jackparkyy

Go for the 4790k, it's better thermal dissipation means what ever cooling you go for will be more effective meaning the fans wont have to spin as fast leading to  a quitter PC and better overclocking potential.

Would he be interested in doing a straw poll to see how interested people are in a GPU bracket of AIO water coolers made by coolermaster because I think he maybe surprised.

Great build! I love that you are doing different things with the case especially the water cooling! Love the pics too. My fav is the one looking down at the three spinning fans!!!

I call horse shit.

Wow much water good build me crying no have good PC.
Looks extremely clean and sexy but those first pictures don't do it justice but then those later ones
What are you taking these pictures with?

Nice build man!!

very pretty

Theres a 16 year old at my high school that has triple sli 780s, previous to that he had quad 680s. Might i say he has an 8...ahem... Unreleased intel desktop processor. All for protein modeling which needs none of that. He invests and he's very good at it. Point being age never matters

Here's those Better photos and night shots I promised...

Hi everyone!
 
I have just finished my second personal build and my first liquidcooling build. It's been a work in progress for three moths now, not all the components are new, as my very first build was just over one year ago now and I didn't feel the need to replace all of the components so I just carried them over from my old system. The components that where carried over are the motherboard (asus maximus v formular), cpu (intel 3770k) and RAM (corsair XMS 2000mh/z DDR3). The new components are the WD red drive, samsung 128gb 840pro, NZXT HUE lighting and 2xtitans in sli (windforce editions).
This first picture was the first of the new components that I bought, I bought them both at the same time and they turned out to be the things that caused me the most trouble. I took them out of the package, ran them for an hour to see if they were defective and they both worked fine so as I bought the windforce editions, once I had done that, I took them out to replace the stock cooler to the windforce cooler.


This was before I had got my new case, ssd or hard drive so the next pic is of them both in my old system, I had spent a good month or so sleeving all the cables in my old system and I did go on to use the same psu in my new system as sleeving takes a very long time! I also decided to start sleeving during my GCSE exams which was probably a bad idea

As you can see my old HAF case wasn't the nicest looking case in the world. later that night I was playing Far Cry 3 and my system crashed, I hadn't overclocked the graphics cards yet so I wasn't expecting it. I tried to reboot my computer but it did nothing at all, I knew that something had to be shorting out and I thought it was probably my cables. after about an hour of trouble shooting, I finally tired taking out the bottom graphics card and the system turned on. I came to the conclusion one of my titans was dead, after a lot of swearing and hitting things, I looked at the warranty and it did cover replacing the cooler. I took it back to scan the next day, they confirmed it was defective and sent it off for a replacement, while I was there, I bought my new case which is a corsair 540d and the HUE lighting. The next pic is of the graphics card config I was running while I was waiting for the replacement and what it looked like when I transferred everything over to the new case.


As you can see, not the most optimal configuration. five weeks later after calling scan ten times and emailing gygabyte I finally got a replacement titan sent, I went to pick it up from scan and this was when I bought the 128gb samsung 840pro. I checked the replacment was working and then replaced the cooler  and installed it into my system, finished sleeving the extra pci-e power cables that I needed going from the 690 to titan sli and then installed it into my system and this was the configuration I was running for about two month before I decided to liquid cool a few weeks ago.

As I said before, I recently decided to liquidcool a few weeks ago but I didn't do it all in one go, I first bought the res (phobya 150 black nickel), rads (EK-CoolStream 40mm PE 360 & 240), tubing (Clear primochill) and the SP120 QE fans. I fitted what I could of the cooling at the time as I only had half the components.

I had to bend the right brive bag in order to fit the tripple rad in the front, meaning I moved the ssd to the back compartment of the case.


two weeks later, I had the money to get the rest of the components. I got the cpu block (EK Supremacy Nickel frosted plexi glass) , gpu blocks (EK-Titan-SE Acetal + Nickel clear plexi glass), Backplate (EK titan SE black backplate), Pump (Alphacool D5 variable speed), pump top (EK D5 X TOP), EK Blood red premix, 6x phobya Compression fittings, 4x phobya 90 degree fitings, 4x phobya 45 degree fittings and a few dust filters. As soon as I got back, I began the 4 hour fitting stage.
 
I began by fitting the pump top and then installing the pump into the case using good old velcro.

I then cracked on with the graphics card waterblocks, I took a few "glamour shots" of all my old air cooler for the titans while I was fitting the waterblocks.

Windforce cooler close ups.



Reference titan cooler close up.

GTX 690 close up.

days of air coolings past.

The waterblocks took about 40 mins each as I cut out all thermal pads for the 24 memory chips separately but other than that, installation was easy, nice instruction from EK!


 
I then installed the CPU block which was again very easy to do and finished all my tubing runs.

