Innovation Cooling 7 Carat Diamond TIM Test results thread *NOW ACCEPTING RESULTS!!*

[Eldonko]

Yeah I found it is much thicker than other stuff as well, I guess that why you need to use so much. I got the exact same temps even after a few days to set, dunno if it is because of H20 cooling or what.

 

[Perineum]

Okay, I've used this on my overclocked BFG 8800GTOC (750 core/1900 shader/2100 mem)

I had to stop folding! OH NOES! :blarg:

Previous compound was whatever came on the Arctic Cooling Accelero Twin Turbo.

Application method was the small pea sized drop in the center of the die. At first the temps were not all that different. I "baked" the compound by running furmark for 3 hours on it's hottest settings. I then let it idle for an hour, then folded for an hour and then retested everything. Temps were astounding. I'm *really* impressed with this compound. All tests for measurements were done for approximately one hour to ensure stasis. The results below that have 2 numbers just means that as the system reached stasis EVGA Precision showed a wavering value for temperature. It would be 46c for 5 seconds then 47c for 5 seconds, etc.

Previous Compound
Ambient Temp: 20c
Idle Temp: 45c - default 40% fan
Folding Temp: 53c - about 86% fan *
Fur Mark Temp: 61c - 100% fan

IC7
Ambient Temp: 20c
Idle Temp: 44c - default 40% fan
Folding Temp: 46c / 47c - about 60% fan *
Fur Mark Temp: 57c / 58c - 100% fan


Now, a word on my setup....

My 8800GT was BIOS edited so that the fan is always on 40% and starts to ramp up at 45c and is 100% at 55c

I've included the "*" to point out that at those temps the fan speeds are not constant. They seem to waver about +/- 3% RPM which is likely the reason for the wavering temperatures.

So... 1c idle difference, at least a 3c or 4c load difference and while folding I got a 6c or 7c drop in temperature and that was even under slower fan speeds! :thumb:

I've triple checked the folding results and it's correct everytime. It just doesn't seem right :biggrin:

 

[ICD7]

Let me know if I have to make any corrections.

 

[ICD7]

Some edits coming over the next couple of days on the following report but it seemed quiet here so I thought I would add something till more results come in-

I am starting a compilation of all forum test data that has been acquired from the forum giveaways to date, the report here is what I am doing at OCN and includes data from Hardware Canucks. Hardware Canucks will get the same kind of report but expanded to include more results.


More credence should be given to larger sample sizes, smaller less so.

1-5 samples 1 or two tests can flip results either way so usually get thrown onto the miscellaneous group.

6-10 samples may start to indicate trends but can be heavily influenced by outliers and so are lightly weighted other than a general trend indicator

11-20 samples - Starting to develop more of a confidence in the trend direction.

20-30 samples - Confidence level improves to around 95% at this point.

What is different in this final report is I am combining results from all forums to date (with the exception of AMDZONE they reworked their site and deleted the thread before I had a chance to copy all data) and breaking out some of the individual compounds.

These are of course preliminary and more of a snapshot as data is being added every day and so is subject to revisions, changes and corrections as needed. All data are independent user results and IC treatment is limited to Averaging and plotting the charts. All data is publicly available for review at Overclockers.com,Overclock.net, Overclockers Austraila, Anandtech, xtremesystms, HardewareLogic and HardwareCanucks.

First up is the generic white compounds and unidentified pastes that shipped with the installed sink.

Generic white or stock paste- 26 users 7 forums

Next AS5 -125 users from 7 forums

Next MX2 - 36 users from 7 forums

Next OCZ Freeze -11 User results from 7 forums

Next TX-2 -3 User results from 7 forums, 3 results from review sites

Next ShinEtsuX23 -7 User results from 7 forums, 3 results from review sites

Next Ceramique -9 User results from 7 forums

Next Thermalright -6 User results from 7 forums

Next Misc -11 User results from 7 forums

Preliminary, some edits, corrections to be added Later

 

[Nademon]

Great results! I imagine we should start seeing more come in now as the 2nd and 3rd batches should be in the hands of the participants. Keep up the great work HWC!!!

