Could I reach 4.5GHZ E6600 Stable on this?

J

joeswm8

New Member
Pump - http://www.petrastechshop.com/laddcwpeddto.html - $100
Radiator - http://www.petrastechshop.com/swmcqposerab.html - $45
CPU Block - http://www.petrastechshop.com/swapexpeunwa.html - $56
Reservoir - http://www.petrastechshop.com/swmchire.html - $20
20ft. Tubing - http://www.petrastechshop.com/7id5odmagepu.html - $10
10xClamps = http://www.petrastechshop.com/ststwohoclwb1.html - $7.50
Extra Barbs

Total=$251


First time writing up a watercooling circuit, i love the potential, maybe i might do it...not too expensive compared to some other things

Numbers:
OCed TDP = 70*(4500/2400)*((1.70-1.3)^2) = 270w of heat given off

(the block gets .10 C/W from swiftech's website)

270w x .10C/W = 27C + 28C(ambient, much lower in real life) = 55C load

Pump goes 2 GPM and ill be using .5 inch fittings on the block for move C/W and better flow. The radiator can handle 300+ watts and gets great C/W so no problem there.

Could I get these numbers and do this? Hows it look?
 
I highly doubt it. Generally speaking, the wall most people hit with the E6600 is around 3.4Ghz, and its not heat-related. Most people who reach 3.4Ghz have temps under 50C, but still cant overclock any higher. It depends greatly on the exact E6600 you get, since E6600's would normally be E6700's or X6800's, but arent stable enough to pass at the higher clock speeds.


You seem to be trying to figure this out mathematically, however most likely the bottleneck wont be the temps, but the reliability of both your motherboard and processor. You would be very lucky to get 3.6Ghz.
 
lucky to get 3.6? i got it the other day with ease...

that was on an evga 680i. We all know the 680i can reach FSBs of higher than 2000. We also know its amazingly reliable and an excellent OCer. So put that FSB on the E6600 with the necessary volts, and you have 4.5
 
The limit will be be the processor, not the motherboard. And reaching 2000? Easy? Stable? 500Mhz is big for any motherboard.
 
How about a DS3 Rev. 3.3? That seems really appealing because I wouldnt go SLI anytime soon.

Is the lowest BUS:RAM ratio 1:1 on the DS3? that would pose a problem
 
That server build isnt at my house, I helped build it with my uncle for his company. When we OCed it though, none of the software worked with windows server 2003, but the BIOS had 31 idle and the software said 50's on not full. This is with a Freezer Pro 7. We took it back down to 3.4 for full time use though. We did reach that FSB however.

http://www.evga.com/community/messageboard/topic.asp?TOPIC_ID=27296&SearchTerms=2000+FSB
this regards the FSB on the 680i. It has many holes, and I was hoping the A1 revision would fix that, but 1800 and 2000 have been reached. 1800 would represent a 4.0GHZ OC on the e6600 which would be the other optimal setting
 
Last edited:
Why would you OC a server, that doesn't make sense to me. Reliability is decreased with an OCed rig no matter how good a person is at it. Oh well, whatever, to each there own.
 
no not that one, thats not even mine lol

i am saying for a future build. how come no one is responding? dont you see the potential in the numbers?
 
Because we're skeptical. There's a fairly good chance that you could reach those FSB speeds, but not with the multiplier at stock. In my judgement, it should top at around 3.8. You never know, though until you try. Maybe it only reaches 3.0, maybe it reaches 4.5.
 
4.0 would be the goal with a 1800FSB.

look at this, this is interesting:
TGraph.gif


The fusion reports a C/W of half of what the apogee says, but they factor in a pressure drop line. What is pressure drop mean?

With a C/W of what they say, theroretically i could get those temps at 39C instead of 55C, but if that thermal resistance was really that good, then it would be the most famous cooling solution ever.

It cant possible be saying it gets around.03 C/W, thats totally insane.

What does that graph mean?
 
That's one hi OC...Why do you need it that high? Are you planing to do some serious folding er somethin?
 
just for fun to get some scores and to have high clocks. its fun i guess, i might not even do it but just playing with the idea is fun

i am really wondering about my previous thermal resistance post, anyone know?
 
Try liquid Nitrogen cooling like that one guy did with his E4600 OC to 4.6GHZ I think it was...It was in the thread "worlds highest overclock" er, something like that...
 
joeswm8 said:
4.0 would be the goal with a 1800FSB.

look at this, this is interesting:
TGraph.gif


The fusion reports a C/W of half of what the apogee says, but they factor in a pressure drop line. What is pressure drop mean?

With a C/W of what they say, theroretically i could get those temps at 39C instead of 55C, but if that thermal resistance was really that good, then it would be the most famous cooling solution ever.

It cant possible be saying it gets around.03 C/W, thats totally insane.

What does that graph mean?
Click to expand...


right here lol
 
Are you saying 50C with Intel TAT or another software program?

50C (if under load), is perfectly fine, and you can take it much higher. I would try and keep it under 65C however.
 
we dont need any more power on that server though, my uncle doesnt want to push it that far and is overwhelmingly happy with the performance.

can someone please answer that thermal resistance question? please just answer that one post
 
You must log in or register to reply here.
Back
Top