I unplugged everything but the pump from the psu, layed down some newspaper and started to fill the loop. All went well, now leaks but lots of air bubbles which took a very long time to get out as I had everything in the least optimal position for getting air bubbles out (triple rad was upside down and I was using the inlet as an outlet on the cpu block so no runs crossed over). I did eventually get all the bubbles out and then left it on overnight to leak test. woke up, nothing had leaked so I powered on the system and everything was A OK!




 
I managed to get my titans upto 1084mh/z and +400mh/z on the memory in game thanks to gpu boost 2.0 which gave me a 13% performance increase in heaven vs my air cooled overclock and this performance increase was about 10% in game. However the performance increase isn't the best bit of watercooling, it was the noise difference. I'm running all my fans at 800rpm and getting temps of 60C max on the top gpu and 55C max on the bottom gpu even with this massive overclock. I'm also able to run the pump at 50% speed and still get great flow, this all adds upto me hardly being able to hear it. It looks sick too, in the night shot above it comes off a lot darker then it actually is so I'll try and get a better night shots with a better camera soon. The 3770k I have isn't a very good overclocker, with the RAM running at 2000mh/z, I'm able to get it upto 4.2gh/z@1.185v but that's as high as it goes with temps at about 70C. The only issue I have is the 90 degree at the bottom doesn't look that great at the tubing comes out at a funny angle to it so I'm going to be replacing it with a 45 degree soon.
 
I'm sorry that the pics aren't of the best quality, I took them on my phone and wasn't planning to do a build log so I'll upload some "glamour shots" taken with a better camera to this thread soon!  Also, I'll upload some screen shot of overclocks, benchmark results and in game fps when I get time.

Looks nice wish i had all dem cards

looks awsome!

looks really good. Like the cables they look really nice in the build.

'Dat color scheme 'doe.

Ghetto SSD mount FTW

unless you have a deeper static pressure fan plan, that's about as quiet as it
will get. raising and lowering the pump speed is small differential (3°-5°) in
temperature maybe more on larger surface area radiators (capacity). high fin
per inch radiators really side with higher RPM fans. the narrow fins grab the
heat from the tubes, but unless the air-flow can carry it away, they just stew.
retest with 1600rpm+ and you'll notice the temperatures are a lot more livable
(unless those are SP120 QE). with standard (60mm) radiators the temps can
be edgy on low fan speed, but easily correctable with more fan speed
(SP120 PE). TTL noted on his LGA2011/Titan build thus overclocked
CPU/GPU that with low speed fans you'll see 70° CPU and 45°-50­° GPU and he
said to use the PE fans to get the lower temps.
 
i think the QE is just too slow. and i doubt another fan set (possibly GT-15 or
NF-F12) might have a couple of degrees of lower temps, but are as loud as the QE,
so the value trade is not there.
 
might want tot try without the cool fusion (MOSFET) in the loop. i'd found that the
trade-off of restriction to temps could be part of your issue, too. unless you are driving
1.8v+ on RAM and holding 2.0v on input regulation, the VRM just won't be a major
issue. one solution i used was the EK (yuck) formula block (less restriction) did have
a better impact than thought to open the flow-rate through the next connection.
just a thought.

Nice rig! You even have WD Red

This was very interesting! Thank you for sharing this information.

All benchmark was gathered from linus's benchmarks
 
Game:       Titan@996mh/z:   780ti@1178mh/z:   780@1025mh/z:   Titan(fps per clock):   780ti(fps per clock):  780(fps per clock):   fps per clock delta(Titan/780ti):   fps per clock delta(780/Titan):
Tombraider    102.5fps                112.3fps                    93.0fps                   0.103                                   0.095                         0.091                           + 7.7%                                               -11.65%
Bio Shock       118.1fps               123.8fps                  107.8fps                   0.119                                   0.105                         0.105                           +11.7%                                               -11.76%
Crysis 3            47.8fps                 50.6fps                    44.6fps                   0.048                                   0.043                         0.044                           +10.4%                                              -  8.33%
Far cry 3           51.4fps                 55.2fps                    50.3fps                   0.052                                  0.047                         0.049                           + 9.6%                                                -  5.77%
                                                                                                     Average:0.081                                  0.073                         0.072                          + 9.9%                                                -  9.38%
                                                         [For Titan to equal 780ti@1178mh/z, titan must be running @1062mh/z]    [For 780 to equal Titan@998mh/z, 780 must be running @1091mh/z]
What this shows:
*How much of an in game performance boost you get from overclocking.
*What clock speeds cards should be running at to get the same performance in game as each other.
*How many frames per second you get per 1mh/z.
*The comparison of fps per clock(how efficient a gpu is at squeezing every frame it can per mh/z)
 
Why this is useful:
*To help get a better understanding at why cards have certain stock clocks.
*Help give you a target overclock to aim for, for your gpu.
 
