Would you try this? 3960X / 3970X Direct-Die Cooling

doyll

[H]ard|Gawd
Joined
Feb 4, 2012
Messages
1,284
No, not unless I had source of free or very low cost CPUs. I didn't watch video, but am guessing only a couple degrees difference if any at all with IHS removed. Not worth the risk to me and most everyone I know of. ;)
 
Joined
Apr 5, 2016
Messages
2,244
No, not unless I had source of free or very low cost CPUs. I didn't watch video, but am guessing only a couple degrees difference if any at all with IHS removed. Not worth the risk to me and most everyone I know of. ;)
Yep. He got results that were about stock. He suspects that he doesn't have good mounting pressure on it anymore since he had to buzzsaw the ILM to make clearance for the edges of the waterblock.
 

clockdogg

[H]ard|Gawd
Joined
Dec 12, 2007
Messages
1,079
Pretty lame. When you delid and go direct die cooling on TR3, I expect to see 5 nano-sized blocks... well.... 4 nano-sized and 1 I/O sized. Not one big monster block made for an IHS that no longer exists.
 

thesmokingman

Supreme [H]ardness
Joined
Nov 22, 2008
Messages
6,396
This "might" be worthwhile or have more importance if there were 4 more chiplets in there.
 

tunatime

Well...OK
Joined
Sep 15, 2011
Messages
4,144
Pointless they allrady run cool my 3960x at 4.45 all cores hits 65-70. I will say this though when you hit them with more v core the temps shoot up like crazy. If I leave pbo on my mb likes to feed this thing like 1.48 single load v core and 1.45 all core for a whooping 4.4 and 4.2 all core load. My manual clock is 4.45 at 1.4 if I back it down 4.3 1.325 v core it's a good 10c droop in temps....so either I have amazing cooling/ bad temps sensers on my CPU or that guy block sucks to be hitting 85c
 

thesmokingman

Supreme [H]ardness
Joined
Nov 22, 2008
Messages
6,396
Pointless they allrady run cool my 3960x at 4.45 all cores hits 65-70. I will say this though when you hit them with more v core the temps shoot up like crazy. If I leave pbo on my mb likes to feed this thing like 1.48 single load v core and 1.45 all core for a whooping 4.4 and 4.2 all core load. My manual clock is 4.45 at 1.4 if I back it down 4.3 1.325 v core it's a good 10c droop in temps....so either I have amazing cooling/ bad temps sensers on my CPU or that guy block sucks to be hitting 85c

Stock they are great and efficient, however things change when you go to an all core overclock. That's what derbauer was testing, a 4.5ghz all core oc. That level of overclock will make your loop cry.
 

somebrains

[H]ard|Gawd
Joined
Nov 10, 2013
Messages
1,403
No, I'd segment the workload to a dedicated machine and reduce drag on the main pc.
 

chithanh

Gawd
Joined
Oct 18, 2010
Messages
814
Seems like not enough pressure from the EK waterblock on the CPU? Why not use a model where the center protrudes more, like the Bykski?
 

M76

[H]F Junkie
Joined
Jun 12, 2012
Messages
11,466
Well, of course it works, the older ones remember direct die cooling was common until the early 2000s. There were of course a lot of chipped CPUs. AMD even added rubber pads to the sides against uneven mounting of coolers.
 

Gideon

2[H]4U
Joined
Apr 13, 2006
Messages
2,760
People are always scared of direct die cooling, yet it was done for years. Problem is removing the heat spreader will reduce the clamping pressure of the block unless you modify the mounting.
 
Joined
Apr 5, 2016
Messages
2,244
People are always scared of direct die cooling, yet it was done for years. Problem is removing the heat spreader will reduce the clamping pressure of the block unless you modify the mounting.
I'm scared of direct die cooling because when I hear that term, this is what I think it means:

| MOTHAFUCKIN' |
| JET OF WATER |
| DIRECTLY ONTO |
/________________\
--------------------------------
BARE-ASS CPU DIE
NO FUCKS GIVEN

When Derbo whipped out the waterblock I was a bit disappointed.
 

vegeta535

Supreme [H]ardness
Joined
Jul 19, 2013
Messages
5,054
People are always scared of direct die cooling, yet it was done for years. Problem is removing the heat spreader will reduce the clamping pressure of the block unless you modify the mounting.
Well it comes from the old AMD days where people cracked core on those bare dies. It was a nerve racking experience back then.
 
Top