Your PC ATM

[PP Mguire]

As long as you use 1 rad, you don't have much choice... You'll be dumping hot water from GPU on the CPU or the other way around.

Also, the water doesn't get enormously hot, especially with some good flowrate. That, and the saturation point of your loop stays the same, no matter in what order you put the components.

He has 2 rads though, so really no point to cool GPU and CPU in the same loop. I was always tought its best to avoid having both in the same loop unless you have an absurd amount of rad power going on and a high flowrate pump.

 

[Thrackan]

He has 2 rads though, so really no point to cool GPU and CPU in the same loop. I was always tought its best to avoid having both in the same loop unless you have an absurd amount of rad power going on and a high flowrate pump.

Hmm crap, didn't see the second rad

Cooling them in 1 loop should be perfectly OK, as long as you don't have an i7 920 @ 4+ Ghz and a GTX285 or something.

I would put one of the rads between CPU and GPU though, as you mentioned. Scientifically seen, the best heat transfer can be obtained when the temp difference between opposite sides is high. So running freshly cooled water (from the rad) over the GPU draws away more heat.
While that makes sense, this also counts for the rad itself. It's better to feed both rads with water as hot as possible (like right after a heated component) then to feed a rad with the cooled water from another rad. If you feed it cooled water from the other rad, rad 2 has much less cooling to do and therefore loses part of its efficiency.

 

[PP Mguire]

Hmm crap, didn't see the second rad

Cooling them in 1 loop should be perfectly OK, as long as you don't have an i7 920 @ 4+ Ghz and a GTX285 or something.

I would put one of the rads between CPU and GPU though, as you mentioned. Scientifically seen, the best heat transfer can be obtained when the temp difference between opposite sides is high. So running freshly cooled water (from the rad) over the GPU draws away more heat.
While that makes sense, this also counts for the rad itself. It's better to feed both rads with water as hot as possible (like right after a heated component) then to feed a rad with the cooled water from another rad. If you feed it cooled water from the other rad, rad 2 has much less cooling to do and therefore loses part of its efficiency.

Exactly and from what it looks like the single 120 rad is doing all the work and feeding to the second rad which then puts back into the loop. I would have put the single 120 rad in the back inbetween the cpu and gpu. And knowing him his i5 will be running 4+ghz.

 

[Kovoet]

Very neat cabling Joe and very clean. Good job

 

[johnnyfiive]

You sure its a good idea to blow your hot cpu water onto your GPU?

Search this topic on XtremeSystems forum Cory, its been covered AND tested many times. The overall water temp only changes depending on ambient temps, water part order doesn't necessarily change temps. What matters is the pump/res location and flow rate. I went for the cleanest loop orientation so naturally I went to the MOSFETs first, then the CPU. At the most, I'm guessing I'm taking a 1C hit on the CPU temps. I've already proven to myself the truth about orientation while doing load testing.

Anyone who disagrees, just read this:
http://www.xtremesystems.org/forums/showthread.php?t=225485

 

[Thrackan]

Search this topic on XtremeSystems forum Cory, its been covered AND tested many times. The overall water temp only changes depending on ambient temps, water part order doesn't necessarily change temps. What matters is the pump/res location and flow rate. I went for the cleanest loop orientation so naturally I went to the MOSFETs first, then the CPU. At the most, I'm guessing I'm taking a 1C hit on the CPU temps, nothing to really complain about when you look at how clean the loop is.

It's been proven that this is correct at the saturation point of the loop i.e. full load on all components all the time.

With real life situations, i.e. varying load, cooling heated components as soon as possible and running the coolest possible water on those components does make a difference though.
I wish I had some studies to prove my point here, but all I have is my previous experience.

 

[PP Mguire]

It's been proven that this is correct at the saturation point of the loop i.e. full load on all components all the time.

With real life situations, i.e. varying load, cooling heated components as soon as possible and running the coolest possible water on those components does make a difference though.
I wish I had some studies to prove my point here, but all I have is my previous experience.

Exactly, i have heard this since 2004, when i first started getting into water cooling and it was "the big thing".

That forum thread though dosent mention anything about what were talking about.

What we are talking about is, having your GPU directly after your CPU. The CPU putting off hot water and trying to cool the GPU with warm water off the cpu. No matter what flow rate you have or rads on the end (or beginning) of the loop your still putting warm water onto the GPU from the CPU. Thats why we where saying it probably would have been better to put the second smaller rad in the back of the case after the cpu to cool off some of the heat off the cpu and then to the GPU.

