I7 3770K delidded Coollaboratory Liquid Ultra @1,54 v #Batch3226C194 Block: Aquacomputer cuplex kryos -laing DDC 18w -Xspc RS360 BLACK Triple Radiator 3x Phobya G-Silent 12 2000 rpm

 
 
Corsair H100 Two AP-15's exhaust
PLL Enabled
Phase Control Extreme
LLC Extreme
Gigabyte UD5H

Can anyone here with a 3930K/3960X post what temperatures, voltages, and clock speed they have during LinX with the latest drivers.
 
I've got a full WC loop dedicated to my CPU and I still run into a thermal barrier.

eternitykh

feniks! delid that baby! it's so worth it! but omg please be careful when doing it, lol. good luck and we'll be waiting for your proper club submission :P

i'll do my submission tomorrow when im bored.

 
thanks buddy
 
Deliding in plans for this Saturday!

Holy $hit 1.5v on an Ivy!!!

3770K 5.00GHz at 1.512v (under load per cpu-z)
Batch: #3224B210
Cooling: custom water, Liquid Ultra TIM under IHS and on the IHS
 

Uploaded with ImageShack.us

eternitykh

D: getting a bit hot there feniks, 104

phew, yeah hot in here
was way worse before ... 105C at 4.7GHz under load ... before I delided it
 
... if only my room ambient was lower, heating out of control, had to use room AC to beat it down from 28C to 24C, not fun, wish I could just turn the heating off and enjoy 13C ambient for some benching

redeemer

How much did temps drop after the De-Lid??

 
for me the difference was 31C down under load at high clocks, with Liquid Ultra TIM used on the die (under IHS) and on the IHS (under cooling block) versus stock TIM under IHS and PK-3 under cooling block.
 
Another great "paste" is Liquid PRO (much harder to apply and even harder to remove).
 
IC Diamond also brings good results, but I would advice against using it on the exposed die (it scratches over time), it's still very good under the cooling block though (mind that it will scratch the IHS though and possibly erase the writings on it over time).
 
The highest I went with my chip was 5.2GHz and SuperPI 32M run completed in 6m 28 secs, CPU-Z validation here (mind the insane vcore I used LOL!):

 
unfortunately it was still unstable for 3dmark physx portion to complete without crashing to desktop.
 
EDIT:
another validation, just for kicks, max speed I can get this chip to run (not stress tested that is) @ 5.38GHz:

redeemer

31 degree drop I am astonished, I wonder just how low the quality of Intel's stock TIM really is!

 
yeah, average drop in temps when using CL Liquid Pro or Ultra is around 25C on average ... even with AS5 on the die, people report 13C drop in temps, here this is the great source of information:
http://www.overclock.net/...lidded-ivy-bridge-club
 
as per stock intel TIM ... it's as good as old dried mustard or ketchup, just grey color, nothing thermally conductive about it LOL!

nookie420

feniks

redeemer

31 degree drop I am astonished, I wonder just how low the quality of Intel's stock TIM really is!

yeah, average drop in temps when using CL Liquid Pro or Ultra is around 25C on average ... even with AS5 on the die, people report 13C drop in temps, here this is the great source of information:
http://www.overclock.net/...lidded-ivy-bridge-club

as per stock intel TIM ... it's as good as old dried mustard or ketchup, just grey color, nothing thermally conductive about it LOL!

that's sad,  i hope to join this club some day.

What cpu cooling do you use Feniks?

AND HOLY MOTHER OF GOD 1.72v, you're brave.

 
using custom water cooling here, but after deliding even a closed loop water cooler or phantecs air cooler should do. anyways, the whole trick was in lowering the ambient temperature as low as possible.
 
I actually went as high as 1.90V trying to hit the round 5.4GHz, but my chip simply refuses to run at such speed no matter if I try 54 multi and 100 BCLK or 53 multi and higher BCLK.

eternitykh

careful with those voltages, feniks, gonna need subzero temps with them or else you'll kill your chip. if it doesn't kill it, it'll probably degrade from electromigration at a very accelerated rate.

that's why im not going above 1.55v xD

 
actually that's some kind of paradox with Ivy Bridge, they don't seem to degrade from high vcore at all. there is a bunch of people that have run them above 1.7 or 1.8v and nothing happened me included. I agree that LN2 is required to cool it down under stress at such voltage, but I don't stress test it there, that's the whole thing.
 
my chip actually degraded before I ever clocked it higher than 4.8GHz with safe 1.33v vcore ... it also carried some brown burn spot on the terminal side (under CPU) the moment I removed it from my EVGA Z77 FTW - it was used on that board for maybe 2 weeks and never clocked higher than 4.8GHz/1.33v, go figure ... that's the only reason I will be RMA'ing it under Intel Overclockers Protection Plan (burn spot caused by a faulty motherboard and possible degradation coming from that). will see what I get in my next round of Intel lottery ... now having a new MB and would like to start fresh with a new CPU too. evga's z77 ftw screwed me up badly (2 chips degraded, 2 hard drives died).
 
however all that needs to wait, my area was devastated by hurricane sandy, there is no power in most of buildings around here (mine and 2-3 more have it somehow), my primary ISP has been down since Monday (running a cellular hotspot now), nearly all stores closed, 99.9% of gas stations closed (no power) and the single ones open sell gas to police only, mass transit has been suspended in both NJ and NY since Monday, many roads blocked by floods and down trees ... it's like after a war around here ...

I was actually showing this to an online friend, but he told me about this 5.0ghz thread and said I should post it here.
 

5.55ghz on an ivy
 

 
 
http://valid.canardpc.com/show_oc.php?id=2596941

 
 

i dont have linx and dont plane on getting it but here is what i got.  hopefully its enough
 
5.4ghz cpu validation

http://valid.canardpc.com/2663970