What this doesn't show:
*That any card is better than another.
*what overclocks you will get.
*That any titans can perform the same as a 780ti.
*That any 780 can perform the same as a Titan.
 
What I gathered from this data was that if you want the best performing card on the market then the 780ti is the way to go. However, If you recently bought a Titan then There's hope
for you yet. The data I have here shows that a Titan clocked at around 1.1gh/z will perform about the same as a 780ti clocked at 1.2gh/z. The way that this seems to work is that The
Titan is able to squeeze about 9% more frames per second out of each mh/z then the 780ti. This doesn't mean the Titan is any better than 780ti, it just means that overclocking will have
a greater influence on your in game performance than the same overclock would on the 780ti. The 780ti is able to overclock much better than most Titans but each mh/z that you are able to
overclock out of the Titan will make more of a difference in game. This really surprised me as the 780ti has all the cuda cores enabled, the theory I have is that the 2688 cores in the Titan was
the sweet spot it terms of in game performance but by enabling them all on the 780ti you are able to spread the heat throughout the GPU more efficient and therefore you are able
to not only increase stock speeds but also overclock extremely well on the 780ti. This would not only explain why Nvidia released the Titan with 2688 cores vs 2880
cores but this theory also explains why the 780ti uses up more power as it simple isn't as efficient per mh/z as the Titan or 780 and it's running at a high clock at stock.
 
I'm going to do the same thing with kyle from newegg results to see if this delta still exists at stock clocks but from Linus's benchmarks, this seems to be the case.

All benchmark was gathered from linus's benchmarks
 
Game:       Titan@996mh/z:   780ti@1178mh/z:   780@1025mh/z:   Titan(fps per clock):   780ti(fps per clock):  780(fps per clock):   fps per clock delta(Titan/780ti):   fps per clock delta(780/Titan):
Tombraider    102.5fps                112.3fps                    93.0fps                   0.103                                   0.095                         0.091                           + 7.7%                                               -11.65%
Bio Shock       118.1fps               123.8fps                  107.8fps                   0.119                                   0.105                         0.105                           +11.7%                                               -11.76%
Crysis 3            47.8fps                 50.6fps                    44.6fps                   0.048                                   0.043                         0.044                           +10.4%                                              -  8.33%
Far cry 3           51.4fps                 55.2fps                    50.3fps                   0.052                                  0.047                         0.049                           + 9.6%                                                -  5.77%
                                                                                                     Average:0.081                                  0.073                         0.072                          + 9.9%                                                -  9.38%
                                                         [For Titan to equal 780ti@1178mh/z, titan must be running @1062mh/z]    [For 780 to equal Titan@998mh/z, 780 must be running @1091mh/z]
What this shows:
*How much of an in game performance boost you get from overclocking.
*What clock speeds cards should be running at to get the same performance in game as each other.
*How many frames per second you get per 1mh/z.
*The comparison of fps per clock(how efficient a gpu is at squeezing every frame it can per mh/z)
 
Why this is useful:
*To help get a better understanding at why cards have certain stock clocks.
*Help give you a target overclock to aim for, for your gpu.
 
What this doesn't show:
*That any card is better than another.
*what overclocks you will get.
*That any titans can perform the same as a 780ti.
*That any 780 can perform the same as a Titan.
 
What I gathered from this data was that if you want the best performing card on the market then the 780ti is the way to go. However, If you recently bought a Titan then There's hope
for you yet. The data I have here shows that a Titan clocked at around 1.1gh/z will perform about the same as a 780ti clocked at 1.2gh/z. The way that this seems to work is that The
Titan is able to squeeze about 9% more frames per second out of each mh/z then the 780ti. This doesn't mean the Titan is any better than 780ti, it just means that overclocking will have
a greater influence on your in game performance than the same overclock would on the 780ti. The 780ti is able to overclock much better than most Titans but each mh/z that you are able to
overclock out of the Titan will make more of a difference in game. This really surprised me as the 780ti has all the cuda cores enabled, the theory I have is that the 2688 cores in the Titan was
the sweet spot it terms of in game performance but by enabling them all on the 780ti you are able to spread the heat throughout the GPU more efficient and therefore you are able
to not only increase stock speeds but also overclock extremely well on the 780ti. This would not only explain why Nvidia released the Titan with 2688 cores vs 2880
cores but this theory also explains why the 780ti uses up more power as it simple isn't as efficient per mh/z as the Titan or 780 and it's running at a high clock at stock.
 
I'm going to do the same thing with kyle from newegg results to see if this delta still exists at stock clocks but from Linus's benchmarks, this seems to be the case.

Brace yourselves gt 610 results coming tomorrow if there are any places left. Also may i ask why only the top 30? I know somw forums where there are more than 30 people with triple 770,680 or 7970s or dual titans. It a bit too much of a pay to win kind of thing.