Overclock.net › Forums › Intel › Intel CPUs › [Official] Delidded Club / Guide
New Posts  All Forums:Forum Nav:

[Official] Delidded Club / Guide - Page 1841

post #18401 of 33673
Newegg has it!
post #18402 of 33673
Thread Starter 
Quote:
Originally Posted by Molokou View Post

Hello y'all!!

Just delidded my 3770K yesterday night because I couldn't wait to do so while my Liquid Metal Pro/Ultra arrives =)
I wanted to join this club because it was all the info in here what made me decide to do it biggrin.gif
That's the hard part of this, so I achieved to have done the worst part by now and just need to wait a few (or should I say several?) weeks to get my order delivered tongue.gif



OCN name: Molokou
CPU: Intel core i7 - 3770K
on die-TIM: Arctic Cooling MX-4
ihs-TIM: Arctic Cooling MX-4
Mhz gained: 0
OC after delid: none so far
Temp drops: they actually got a little bit worse than before
CPU-Z validation of max OC: no OC yet

I thought seriously on not joining the club just yet 'cuz I haven't got any improvement just yet :guiltysmile: and it makes me feel ashamed tongue.gif
Just need to get the LMP/U to make this work worth the price (and a better thermal solution, I have my eyes settled on the Swiftech H220 but it continues out of stock mad.gif) rolleyes.gif

See ya around =P

You're in Buddy! No prob on the temps wait for CLU/P and show me then with the new cooling thumb.gif
post #18403 of 33673
...temps / core differentials on a delidded Ivy Snorkle.gif

...as some of you know, I have been playing around with both the delidded Ivy 3770K and the Sandy-E 3970X as part of the Siamese Twins build (...btw. after seeing the contraption with open boards and all the tubes and wires, a friend of mine spontaneously called it 'The Borg')....

...I recently went back an tried both 'bare die' and 'lapped IHS' (both with CL-U) on the Ivy...while bare die beat lapped IHS by a couple of degrees in temps, I'm running with IHS - for now - as I can switch out the water-blocks between the Ivy and the 3970X in just a few minutes the way it is arranged without draining the system.

I do think that I got the lapping and mounting right this time, noting the identical core temps in both 'current' and 'min' temps...the Cinebench run at 5.1GHz was with iGPU (no discreet GPU for that test) which tends to increase core temp differentials a bit under load...though I'm very happy with the 'max temps' for Cinebench at 5.1 giggles (under 70 C w/ambient at 20 C) ...I really do think that lapping the IHS thoroughly was well worth the effort because prior of doing so, even idle temps varied between cores by 4+ C. smile.gif

LEAD Technologies Inc. V1.01
Edited by Joa3d43 - 5/16/13 at 4:15pm
post #18404 of 33673
Haha ever since someone posted that video, I've been told that I'm a idiot for thinking a 30c temp difference from cores WASN'T due to a poor mount, but instead due to Intel and their bad paste. I'm really laughing so much, it's like trying to teach a bunch of ignorant kids:
http://www.youtube.com/all_comments?v=XXs0I5kuoX4&feature=em-comment_reply_received
post #18405 of 33673
My own 2cents.gif

PROGENITOR
(17 items)
 
Emulator Box
(17 items)
 
The contender
(15 items)
 
CPUMotherboardGraphicsGraphics
i7 4770k Gigabyte GA-Z87X-OC Gigabyte Windforce 780Ti 3072mb Gigabyte Windforce 780Ti 3072mb 
RAMHard DriveHard DriveOptical Drive
16GB Avexir Orange 2133mhz 1TB Crucial M550 2TB Seagate Barracuda 7200 LG BDR/DVDR 
CoolingCoolingOSMonitor
OCUK Fathom with HW Labs Black Ice 240 rad Mayhem's Gigabyte Orange 8.1 64 OEM Dell U2713HM 
KeyboardPowerCaseMouse
Ducky DK-9008 Shine III MX Brown Superflower Leadex 1000w Platinum Black NZXT Switch 810 Logitech G602 
Audio
SoundBlaster Z PCI-E 
CPUMotherboardGraphicsRAM
AMD FX 8320 Vishera ASUS M5A99X EVO R2.0 Sapphire R9 290 Tri-X 4096MB Avexir Blitz 2000mhz 8GB 
Hard DriveOptical DriveCoolingOS
1TB Hitachi Deskstar + 500GB Hitachi Deskstar 7... Samsung DVDR/CDR Noctua NH-U12P windows 7 Ultimate 64-bit 
MonitorPowerCase
AOC 22" LED 1920x1080 SuperFlower Golden Green HX 550W 80 Plus Gold Xigmatek Utguard 
  hide details  
Reply
PROGENITOR
(17 items)
 
