• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

[OFFICIAL] Free ICD7 thermal compound for OCF members Results Thread

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Mr.Guvernment

Member
Joined
Feb 26, 2003
Just making this thread for all of those who will be receiving their free ICD7 thermal compound from this thread:

http://www.ocforums.com/showthread.php?t=538622


This will be the place it post your bench's, test, results from your IC Diamond 7 Carat thermal compound you received.


I am also sure if you went out and bought some you can tell us how well it did for you also!
 
GAH you got me all excited i thought this was the results of the people who would be receiving the stuff! either way if i get some ill post up results in this post!

RESERVED for results!

EDIT... Results (new as5 vs new ICD7 on my HTPC - e4300 @ stock clocks pad modded to 1.2v default VID, with speedstep enabled)

Paste ---- Coretemp ------TAT

AS5
Idle ------ 36/39*C ------- 39/40*C
Load ----- 69/69*C ------- 67/68*C

ICD7
Idle ------ 38/38*C ------- 37/40*C
Load ----- 68/68*C ------- 67/67*C

Ambient temp during all testing was 69.8-70.2*F... All said the ICD7 produced maybe 1*C difference depending on what program you go by. And to be honest (ikikuinthenutz brought this up in chatting w/ him) ICD7 is made up of well diamond powder.... the hardest substance on our planet... squish that between 2 soft pieces of copper and well ur gonna get lil itty bitty indentations, AS5 on the other hand is made of silver which is super soft and will be squished out by the coper to fill in any lil indentations there may be.

AS5 Idle
as5idle.JPG


ICD7 Idle
ic7idle.JPG


AS5 Load
as5load.JPG


ICD7 Load
ic7load.JPG


PS i used the application method stated (pea sized drop, but i heated the tube up with a desk lamp and the stuff flowed out so much better than when it is cold)

More testing will hopefully follow.
 
Last edited:
Sorry, no fancy graphs but I did drop a solid 2 degrees going from AS Ceramique to ICD7 on my most recent build (Core 2 Duo E6400 at stock when tested).
 
Last edited:
Ambient 23 C:

e6300 allendale @ 3.2 MHz, 1.384V
tuniq 120
37 C idle, 65 C load

HD 3850 256MB @ 749MHz/1017MHz, stock volts
stock cooler @ 70% fanspeed, ICD7 on GPU, stock thermal pads on memory
34 C idle, 57 C load
 
Last edited:
Reserved. Testing AS5, mx2, and IC diamond on a pentium4 550 (prescott core) under a TR120 Extreme.
 
Ambient 23 C:

e6300 allendale @ 3.2 MHz, 1.384V
tuniq 120
37 C idle, 65 C load

HD 3850 256MB @ 749MHz/1017MHz, stock volts
stock cooler @ 70% fanspeed, ICD7 on GPU, stock thermal pads on memory
34 C idle, 57 C load



what were you using before, do you recall the temps? or just went straight to icd7?
 
once we get a nice collection it can be consolidated into a nice purdy format / chart :)
 
I just went straight to ICD7, after having good results replacing cooked AS5 on my old x1900xt (~1 C lower, 24 hour breakin).

3850 idle temps were crappy before, but that was with stock TIM and auto fan @29% idle, which can't really compare to my 70% constant fan now.
 
reserved.

going to be testing again a the northwood in my sig w/ AS5, and a iBook G4 @ 1.2 Ghz, as well as its video card. and maybe ill try it out vs the stock thermal glue on my 5900 ultra
 
Just making this thread for all of those who will be receiving their free ICD7 thermal compound from this thread:

http://www.ocforums.com/showthread.php?t=538622


This will be the place it post your bench's, test, results from your IC Diamond 7 Carat thermal compound you received.


I am also sure if you went out and bought some you can tell us how well it did for you also!

Thanks Mr. Guverment for starting this thread

We would like to also thank all those who are participating with their test results in our Christmas Promo Giveaway If anybody has other ideas and/or recommendations for testing I would like to hear them.

There is always a lot of discussion on different methods of application. The criteria IC is recommending was drafted to simplify the compound application as much as possible to effectively provide optimal performance. The debate often centers on how much is a pea sized amount? Unfortunately this is subject to a final artful interpretation. In ICD7's case, more is better than less, by which we mean enough to get the whole IHS covered when the sink is mounted. You can jiggle into it if you are unsure by pulling the sink off and adding a little more if necessary and re clamping it, excess will just squeeze out.

we are investigating a couple of metered delivery systems that would take the guess work out of it.

This compound is very thick. when pushing the plunger, be patient and do not force it or you will break the plunger. It will come out at a slower pace than what most are used to. The reason the compound is formulated so densely is for performance and to minimize pump out to provide for long term stability. Thinner compounds (those with bulk loadings under 90%) while giving excellent performance on initial application are subject over time with thermal cycling causing an expansion and contraction of components to act like a "pump" and which then pushes the grease out of the joint and can cause a degradation of thermal performace of several degrees after only a couple months. The thickness/density is just the nature of the beast

Do not use the newly popular single line or other methods with ICD7 as spreading will be uneven. If everyone uses the recommended method it will settle into an even bond line which will give an optimal result and should provide more consistency between everybody's test results with fewer outliers.

Note your before and after temps, board and case temps, ambient temps if possible. It being winter be aware of alternate heat sources like baseboard hot air or water and when when your furnace is running etc. This ambient fluctuation can skew results. Be suspicious of of overly positive or overly negative results. Recheck the setup.

My comments from here on out will be somewhat limited so as not to influence results. It is what it is. I will answer questions but not want to enter into any debates.

syringes were mailed Monday so I expect this weekend or next week we will begin to see some results trickle in.



