HD6970 Cooler comparison Review coming soon / First look

[sneekypeet]

From your link "does placement of the pump in relation to the CPU block matter? Specifically, does the increased inlet pressure from placing the pumps directly before the block increase performance as suggested by many". that is one block one rad, not what we are discussing in this thread

 

[crazyeyesreaper]

ding ding we have a winner

 

[EarthDog]

Again, conceeded the fact that if there is low flow (pressure) this could be an issue... assuming its not in this testing enviroment.

 

[EarthDog]

Here you go gents...



Lulz..damn your attachment really shrinks down the picture. Click plz.

 

[sneekypeet]

just because he said it doesn't mean its factual. If my loop only produces 1 degree increase out of my CPU block and drops right into my GPU block showing more than 1*C change, how do you explain that by air? It hasn't even hit the rad yet, so his take is just odd!

 

[EarthDog]

Not sure man... If there is anyone to believe, skinee, Vapor, Martin, etc are the men for watercooling. If you cant accept that info, well, lets say Im quite clearly not the one who is being stubborn here (on those points).

Thanks for the discussion, and Im glad it mostly stayed within forum rules.

Soooooooo about this guys testing... can we see it already?!!!!!! hahaha!

EDIT: I still think his testing is just fine IF the scope were differences between the blocks and not how cool does the block keep a card, but at this point, its best to just let this rest.

 

[cadaveca]

When you find an actual physicist, then you can say that, but, without that, those guys posting stuff is no different than you or I posting it.

The amount of "wattage" that rasies water temps depends on flow, water volume, and loop length, as well as rad size and fan airflow on the rad.


The point you are missing is that when testing ANYTHING, the "proper", scientific approach is to remove any variables that might affect the results. As the CPU block can add restriction, affecting flow, and this restriction will change from block to block(because restriction affects flowrates), the results obtained with one CPU block may not be the same as with all other CPU blocks. When 1-2c may be all that these blocks differ, everything matters.

That is all.

 

[EarthDog]

So now Vapor isnt good enough? It was in post #90 from you...why isnt it good enough now? (you really shouldnt answer that... )

Yes, yes... keep bringing up points I agree with...with that thinking all the GPU block only tests are not appropriate to take outside of that test either since every loop is different... notably most contain CPU blocks.

Anyhoo...

 

[cadaveca]

keep bringing up points I agree with

Actually, I've had the exact same point all along. And since you say you now agree, I guess you're just trolling?

 

[bbmarley]

did anyone else notice thats a 6950 on the desk

 

[cadaveca]

did anyone else notice thats a 6950 on the desk

reference 6950 = 6970, basically.

 

[EarthDog]

The amount of "wattage" that rasies water temps depends on flow, water volume, and loop length, as well as rad size and fan airflow on the rad.

Thats what I agree with.

Not a troll dude. Just trying to sort through the trees to get to the forest.

 

[cadaveca]

Thats what I agree with.

Not a troll dude. Just trying to sort through the trees to get to the forest.

OK, but listen...a CPU block affects flow. So it will affect results. So I asked that GPU-only testing takes place. Elimination of factors that affect results.



I think you see the forest, and not the trees, except those you chose to look at.

 

[EarthDog]

But if its the same block in the same loop with the ONLY difference being the GPU blocks........I dont get whats wrong with the testing OR what makes isolating the GPU block a better method (considering most loops contain CPU blocks and more load!!!). Sure the CPU block adds restriction and impedes flow and does NOT isolate the GPU block. BUT, and back to yet ANOTHER point I and sneeky brought up, is that we dont know the goal of the OP. If he is testing for absolute temps, meaning this block will hold temps of this card at these clocks, with this rad, and these fans, at these ambients, then yes, you and sneeky are correct. If he is just testing the differences between the blocks which is relevent to most water users OUTSIDE of those with poor flow, then his testing is just fine.

You edit like a madman... LOL love that feature in vbulliten

Ok... seriously... I have been stuck in the burnout area in this thread all day. I am done. I feel confident about my points that vapor brought up should make some people here think a bit about their thoughts of the behavior of a water loop.

 

[cadaveca]

Yes, I am typing with one hand, so you gotta give me a minute.

OK, so listen.