Emulator Box
(17 items)
 
The contender
(15 items)
 
CPUMotherboardGraphicsGraphics
i7 4770k Gigabyte GA-Z87X-OC Gigabyte Windforce 780Ti 3072mb Gigabyte Windforce 780Ti 3072mb 
RAMHard DriveHard DriveOptical Drive
16GB Avexir Orange 2133mhz 1TB Crucial M550 2TB Seagate Barracuda 7200 LG BDR/DVDR 
CoolingCoolingOSMonitor
OCUK Fathom with HW Labs Black Ice 240 rad Mayhem's Gigabyte Orange 8.1 64 OEM Dell U2713HM 
KeyboardPowerCaseMouse
Ducky DK-9008 Shine III MX Brown Superflower Leadex 1000w Platinum Black NZXT Switch 810 Logitech G602 
Audio
SoundBlaster Z PCI-E 
CPUMotherboardGraphicsRAM
AMD FX 8320 Vishera ASUS M5A99X EVO R2.0 Sapphire R9 290 Tri-X 4096MB Avexir Blitz 2000mhz 8GB 
Hard DriveOptical DriveCoolingOS
1TB Hitachi Deskstar + 500GB Hitachi Deskstar 7... Samsung DVDR/CDR Noctua NH-U12P windows 7 Ultimate 64-bit 
MonitorPowerCase
AOC 22" LED 1920x1080 SuperFlower Golden Green HX 550W 80 Plus Gold Xigmatek Utguard 
  hide details  
Reply
post #18406 of 33673
Quote:
Originally Posted by Joa3d43 View Post

...temps / core differentials on a delidded Ivy Snorkle.gif

...as some of you know, I have been playing around with both the delidded Ivy 3770K and the Sandy-E 3970X as part of the Siamese Twins build (...btw. after seeing the contraption with open boards and all the tubes and wires, a friend of mine spontaneously called it 'The Borg')....

...I recently went back an tried both 'bare die' and 'lapped IHS' (both with CL-U) on the Ivy...while bare die beat lapped IHS by a couple of degrees in temps, I'm running with IHS - for now - as I can switch out the water-blocks between the Ivy and the 3970X in just a few minutes the way it is arranged without draining the system.

I do think that I got the lapping and mounting right this time, noting the identical core temps in both 'current' and 'min' temps...the Cinebench run at 5.1GHz was with iGPU (no discreet GPU for that test) which tends to increase core temp differentials a bit under load...though I'm very happy with the 'max temps' for Cinebench at 5.1 giggles (under 70 C w/ambient at 20 C) ...I really do think that lapping the IHS thoroughly was well worth the effort because prior of doing so, even idle temps varied between cores by 4+ C. smile.gif

LEAD Technologies Inc. V1.01

Fantastic result mate, top job that thumb.gif

Is this on water?
PROGENITOR
(17 items)
 
Emulator Box
(17 items)
 
The contender
(15 items)
 
CPUMotherboardGraphicsGraphics
i7 4770k Gigabyte GA-Z87X-OC Gigabyte Windforce 780Ti 3072mb Gigabyte Windforce 780Ti 3072mb 
RAMHard DriveHard DriveOptical Drive
16GB Avexir Orange 2133mhz 1TB Crucial M550 2TB Seagate Barracuda 7200 LG BDR/DVDR 
CoolingCoolingOSMonitor
OCUK Fathom with HW Labs Black Ice 240 rad Mayhem's Gigabyte Orange 8.1 64 OEM Dell U2713HM 
KeyboardPowerCaseMouse
Ducky DK-9008 Shine III MX Brown Superflower Leadex 1000w Platinum Black NZXT Switch 810 Logitech G602 
Audio
SoundBlaster Z PCI-E 
CPUMotherboardGraphicsRAM
AMD FX 8320 Vishera ASUS M5A99X EVO R2.0 Sapphire R9 290 Tri-X 4096MB Avexir Blitz 2000mhz 8GB 
Hard DriveOptical DriveCoolingOS
1TB Hitachi Deskstar + 500GB Hitachi Deskstar 7... Samsung DVDR/CDR Noctua NH-U12P windows 7 Ultimate 64-bit 
MonitorPowerCase
AOC 22" LED 1920x1080 SuperFlower Golden Green HX 550W 80 Plus Gold Xigmatek Utguard 
  hide details  
Reply
PROGENITOR
(17 items)
 