Happy Holidays to all:santa::bday:

Application notes for those participating the giveaway test promo.



IC Diamond is composed mostly of diamond powder 94% with other along with other bulk material(carbon black+), and as such is quite thick. Proper application is critical to optimum performance. Squeeze onto the center of the CPU an amount of IC Diamond compound about the size of a pea The amount applied should be enough to cover the entire IHS - the center of the CPU/IHS is where most heat is concentrated. Place the heatsink on the CPU and push down to spread IC Diamond over the CPU's surface. Clamp the heatsink and power up the PC.



Curing Time: IC Diamond requires minimal time to attain peak performance; in most cases, IC Diamond will reach peak performance after two hours of use. After 24 hours an addition 1/2 C drop. may be noted but realistically It suffices enough to call it two hours for optimal performance

Stability: IC Diamond is designed for stability - it will not bleed or separate in normal use


.:santa::bday:
 
Thanks Mr. Guverment for starting this thread

We would like to also thank all those who are participating with their test results in our Christmas Promo Giveaway If anybody has other ideas and/or recommendations for testing I would like to hear them.

There is always a lot of discussion on different methods of application. The criteria IC is recommending was drafted to simplify the compound application as much as possible to effectively provide optimal performance. The debate often centers on how much is a pea sized amount? Unfortunately this is subject to a final artful interpretation. In ICD7's case, more is better than less, by which we mean enough to get the whole IHS covered when the sink is mounted. You can jiggle into it if you are unsure by pulling the sink off and adding a little more if necessary and re clamping it, excess will just squeeze out.

we are investigating a couple of metered delivery systems that would take the guess work out of it.

This compound is very thick. when pushing the plunger, be patient and do not force it or you will break the plunger. It will come out at a slower pace than what most are used to. The reason the compound is formulated so densely is for performance and to minimize pump out to provide for long term stability. Thinner compounds (those with bulk loadings under 90%) while giving excellent performance on initial application are subject over time with thermal cycling causing an expansion and contraction of components to act like a "pump" and which then pushes the grease out of the joint and can cause a degradation of thermal performace of several degrees after only a couple months. The thickness/density is just the nature of the beast

Do not use the newly popular single line or other methods with ICD7 as spreading will be uneven. If everyone uses the recommended method it will settle into an even bond line which will give an optimal result and should provide more consistency between everybody's test results with fewer outliers.

Note your before and after temps, board and case temps, ambient temps if possible. It being winter be aware of alternate heat sources like baseboard hot air or water and when when your furnace is running etc. This ambient fluctuation can skew results. Be suspicious of of overly positive or overly negative results. Recheck the setup.

My comments from here on out will be somewhat limited so as not to influence results. It is what it is. I will answer questions but not want to enter into any debates.

syringes were mailed Monday so I expect this weekend or next week we will begin to see some results trickle in.



Happy Holidays to all:santa::bday:

Application notes for those participating the giveaway test promo.



IC Diamond is composed mostly of diamond powder 94% with other along with other bulk material(carbon black+), and as such is quite thick. Proper application is critical to optimum performance. Squeeze onto the center of the CPU an amount of IC Diamond compound about the size of a pea The amount applied should be enough to cover the entire IHS - the center of the CPU/IHS is where most heat is concentrated. Place the heatsink on the CPU and push down to spread IC Diamond over the CPU's surface. Clamp the heatsink and power up the PC.



Curing Time: IC Diamond requires minimal time to attain peak performance; in most cases, IC Diamond will reach peak performance after two hours of use. After 24 hours an addition 1/2 C drop. may be noted but realistically It suffices enough to call it two hours for optimal performance

Stability: IC Diamond is designed for stability - it will not bleed or separate in normal use


.:santa::bday:


Sweet dealio, i take it you guys shipped them out via USPS?
 
The compound arrived yesterday and I was pleased to see that the container is larger than my old tube of AS5.

comparison.jpg


I blew out what dust was in my case and cleaned it up for photo day. Then I rerouted the hoses, trimming out about 8 inches of needless length to prepare for the testing.

ugly.jpg


Test system:

Motherboard: Gigabyte GA7N400 Pro2 R1
CPU: AMD Athlon XP Barton 2500+ @ 3200 (2204Mhz)
Ram: 2x512 Kingston HyperX
Video: 6800 vanilla
Cooling: BipII 240, TDX, Maze4 GPU, MCP655. A fresh application of AS5 on CPU, GPU, & NB (Nearly clear water in lines.)

With ambient temp staying at a range of 22C~24C, it is idling at 32C and after 25 minutes of 95% CPU load with ./nbench, it peaked to 35C. I will let the AS5 settle out for a few days to a week and rerun the test, strip it all off and try again with 7^D.


In an attempt to recreate these tests/results in a Windows XP (Home SP2) environment, I have installed Speedfan and SuperPI. Speedfan reports these idle temps:
CPU: 34C
NB: 30C
Ambient: 23C
GPU Core: 40C
GPU ambient: 31C

Results with SuperPI remind me of ./nbench. 35C was the max load after 'bout half-hour.
I noticed Nebulous is using TOAST, and decided I'd try that. Two instances of TOAST gave a bit of increase. Fianlly, some results.

Idle-
CPU: 34C
Ambient: 23C
GPU Core: 40C
GPU Ambient: 31C

Load-
CPU: 37C
Ambient: 27C
GPU Core:41C
GPUAmbient:31C

I will begin trying to cook the GPU next. More to follow...
 
Last edited:
Back