GPU temps will increase, say, 15c, from water being 1c hotter. They go up because the capacity of the water to hold heat is smaller, and the water, at that point, is less effective/efficient.

Now, end flow rates are the sum of all restrictions in the loop.

So, say one GPU block allows 2.5 GPM, but the next allows only 2 GPM, overall, temps should increase, not only on the GPU, but the CPU as well, due to lower flow rate. This will lead to skewed results, again, because the CPU also affects flow rate, so the GPU block's capabilities may not be shown properly, due to the CPU block. Will the higher tmeps @ teh CPU, becuase of lower flow, affect GPU temps as well?


The only way to know what's going on for sure, is to test blocks by themselves, as well as with others, so as to ensure that the results are truly accurate.


Edit: maybe you need to check on the second law of thermodynamics.

 

[EarthDog]

Ok... I think I have a grasp on it the thinking behind the iso method. Finally. Thank you.

BUT, I guess I just differ in methodology. I wouldnt test a block, with no other blocks in the loop. Performance will only get worse inside a loop with other items in it (right?). The testing here, while isolating the block, is a best case scenario for the block. Im guessing there are MANY people happy I dont do cooler reviews all the time. LOL!!!

Thank you, all, again for the conversation. It was enlightening for me, regardless if I agree with the method used here or if its right or wrong. I hope the information I provided about inside the loop temps and why gpu temps go up when only CPU under load will open some eyes too.

 

[silkstone]

They go up because the capacity of the water to hold heat is smaller, and the water, at that point, is less effective/efficient.

I'm pretty sure that the relationship is linear up until you get a phase change. Hotter watter, in theory should still be as effective at holding heat.
i.e. it takes just as much energy (and hence power) to heat water from 4C to 5C as it does to heat from 79-80C.

In theory, i think that so long as the water is cooler than the gpu, the temp of the water shouldn't matter.

 

[trickson]

I hope the op starts a new thread this one is all messed up . I would like to see the results at least .

 

[cadaveca]

I'm pretty sure that the relationship is linear up until you get a phase change. Hotter watter, in theory should still be as effective at holding heat.
i.e. it takes just as much energy (and hence power) to heat water from 4C to 5C as it does to heat from 79-80C.

In theory, i think that so long as the water is cooler than the gpu, the temp of the water shouldn't matter.

I think the volume of water matters. I want to say "Specific Heat"..i think that's the term here...and specific heat depends on temperature, no?

One BTU refers to the increase required to go from 59.5 to 60.5? One kilo-cal for heating one kilo of water from 15c-16c?

I hope the op starts a new thread this one is all messed up . I would like to see the results at least .

I agree..I thought this was the preview thread?

 

[trickson]

Whatever it is it has gone way off track and is so fed up now it would be to hard to find any thing of real value here .

 

[silkstone]

I think the volume of water matters. I want to say "Specific Heat"..i think that's the term here...and specific heat depends on temperature, no?

One BTU refers to the increase required to go from 59.5 to 60.5? One kilo-cal for heating one kilo of water from 15c-16c?

From memory specific heat capacity does not depend on temperature.
more reading here: http://en.wikipedia.org/wiki/Heat_capacity

The specific heat for water is cp = 4.1855 [J/(g·K)] (15 °C, 101.325 kPa)
if the flow rate is known, it should be fairly easy to calculate the maximum amount of heat (as power) that the water could carry away from the block. I imagine that most of the time it would be much greater than the actual heat (power) put out by a gpu.

 

[cadaveca]

Well, your link says:

The heat capacity of most systems is not a constant. Rather, it depends on the state variables of the thermodynamic system under study. In particular it is dependent on temperature itself, as well as on the pressure and the volume of the system.

Kinda vague.

 

[silkstone]

yea, i just noticed that when i read through it, i noticed the value of specific heat for water i quoted stated both at what temperature and pressure. So i guess it does depend on temperature, but i can't imagine it would be by very much.

I think i need to spend some more time with my head in my old physics books.

 

[cadaveca]

Dude, no need for texts. Just use iTunesU, and watch/listen to actual university courses/lectures.

If you are really interested, that's all you need. I tend to have them playing in the background while working on reviews.

I am, however, a bit geeky, I suppose. lulz.

 

[LordJummy]