3 dimensional problem basically and by using a much different TIM joint on the die sim (solder) than a real CPU I think you'd get pretty different results in terms of spreading. Also how does Intel do the soldering so that the system is flat when compressed via hs/fan? I don't know how one would reproduce that process?

Oh and you should change your signature to:

Procooling: Where people with high school level physics teach engineers and Ph.D chemists about the scientific method.

What arrogance! Since you bring up the scientific method though and have formed the hypothesis that Intel can't simulate their processors then by all means experimentally test that!

//edit: You realize that you're setting up a situation where everyone with technical expertise will refrain from publicly posting and the ignorant will take upon the role of "experts" right? Why not just cut out the middle man and go to one of the many other forums that are further along in that process?

[TerraMex]

hmm.

[Voodoo Rufus]

As long as the replacement solder suggested for the emulated die sim is close in thermal conductivity to the original, it should work fine.

[pHaestus]

How do you test for that Rufus? Remember that you cannot rely on any information from Intel to make an estimate...

[Incoherent]

Originally Posted by pHaestus

How do you test for that Rufus? Remember that you cannot rely on any information from Intel to make an estimate...

Don't.

Model the problem. Make the test result conditional...

Assume it's perfect. Then assume its so and so much. Then assume it's so and so much + 30 banana's. Then ...+60 bananas...+90 banana's

Pretty soon you have a curve, dependent on what it might be. Compare two curves, one will be better than the other over a certain value of banana's. That can be your FOM for a given block if your model is valid... Performance/unknown.

That's how I would do it in this situation. Make the unknown a variable, see what happens. Chances are the better block will be obvious.

Pondering on...

[Joe]

Ok... I dont have a banana button on my calculator here... I am all kinds of confused!