 

[ICD7]

Okay, I've used this on my overclocked BFG 8800GTOC (750 core/1900 shader/2100 mem)


My 8800GT was BIOS edited so that the fan is always on 40% and starts to ramp up at 45c and is 100% at 55c

I've included the "*" to point out that at those temps the fan speeds are not constant. They seem to waver about +/- 3% RPM which is likely the reason for the wavering temperatures.

So... 1c idle difference, at least a 3c or 4c load difference and while folding I got a 6c or 7c drop in temperature and that was even under slower fan speeds! :thumb:

I've triple checked the folding results and it's correct every time. It just doesn't seem right :biggrin:

Interesting about the fan speeds. I tried to get people to post their fan speeds for the last 3 giveaway's for the data but the informal nature of the giveaway's it get's overlooked in the temperature hunt.

 

[Jacka]

System: Xeon 3110 @ 4.2Ghz and 1.44v
Cooling: Core Contact 4 heatsink
Stress and Temps: CoreTemp 0.99.4 & Orthos Small FFT Prio 7

Pre-IC7 Compound: AS5
Abient Temp: 20C Idle
Temp: 38C
Load Temp: 54C

IC7
Ambient Temp: 19C
Idle Temp: 35C
Load Temp: 52C

These stuff are thick! Warm up tube and apply more than the regular AS5, stress test overnight. Awsome cooling!

 

[hausner]

System: i920 @4Ghz 1.34v
Cooling: Enzotech Luna/Feser One X-360 TFC Xchanger Triple
Stress: LinX 0.5.8
Sensor:Real Temp 3.20 RC1
Pre-IC7 Compound: MX-2
Abient Temp: 23C Idle
Temp: 33C
Load Temp: 60C

IC7
Ambient Temp: 22C
Idle Temp: 32C
Load Temp: 57C


Sorry about the delay and thanks again for great product!

 

[sswilson]

Pre-IC7
Compound: MX2
Abient Temp: 21c
Idle Temp: 26/25/23/25
Load Temp: 62/62/62/62

IC7
Ambient Temp: 20c
Idle Temp: 23/22/21/22
Load Temp: 60/60/60/60

​

​

This is under water, Q6600 @ 3745 1.55 Vcore / MCR220-QP-Res / DD DDCPX-Pro / Apogee GT / YL 1350 X4.​

Testing procedures were:

• Fresh mount on both MX2 & IC7​
• Both applications were done using the "blob in the center" method​
• Both mounts were "burnt in" overnight with rosetta (100% on all cores, but mid-range temps)​
• Temp benchmarking done from a fresh restart after minimum 1 hour power down​
• Recorded temps are min/max as recorded by Real Temp​
• System loaded with 1 hour full priority OCCT CPU testing​
• System allowed to cool down for .5 hours after Load complete to allow for min temps​

Observations......​

• Initial tests (immediately after mount) were pretty well neck and neck. IC7 seemed to improve a bit overnight while MX2 stayed pretty much the same.
• Min/Max temps don't do IC7 justice. I'm going to see if I can compare the two OCCT logs, but my initial impressions are that the IC7 mount seemed to have an overall lower average temp under load. The MX2 seemed to consistantly stay at max temp under load, while the max temp for the IC7 is a peak rather than the average.​
• The mounting mechanism for my GT might be an issue here as well.... if IC7 performance is strongly influenced by pressure, the fact that my mount is always "maxed out" by the plastic spacers might be preventing a little extra pressure to improve temps.​

Overall I'm impressed. I found I had to take some pressure off of the IC7 mount to "wriggle it in" in order to get the best mount.​

Sorry it took so long for me to get these results in. As it stands, I ended up having to redo my WC loop to get some consistant results. ​

 

[Nademon]

Thanks SS! Very detailed results you've posted. I imagine Andrew will be interested in seeing your post. I've received a few PM's from participating members letting me know they are intending to post some results shortly, so there still should be plenty more data to collect.

Keep the results coming folks!