DIY Corsair H50 killer...?

[skinnee]

Why would a waterblock be different though?

No two spread patterns of TIM are alike either - just like snowflakes.

What's frustrating is you have good temps when you think you can do better and never really get them down as good as they were previously.

It's like messing with collimators where one side is not as accessible as the other. The output beam is never devoid of speckle once it gets touched.

Yup, TIM print are like snowflakes. I don't know why the Air coolers aren't showing much variation, for the most part I'm getting less than a degree variation between mounts. I'm either one of the worlds most consistent cooler mounters, there are other thermal/mechanical factors in Air Coolers that reduce variations, or the mounting systems for Air Coolers are just that good... someone smarter than me can chime in anytime.

 

[nitromullet]

AT has spawned a monster...

What started out as an attempt to build something similar to the H50 has gotten a little more involved...

Enzotech Sapphire Rev. A cpu block
XSPC RX120 rad
Swiftech MCR420 QP rad
Swiftech MCP655 (D5) pump
XSPC single bay reservoir
BitsPower 480 rad grill
8x Scythe S-Flex F 120mm fans

...now I need a decent video card and block...

With a closed case, max load temp 74C on core 0, most cores hover around 65-68C. Ambient ~20C.

Temps are pretty much in line with what I expected from this block:

 

[Rubycon]

That's a lot of volts to be at 4.2.

Is that what yours is currently running?

 

[nitromullet]

That's a lot of volts to be at 4.2.

Is that what yours is currently running?

Nope. I'm at 1.375, which surely has an impact on temps.

Skinnee's i7 used in his testing is a C0. Mine is a D0.

The processor I'm using for this test is my C0/C1 i7 920. I'm running it at 21x196 (4120MHz) at 1.46V

http://skinneelabs.com/i7-blocks-1.html

 

[Rubycon]

Nope. I'm at 1.375, which surely has an impact on temps.

Skinnee's i7 used in his testing is a C0. Mine is a D0.



http://skinneelabs.com/i7-blocks-1.html

Ouch. Yeah I settled at 1.325V at 4.4GHz air. The chip was pot tested (CO2) well north of 5GHz but those guys LOVE to tease me. Like hanging a canteen in front of a miner that's been trekking through the dessert for two days. 🙁

 

[Red Squirrel]

looking at all these watercooling setups makes me want to try it. Do these all use standard pex fittings or is it different? Been debating on if I want to spend 200 bucks to get a pex crimper for various plumbing projects around the house, but they're pricy, and I don't know how often I'd actually use it.

I'd probably use a chiller and dexcool though, just because. 😀

 

[Rubycon]

looking at all these watercooling setups makes me want to try it. Do these all use standard pex fittings or is it different? Been debating on if I want to spend 200 bucks to get a pex crimper for various plumbing projects around the house, but they're pricy, and I don't know how often I'd actually use it.

I'd probably use a chiller and dexcool though, just because. 😀

Use Parker prestolok fittings!

Barbs, clamps, and even compression fittings are mere child's play in comparison. That's what I would use IF I could water cool. 🙁

 

[nitromullet]

Ouch. Yeah I settled at 1.325V at 4.4GHz air. The chip was pot tested (CO2) well north of 5GHz but those guys LOVE to tease me. Like hanging a canteen in front of a miner that's been trekking through the dessert for two days. 🙁

that's a sweet OC on air at 1.325.

Thanks for the reminder actually, I've been meaning to drop the voltage a bit. I was running 1.375 vcore to test the wc loop because I knew it would be stable. Giving 1.325 a shot now.

 

[Gillbot]

looking at all these watercooling setups makes me want to try it. Do these all use standard pex fittings or is it different? Been debating on if I want to spend 200 bucks to get a pex crimper for various plumbing projects around the house, but they're pricy, and I don't know how often I'd actually use it.

I'd probably use a chiller and dexcool though, just because. 😀

No, typically standard G1/4 threaded barbs are used. No pex fittings, or crimp connections.

 

[Rubycon]

that's a sweet OC on at 1.325.

Thanks for the reminder actually, I've been meaning to drop the voltage a bit. I was running 1.375 vcore to test the wc loop because I knew it would be stable. Giving 1.325 a shot now.

Yes it is and I have 12GB Dominator GT running 7-7-7-18-1T at 1693 MHz on the 1366 strap! Memtest 4.0 shows almost 20GB/S for memory. The 4X Classified X58 is a beast of an overclocker. Finally got it up and running LinX stable at that speed. :awe:

 

[Yellowbeard]

Yes it is and I have 12GB Dominator GT running 7-7-7-18-1T at 1693 MHz on the 1366 strap! Memtest 4.0 shows almost 20GB/S for memory. The 4X Classified X58 is a beast of an overclocker. Finally got it up and running LinX stable at that speed. :awe:

Very cool!

 

[Rubycon]

Very cool!

Tell me about it!

The Dominator GT 1600 beats the Dominator 1866 (non GT) version with timings. I can get higher speed (MHz) out of the 1866 but not with the lower timings. To summarize the GT has more throughput. The red color along with the spine of the HD5970 and the all red PCI-E slots on the Classified look very racy if that makes sense! Now all they need to do is make Dominator GTs with LEDs. I had some DDR2 and DDR Corsair memory with LEDs. Those were much better than the Crucial ones which can cause you to have a seizure when they aren't burning out! 😀

 

[Anubis]

Tell me about it!

The Dominator GT 1600 beats the Dominator 1866 (non GT) version with timings. I can get higher speed (MHz) out of the 1866 but not with the lower timings. To summarize the GT has more throughput. The red color along with the spine of the HD5970 and the all red PCI-E slots on the Classified look very racy if that makes sense! Now all they need to do is make Dominator GTs with LEDs. I had some DDR2 and DDR Corsair memory with LEDs. Those were much better than the Crucial ones which can cause you to have a seizure when they aren't burning out! 😀

hahaha

i love the differing opinion on this, i cant stand the LEDs on anything, its a computer not a lamp

 

[Rubycon]

hahaha

i love the differing opinion on this, i cant stand the LEDs on anything, its a computer not a lamp

Well on the Corsair sticks the LEDs could easily be associated by actions of the PC. Once these were memorized one can tell quickly if something has gone amiss when their display isn't responding, for example.

Post LEDs are nice on motherboards.

I don't mind the LEDs at all - then again in the 70s I loved my lava lamps and oil "rain" lamps too! 😛

Things I do NOT LEDs on are TVs with logos that light up, same goes for notebooks/laptops. Cannot stand that crap. Oh and all blue LEDs on everything. Green for power, yellow for (then turbo can mean other things now like network traffic if the board has ethernet activity headers), and of course red for disk activity. When everything is blue I get the blues. I don't mind singing or playing them (one of my favorite genres) but it's depressing when ALL of the lights are blinking blue and you have to STARE at something to see what the hell is going on.