Innovation Cooling's Diamond 7 TIM test results

[tastymannatees]

Hazaro OCF- waiting on thermals to link data

[tastymannatees]

Updated some of the thermals

[123bob]

OK, New machine here. This is El-machino-3, which is my workbench machine. It is in a Sonata III case with a G0 Q6600 clocked at 2.97 gig. It runs on a Gigabyte DS3L with a Zalman 9700 cooler. The mobo is in the vertical orientation, the 9700 horizontal. I ran coretemp and realtemp on this machine.

I tested 3 things here.
1.The first was the original ceramique mount with the 9700.
2. The second was D7 on the 9700.
3. The third was a D7 mount on a new TRUE which replaced the 9700.

Pics first, then data.

The original mount is Ceramique, spread pics below.




Film pressure test resulted in the following wimpy pressure.

Pretty krappy!


The following are the spread pics of the D7 on the 9700. Note I managed to miss the top a bit, I think due to the way the 9700 clamps down. Looks like a bit of squish and twist would have helped here. It may have thrown off my comparison results a little, although I think we can conclude the wimpy mount pressure has far more to do with it, based on other data.




Here's a pressure test pic of the new TRUE cooler. Much better pressure...




Test #1 Ceramique, Pea blob 9700 cooler
Mount age about 5 months old.

Idle:
Ambient 22.4C at case intake
CoreTemp – 42,40,36,39
RealTemp - 37,35,31,34

Load:
BOINC as load – weeks of time running
Ambient 22.4C at case intake
CoreTemp – 59,58,56,56
RealTemp - 54,52,50,51


Test #2 D7, Pea blob, 9700 cooler

Idle:
Ambient 22.2C at case intake
CoreTemp – 40,40,35,38
RealTemp - 35,35,30,33

Load:
BOINC as load – 45 min. time running
Ambient 22.2C at case intake
CoreTemp – 59,57,55,55
RealTemp - 54,52,50,51


Test #3 D7, Big Pea blob, TRUE cooler

Idle:
Ambient 23.0C at case intake
CoreTemp – 40,40,35,38
RealTemp - 35,35,30,33

Load:
BOINC as load – 45 min. time running
Ambient 23.0C at case intake
CoreTemp – 57,55,52,52
RealTemp - 51,50,47,47

Looks like for the direct 9700 comparison, the D7 won by close to 1 degree. It did much better on the TRUE mount, partly because it's a TRUE and partly because I was generous with the paste and had much better mount pressure.

I've also come to the conclusion that this case is somewhat performance limited due to crummy input air into the case. I'll have to through a good fan up front to improve it.....Pretty high deltas over ambient.

Regards,
Bob

[tastymannatees]

Looks like one of the better ones on both contact and pressure. Little high on the one edge- see how it looks on imaging. Bail_W is tops in contact and pressure so far and in line to win the prize although you may nudge him out of first place with that test

[G.Foyle]

My pressure tests were returned to me by Polish Post... those lazy bums have been on strike last week.
I will re-ship them asap.

[tastymannatees]

Informal survey

How many of you disabled the Intel fan PWM for your test? I am a total **** on the subject so any input would be helpful

I am always looking for reasons why people might not see a thermal result with the IC diamond paste with good progress on the contact side of it, but with some outliers outstanding I have been noodling over, this was a long shot idea I had and I am not sure if it has any merit or not

The temp of the diode controls the fan speed, so if the cpu is cooler the fan speed would slow down in response, in effect allowing the cpu temp to catch up temp wise to a baseline sink comparison. Make sense? I don't know if that actually is what happens But I would assume it would introduce some variability in the results

And how would resetting or calibrating the temps under hardware monitor vs real temp and others affect the PWM duty cycle?

I am guessing from the Intel docs ranges are stepped either 5 or 10C increments or is there more flexibility like 2 or 3 C?.

I suppose that you would only see that if you happened to be straddling ranges on the before and after result?

[tastymannatees]

We have been organizing results from our pressure testing into a series of articles for a final report of results the first of which has been posted here

http://overclockers.com/articles1514/index02.asp

[littleowl]

I am trying to remember all my tests and I don't think I turned it off but maybe one time. If I did then it would have been between the Q6600 GO & B3
and I beleave you said you couldn't use that data anyways.

[tastymannatees]

I am trying to remember all my tests and I don't think I turned it off but maybe one time. If I did then it would have been between the Q6600 GO & B3
and I beleave you said you couldn't use that data anyways.

I wonder if it would show a difference in temp and/or fan speed between enabled and disabled at load?

[littleowl]

I wonder if it would show a difference in temp and/or fan speed between enabled and disabled at load?

hum that is a good question I may check on the E6600 tonight if I have time.

[tastymannatees]

ACHHH!

We need incentive here - A free tube of IC Diamond 24 Carat for the participating users on this forum who currently have IC diamond installed on an aircooled system - I need test results run with the PWM enabled and disabled at load and at Idle.

Data required - Ambient temps at the time the test was taken. Fan speeds at both Idle and Load PWM enabled and disabled and CPU temps with PWM enabled and disabled.

After qualifying members post send me a PM with your mailing address and I will send you your paste

[64dragon]

wish i could help, since i have amd i'm not aware that i can help. my board does have a 4 pin fan connector for the cpu but i dont have a 4 pin pwm fan.

since my last post, i did receive the new retention mechanism for my heatsink but dont think it will fit even if i dremel some of it off i've not pulled my heatsink off so i'm not 100% positive that it wont fit but i'm pretty sure that it wont.