 

[johnnyfiive]

I get the point you guys are trying to make, but the temperature difference would be so little, its not even worth it. It really isn't, we are talking maybe 1C at the most. If someone wants to take the time and effort to prove the truth feel free, I sure as hell won't lol. I've done enough reading to know what suits my preference.

 

[xrealm20]

I get the point you guys are trying to make, but the temperature difference would be so little, its not even worth it. It really isn't, we are talking maybe 1C at the most. If someone wants to take the time and effort to prove the truth feel free, I sure as hell won't lol. I've done enough reading to know what suits my preference.

Even luke warm water from the CPU being used to cool the GPU is going to be far more effective than air - the simple reason is that water has a much higher specific energy than air. Therefore it can take on much more energy before changing a degree in temperature. Passing the water through a single radiator before the GPU will not make much difference (read less than 1 degree Celcius) under load.

Just my 2 cents.

 

[Chicken Patty]

I expiremented this on a old water setup I had and switching rads around didn't do anything. But it was a while ago, I probably don't even remember correctly.

 

[aCid888*]

I had a Feser 360 cooling 4870 Xfire & 945ES and I had great temps before my 4870's decided to die to spite my computers great performance.



On another note, my laptop at the moment:

I was doing some baking and I left my oregano and herb grinder next to my laptop by mistake.

 

[Fitseries3]

toying around today...

 

[johnnyfiive]

Even luke warm water from the CPU being used to cool the GPU is going to be far more effective than air - the simple reason is that water has a much higher specific energy than air. Therefore it can take on much more energy before changing a degree in temperature. Passing the water through a single radiator before the GPU will not make much difference (read less than 1 degree Celcius) under load.

Just my 2 cents.

Exactly, the difference is barely marginal.

 

[cadaveca]

Personally, I noticed a 10c difference in gpu temps adding a 120mm rad between cpu and gpu, rather than having it inline with a 360 rad.


Did it really make an difference in clocks or anything though?

NOPE.

 

[Chicken Patty]

Everybody is going to get different results. My advice is try it an keep it the way it works best for you.

 

[theorw]

Personally, I noticed a 10c difference in gpu temps adding a 120mm rad between cpu and gpu, rather than having it inline with a 360 rad.


Did it really make an difference in clocks or anything though?

NOPE.
http://forums.techpowerup.com/attachment.php?attachmentid=33003&stc=1&d=1265648454

wow!!!
Keep those 1900s some months more till i get W/C and then would u sell them to me?

 

[Velvet Wafer]

I had a Feser 360 cooling 4870 Xfire & 945ES and I had great temps before my 4870's decided to die to spite my computers great performance.



On another note, my laptop at the moment:

http://img.techpowerup.org/100208/P1040429MOD254.jpg


I was doing some baking and I left my oregano and herb grinder next to my laptop by mistake.

open up that box heres the police!

 

[PP Mguire]

 

[Thrackan]

What's with the huge ribbon?

 

[PP Mguire]

Oh it just goes to the little board evga gives you. When i wake up (havent been to sleep in over 24 hours) ill fix that and redo my cables cause i added the second 8800GTS and kinda tossed the cable in there for it as well as that extra board and ribbon cable.

I just wanted to see what i could do with my new toy without reading the manual.

 

[cadaveca]

wow!!!
Keep those 1900s some months more till i get W/C and then would u sell them to me?

Nope, not for sale, sry! I am collecting ATi vgas. Every once in a while I revist them...usually when a new cpu comes out.

 

[theorw]

Nope, not for sale, sry! I am collecting ATi vgas. Every once in a while I revist them...usually when a new cpu comes out.

Damn,and i just had a crush on them...beautiful babes...

 

[p_o_s_pc]

here is my PC ATM well more like the case.. My PC will be in it soon... I see a few places i missed. Looks like this summer when its warmer i will be doing it again

 

[(FIH) The Don]

it looks SO much better than the red you had before,

just put some tape around those red spots and respray them

did you remember the clear coating?

 

[p_o_s_pc]

it looks SO much better than the red you had before,

just put some tape around those red spots and respray them

did you remember the clear coating?

nope i didn't clear coat it. But this paint that i used doesn't scratch easy... well atleast it didn't on my other case.
Thanks I think it looks better too. I'm not worried about the little red spots right now. I'm sure by the time summer gets here it will need painted again anyways