Emulator Box
(17 items)
 
The contender
(15 items)
 
CPUMotherboardGraphicsGraphics
i7 4770k Gigabyte GA-Z87X-OC Gigabyte Windforce 780Ti 3072mb Gigabyte Windforce 780Ti 3072mb 
RAMHard DriveHard DriveOptical Drive
16GB Avexir Orange 2133mhz 1TB Crucial M550 2TB Seagate Barracuda 7200 LG BDR/DVDR 
CoolingCoolingOSMonitor
OCUK Fathom with HW Labs Black Ice 240 rad Mayhem's Gigabyte Orange 8.1 64 OEM Dell U2713HM 
KeyboardPowerCaseMouse
Ducky DK-9008 Shine III MX Brown Superflower Leadex 1000w Platinum Black NZXT Switch 810 Logitech G602 
Audio
SoundBlaster Z PCI-E 
CPUMotherboardGraphicsRAM
AMD FX 8320 Vishera ASUS M5A99X EVO R2.0 Sapphire R9 290 Tri-X 4096MB Avexir Blitz 2000mhz 8GB 
Hard DriveOptical DriveCoolingOS
1TB Hitachi Deskstar + 500GB Hitachi Deskstar 7... Samsung DVDR/CDR Noctua NH-U12P windows 7 Ultimate 64-bit 
MonitorPowerCase
AOC 22" LED 1920x1080 SuperFlower Golden Green HX 550W 80 Plus Gold Xigmatek Utguard 
  hide details  
Reply
post #18407 of 33673
Yeah, the TIM Intel uses isn't that bad. It's just not as good as Liquid Metal and of course a lot worse than soldering.
Also mounting quality varies. My chip was very very very tight, so before I applied CLU to the Die, I've made a test with AS5, and the difference was negligible to not delidded (2-4°C). Only after appliying CLU I saw a noticeable difference of 11°C. And actually the latter was with resealing the gap with black silicone, while the former was the IHS just being hold in place by the socket's fixing mechanism.

People that notice 20+ difference probably have had a worse production/mounting quality.
post #18408 of 33673
Quote:
Originally Posted by spoonium View Post

Yeah, the TIM Intel uses isn't that bad. It's just not as good as Liquid Metal and of course a lot worse than soldering.
Also mounting quality varies. My chip was very very very tight, so before I applied CLU to the Die, I've made a test with AS5, and the difference was negligible to not delidded (2-4°C). Only after appliying CLU I saw a noticeable difference of 11°C. And actually the latter was with resealing the gap with black silicone, while the former was the IHS just being hold in place by the socket's fixing mechanism.

People that notice 20+ difference probably have had a worse production/mounting quality.

Well to be honest.
I can understand 20-40c (yup 40) drops in temps through the cores from the original ones with CLU and DIRECT DIE mount.
However...a loss of only 5c for putting normal paste on 2 CORES and then the other 2 CORES to have dropped 30c? Yeah sure - it was Intel's fault...
That said who knows - but from my time of being part of this club, OC'ing PC's and building PC's for the last decade, I haven't notice ONCE a core that had a 30c difference from its other core, without touching anything (ie not de-lidding etc).
post #18409 of 33673
Quote:
Originally Posted by Totally Dubbed View Post

Well to be honest.
I can understand 20-40c (yup 40) drops in temps through the cores from the original ones with CLU and DIRECT DIE mount.
However...a loss of only 5c for putting normal paste on 2 CORES and then the other 2 CORES to have dropped 30c? Yeah sure - it was Intel's fault...
That said who knows - but from my time of being part of this club, OC'ing PC's and building PC's for the last decade, I haven't notice ONCE a core that had a 30c difference from its other core, without touching anything (ie not de-lidding etc).
Yeah, me neither. The temp difference in that video before delidding is really extreme, and it could've been an error while mounting the CPU (not enough TIM, cooler not straight, etc) or indeed a production failure where the IHS doesn't have decent contact with the first cores. Although to be honest a user error is just so much more likely.
It's not entirely impossible though, I've heard reports where basically the whole inside of the PCB-IHS space was filled with termal paste.
post #18410 of 33673
Very valid points. Although if I ever had a chip like that, I would return it. But the guy thought, why waste time and delidded it. I wouldn't have ever done that. If the TIM application is done badly from the factory, what else could there be wrong with the chip?
I wouldn't have even thought twice about it.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › [Official] Delidded Club / Guide