E6700 Lapped IHS = 2-7C cooler under load

[eva2000]

Decided to lap my E6700 B1 IHS and it's the best lap job i've done so far compared to what I did on E6600s. E6600s didn't gain any clocks or temp reductions when lapped as the IHS was already flat.

But E6700 cpu when i started to lap only the 4 corners went to copper for a while so the IHS was definitely concave! See here the edges were first to go.

http://fileshosts.com/intel/E6700_photos/lapped/thumbnails/thumb_E6700_002.JPG (http://fileshosts.com/intel/E6700_photos/lapped/photos/E6700_002.html)

Not perfect as there's still some slight scratches but enough to produce this

http://fileshosts.com/intel/E6700_photos/lapped/thumbnails/thumb_E6700_003.JPG (http://fileshosts.com/intel/E6700_photos/lapped/photos/E6700_003.html) http://fileshosts.com/intel/E6700_photos/lapped/thumbnails/thumb_E6700_004.JPG (http://fileshosts.com/intel/E6700_photos/lapped/photos/E6700_004.html) http://fileshosts.com/intel/E6700_photos/lapped/thumbnails/thumb_E6700_005.JPG (http://fileshosts.com/intel/E6700_photos/lapped/photos/E6700_005.html)

So far prelim tests at 3600mhz at 1.394v show a 3-4C reduction in idle temps and 5-7C reduction in load temps but i forgot what the room temps where before lapping so who knows heh.

http://i4memory.com/reviewimages/cooling/cpu/Corsair/Hydrocool200EX_LGA775/html2/photos/hydrocool200ex_003.jpg (http://i4memory.com/reviewimages/cooling/cpu/Corsair/Hydrocool200EX_LGA775/html2/photos/hydrocool200ex_003.html)

E6700 @3600Mhz at 1.394v with Corsair Hydrocool 200EX 1/4" OD tubing

Before lapping
cpu idle = 37-40C
cpu load = 65-67C

After lapping
cpu idle = 34-36C
cpu load = 60-63C

http://fileshosts.com/intel/Intel975x/975xbx_results/E6700/HC200EX/gskill/pc6400phu2/17_18/1334/10x/533/360-360-4448_1.394-1.65-1.395-2.177/stressprime_load_tn.png (http://fileshosts.com/intel/Intel975x/975xbx_results/E6700/HC200EX/gskill/pc6400phu2/17_18/1334/10x/533/360-360-4448_1.394-1.65-1.395-2.177/stressprime_load.png)

E6700 @3734Mhz at 1.418v with Corsair Hydrocool 200EX 1/4" OD tubing

Before lapping
cpu idle = 36-39C
cpu load = 69C

After lapping
cpu idle = 36-39C
cpu load = 65-67C - 2-4C improvement

http://fileshosts.com/intel/Intel975x/975xbx_results/E6700/HC200EX/gskill/pc6400phu2/17_18/1334/10x/533/373-373-4448_1.418-1.65-1.395-2.177/stressprime_load_tn.png (http://fileshosts.com/intel/Intel975x/975xbx_results/E6700/HC200EX/gskill/pc6400phu2/17_18/1334/10x/533/373-373-4448_1.418-1.65-1.395-2.177/stressprime_load.png)

Now that the E6700 is lapped i'm about to pull off 4000mhz at 1.55v (1.6v bios) on Intel D975XBX rev304 1334 bios which just my puny Corsair Hydrocool 200EX 1/4" OD tubed unit. Before E6700 unlapped for 4004mhz took 1.65v bios for P5W DH DX on Swiftech G4 Storm/MCP655/Thermochill 120.3 (this DIY H20 averages 7C cooler temps under load than Hydrocool 200EX)

Click image for larger full screenshots...

Before lapping
- E6700 B1 on P5W DH with Swiftech G4 Storm/MCP655/Thermochill 120.3
@4004mhz at 1.65v bios set (1.62v)

Single 32M
http://fileshosts.com/intel/Asus/P5WDH/results/E6700_2/SwiftechG4/gskill/pc6400phu2/19_20/0801/10x/533/400-400-4448_1.65-1.3-1.65-2.35_std/superpi-32min_single_13min20s828ms_tn.png (http://fileshosts.com/intel/Asus/P5WDH/results/E6700_2/SwiftechG4/gskill/pc6400phu2/19_20/0801/10x/533/400-400-4448_1.65-1.3-1.65-2.35_std/superpi-32min_single_13min20s828ms.png)