[littleowl]

Dang dragon I was so hoping that it would work for you!!!

ok so I did my little test and did every bit of it at 100% load and I am sad to say that it didn't make any difference.

ok first my cpu is a Q6600 B3 running on stock cooler
the mobo is an asus p5wdg2 ws pro
Room temps is 31.1

I had cpu Q-fan control (pwm) disabled
my temps with it off were
70 70 66 68

I enabled cpu Q-fan control (pwm)
and temps were about the same after 10 min of running
70 70 67 67

I also notices that cpu internal thermal control was on auto
so just for fun I disabled it to see if there would be any change
needless to say no change. still running same temps on all cores.

I hope that helps some Andrew!

[tastymannatees]

thanks, I still need more samples though.

I would not expect to see more than a degree or so. I wish you had noted your fan speeds as a check point

[D_A]

Bummer, no PWN fans here either.

[littleowl]

ok so I did some more testing for you today noting the fan speeds using speedfan 4.34
room temp is 30.6c
humidity is 77%
Barometric pressure is 30.07

how the funny stuff. I didn't see any change in temps other then my first core.
with Q-Fan enabled my temps were 70 70 67 68
fan speed was 2744 rpm

with Q-Fan disabled my cores were
71 70 67 68
fan speed was 2657 and jumped maybe 2 times to 2700

I found that very very interesting!!

[tastymannatees]

ok so I did some more testing for you today noting the fan speeds using speedfan 4.34
room temp is 30.6c
humidity is 77%
Barometric pressure is 30.07

how the funny stuff. I didn't see any change in temps other then my first core.
with Q-Fan enabled my temps were 70 70 67 68
fan speed was 2744 rpm

with Q-Fan disabled my cores were
71 70 67 68
fan speed was 2657 and jumped maybe 2 times to 2700

I found that very very interesting!!

By those temps and humidity I would think you live in a rain forest. It is interesting speeds changed disabled vs. enabled I would have thought the reverse.

In any event looks only to be about 1/2 cfm or less.

I think I am pushing on a string here with this type of test thinking it through.

The test that would tell me something would be the before baseline and after new app of IC diamond, but I think it may be a long shot. I may do it as a focus on another beta test group. Sometimes just knowing what a particular feature is'nt is just as valuable as knowing what it is.

Thanks for your extra effort much appreciated.

[Movieman]

Doing my bit..
2 tubes of IC7 just shipped from Petra's to one of our members in Salem,Oregon from me.
Ok, now you don't have to worry about the mortage!

[tastymannatees]

Doing my bit..
2 tubes of IC7 just shipped from Petra's to one of our members in Salem,Oregon from me.
Ok, now you don't have to worry about the mortage!

Thanks. Tube here, tube there all adds up. Don't expect a commission on two tubes though

[jgmeeks01]

I bought 2 tubes of IC7 from frozen cpu and im having a little trouble finding a good way to apply this to an OCZ Vendetta 2. Any tips from people who are actually using this stuff? I got it put on the best I could figure out, (ill try to post some pics later that I took) anyways the CPU is lapped but I didnt lap the cooler because it seems pretty flat and it seemed like it would be a huge pain to try and lap anyways.

What I did was try to fill in some of the gaps with a razor blade and then lightly scraped the excess off and applied 2 thin strips down the 2 middle aluminum bars. My temps actually went up a degree or so but the gap between cores is bigger than it was aswell....anyways im gonna reseat tonight and was looking for some ideas.

Thanks

[Jaco]

Most of are using the 'blob' method. pea-sized blob of IC D7 in the middle of the heatspreader.
A good mounting with enough pressure will spread out the paste evenly.

[tastymannatees]

Heatsink Mounting Pressure and Flatness vs Performance

Initial summary of pressure tests - others to follow

Many thanks to those on the forum that took the time and effort to test. much appreciated

[D_A]

I bought 2 tubes of IC7 from frozen cpu and im having a little trouble finding a good way to apply this to an OCZ Vendetta 2. Any tips from people who are actually using this stuff? I got it put on the best I could figure out, (ill try to post some pics later that I took) anyways the CPU is lapped but I didnt lap the cooler because it seems pretty flat and it seemed like it would be a huge pain to try and lap anyways.

What I did was try to fill in some of the gaps with a razor blade and then lightly scraped the excess off and applied 2 thin strips down the 2 middle aluminum bars. My temps actually went up a degree or so but the gap between cores is bigger than it was aswell....anyways im gonna reseat tonight and was looking for some ideas.

Thanks

What I did (yes I managed to get some here in Auz, though there's no local distributor ) on this Opty 170 was a hemispherical blob about 4mm across smack in the centre of the IHS, then carefully mount the HS directly down on top with no twisting and as evenly as possible.

I have a lapped HS and on my first mount I got 1-2C higher temps. I pulled the mount down and discovered that the CPU had stuck to the HS and they had both pulled out together. Thankfully no damage, but just be careful. On my first attempt I twisted the HS, thinking it would spread the TIM better but all I did was introduce air pockets. Don't do it. The considerable viscosity of this stuff has to be considered in your mounting technique.

My end result is a drop of 1-2c over the AS5 I was using before.

[ceemic]

Hi!
First sorry, I don't have time to read all the thread.

I would like to know, how IC7 compared Ceramique in subzero cooling (LN2/DICE)?
Liquid Metal Pro should be probably better than Ceramique @LN2.
Hopefully someone have used IC7 in extreme cooling.

Thanks in advance

[Deux]

Hi!
First sorry, I don't have time to read all the thread.

I would like to know, how IC7 compared Ceramique in subzero cooling (LN2/DICE)?
Liquid Metal Pro should be probably better than Ceramique @LN2.
Hopefully someone have used IC7 in extreme cooling.

Thanks in advance

I'm benching DICE tomorrow so I'll give it a quick test.