After lapping
- E6700 B1 on Intel D975XBX rev304 with Corsair Hydrocool 200EX
@4000mhz at 1.6v bios set (1.55v)

http://fileshosts.com/intel/Intel975x/975xbx_results/E6700/HC200EX/gskill/pc6400phu2/17_18/1334/10x/533/400-400-4434-1.55-maxed-2.29/cpuz_validated_tn.png

http://fileshosts.com/intel/Intel975x/975xbx_results/E6700/HC200EX/gskill/pc6400phu2/17_18/1334/10x/533/400-400-4434-1.55-maxed-2.29/superpi-32m_single_13m11s016ms_grey_tn.png (http://fileshosts.com/intel/Intel975x/975xbx_results/E6700/HC200EX/gskill/pc6400phu2/17_18/1334/10x/533/400-400-4434-1.55-maxed-2.29/superpi-32m_single_13m11s016ms_grey.png)

Yes different timings used but just comparing vcore needed to hit 4Ghz

[before]

That's a nice work eva... as usual ;)

[El<(')>Maxi]

How much further does it overclock after the deck polish Eva?

[eva2000]

Hard to say as Badaxe board only has max 1.6v vcore option in bios = 1.55v vcore.

On P5W DH DX 1.7v vcore (1.66v) max.. but that board is covered in insulation mess from phase adventures right now heh

[Mr. Roboto]

wow, that looks really nice. There has been quite a few people lapping their conroes on here with great results. I'd love to try it, but my temps are great right now, so i'm not going to bother.

[enz660]

well done, im so jealous of you ppl clocking at over 3.6 while under 1.45vcore :bang head

nice lap job anyways :beer:

[Maviryk]

Now just imagine what would happen if you took a blowtorch to your conroe and remove the IHS! It would run even colder!

[enz660]

Now just imagine what would happen if you took a blowtorch to your conroe and remove the IHS! It would run even colder!

hell yeah!! watch that conroe melt off like butta!!! :beer:

[Epox4life]

I had about the same results using a vapochill.

http://www.ocforums.com/showpost.php?p=4718960&postcount=24

[Phrenetical]

nice,

[Stratcat]

5 - 7*C under H20 is a huge difference! :eek:

Undoubtably you did have a less than true IHS.

I'll be pulling my E6600 out to change mobs this weekend, & now that the prelim stability work on this CPU is done, I'll definitely do a "razor blade" test, to check my IHS.

Thanks for posting this. :beer:

An IHS warped enough to cause a 7*C change in temps under H20 seems bordering on defective, imho.

Unless there's some truth to the "concave TIM puddle" theory from a while back.

~ Strat ~

[bing]

If you have small >10mm thick flat glass (not mirror), then put some goop (heck, even works better with tomatto sauce) :D on the ihs and push the thick glass on it.

It will shows you how flat it is by looking at that red goop spreads when squezzed between the glass and IHS, and this method also can be used for your HSF base to see it's flatness.

Assuming you pressed the glass evenly on the IHS, it will shows you visually how is that tomatto sauce thickness spreaded on the surface, deep red means thicker than lighter red. :D

No, I'm not joking, that sauce is better than AS5 goop since it is more difficult to differenciate the color spreading as the AS5 goop silver color is almost similiar to the IHS surface.

UPDATE :
There is feedback from Spawn-Inc with pics HERE (http://www.ocforums.com/showthread.php?t=529941) that printer ink is better than ketchup ! Thanks & credit goes to Spawn-Inc ! :thup:

Update : Another great thread by Spawn_Inc HERE (http://www.ocforums.com/showthread.php?t=564355) !

[eva2000]

I had about the same results using a vapochill.

http://www.ocforums.com/showpost.php?p=4718960&postcount=24
yeah this lap job has done wonders for my E6700

thanks bing for the info :)

[bing]

Just make sure you wipe that ketchup clean, or you'll get a rotten, smelly and fungus infected rig !..yikes... LOL :D

[eva2000]

heh

Some PCmark05 fun now :)

http://service.futuremark.com/compare?pcm05=495469

http://fileshosts.com/intel/Intel975x/975xbx_results/E6700/HC200EX/gskill/pc6400phu2/17_18/1334/10x/667/384-480-4438-1.55-maxed-2.29/7800GTX_SLI/3dmark06_473-1350_10558_grey_tn.png (http://fileshosts.com/intel/Intel975x/975xbx_results/E6700/HC200EX/gskill/pc6400phu2/17_18/1334/10x/667/384-480-4438-1.55-maxed-2.29/7800GTX_SLI/3dmark06_473-1350_10558.png)

[xTrEmEoVrClOcKr]

Wish I had a better hdd setup, that score is smokin :D.

I can run PCMark05 up to 3900mhz, requires a lot of voltage though.

[greenmaji]

eva2000.. what kind of HD setup did you have for that run?
I remember your I-Ram runs being a bit wild :p