Innovation Cooling IC Graphite Thermal Pad Review
Never Buy Thermal Paste Again?
“From that fateful day when stinking bits of slime first crawled from the sea and shouted to the cold stars, ‘I am man’, our greatest dread has always been the fear of using the incorrect thermal paste. But tonight, we shall hurl the gauntlet of science into the frightful face of thermal degradation itself.”
I ask that you forgive my overly dramatic introduction, and my paraphrasing of one of the greatest scenes in cinematic history. I simply feel that too many new PC builders – and even more experienced enthusiasts – may have been spending too much time stressing over their choices of thermal interface material.
“Are you mad?!”
Over the last several years, both CPU efficiency and thermal paste technology have improved. In this time, many tech publications have presented thermal interface material testing charts. From any of these it has become pretty obvious that there is generally not a vast temperature distance between the best pastes on the market and some very budget-friendly offerings.
The advanced thermal management and self-boosting features of modern processors mean that, unless someone is overclocking their CPU to the hairy edge of voltage sanity, you could use just about any recent thermal paste and a decent cooler and not get in range of temperatures that would hold back performance.
Some of you may now be asking, “If that’s the case, then why even do this test?” and you would be justified in doing so. To that, I would just point to questions I’ve been asked regarding my choice to use Innovation Cooling’s ID Graphite Thermal Pads for testing purposes in my reviews of CPU coolers.
Putting the IC Graphite Thermal Pad to the Test
While I’ve previously expressed my thoughts that the IC Graphite Pads eliminate the variable of paste application consistency in testing, I have also heard concerns that their thermal performance was far enough behind traditional pastes that they couldn’t be relied on as a good benchmark. There were also questions about the reusability of these thermal pads, and if there is performance degradation after repeated uses.
Here I will be honest in that I had never personally tested thermal pads against standard thermal compounds, or to see if there was degradation after usage, so I was unable say for certain that these concerns were without merit. At last, I decided to do a comparison myself, in the most controlled and equalized testing that I could manage without the use of a climate-controlled laboratory setting.
I also resolved to test several traditional thermal compounds in the process, which should give a fairly accurate benchmark of where the IC Graphite pads fall in relation to the rest of the market. It would also allow me to provide my thoughts on these pads, as well as some other popular thermal compounds.
I wished to insure that the cooling used during the test would not be a limiting factor, so I put together a closed loop using an Optimus Foundation full copper CPU block and a 360 EK Performance Edition radiator.
The Thermal Interface-off
Editor’s Note: In the following portion of today’s review, Kent places himself at the mercy of internet critics as he has taken close-up photos of his thermal paste application. I hope he is at peace with this. Also, please note that results from thermal testing are peppered throughout this overview – with charted results in their own section below this because I was attempting to match the formatting of typical reviews.
Innovation Cooling ID Graphite Thermal Pad
The application of these pads is pretty straightforward as it does not require even spreading or guessing the correct amount to apply. Just take the pad and place it on the CPU, then attach your cooler.
I have had some issues in the past where the pad would move about when attaching one of the more finicky air coolers I’ve tested, but I had no such issues with either pad tested with the waterblock. The new pad was a previously unopened pack, and the old was the very same pad I have used in all of my CPU cooler testing since I switched to an AMD system.
My estimate is that this pad has been used, then stored, then used again at least 20 times. To my surprise, the old pad actually outperformed the new pad, albeit it by a very small range (.3° C max), which is honestly probably within margin of error for the test. The two thermal pads were the worst performers of the group tested, but the difference from the worst to the best was under 4.5° C.
I will also note that if you are using an AMD AM4 or earlier socket CPU with the thermal pads, you will never have to worry about pulling your CPU out of the socket when changing a cooler.
Kingpin Cooling KPx
Since hitting the market a couple of years ago, the KPx has been the default go-to compound for extreme overclockers. I have been using it for a bit over a year for all systems I’ve built with the exception of my personal, delidded Core i9-9900K (as that is direct die cooled it is much better to use a liquid metal coolant such as Thermal Grizzly Conductonaut).
Out of this group, the KPx is not only the hardest to apply in a smooth even layer, it was actually an annoying pain. It is very thick and sticky compared to all the other compounds. This also made it the most difficult to clean of the group as well. Still, it was the best performing of the compounds (SPOILER ALERT – Ed.), with it and the Thermal Grizzly Kryonaut beating out all the rest by a reasonably healthy margin.
Thermal Grizzly Kryonaut
Thermal Grizzly really seemed to change the playing field when they hit the scene just a few years ago. Kryonaut has been their top performing, non-conductive compound for some time and was the standard for extreme overclocking until the KPx (and the new Kryonaut Extreme) came out.
It was my thermal paste of choice until KPx, and after the results of this test, I may switch back. The Kryonaut was only beaten by the KPx in this test by one tenth of a degree Celsius, which is certainly within margin of error. Add to that the fact that the Kryonaut was much easier to apply and clean up than the KPx.
In the entire industry of CPU cooling, there may not be another company with a reputation as sterling as that of Noctua. While I doubt that Noctua actually manufactures their own thermal compounds (most likely a chemical company contracts to their specifications), the NT-H1 is certainly worthy of the Noctua name and reputation.
While it wasn’t the best performing compound (it finished third in the group) its performance was still very good and it was by far the easiest compound to apply and clean up.
Arctic Silver Ceramique 2
Ten years ago if you asked any PC building enthusiasts what compound to use, their answer would most likely have been a product from Arctic Silver, with both the Arctic Silver 5 and the Ceramique 2 being regarded as two of the very best on the market. Fast forwarding to 2022 and several other companies seem to have pushed past Arctic Silver on the recommendations list.
The Ceramique 2 was almost as difficult to apply evenly as the KPx, but I was able to achieve a more even and smooth layer with the Ceramique in the end. The Ceramique was also just as sticky as the KPx, but slightly thinner and almost as difficult to clean. The thermal performance of the Ceramique 2 was very slightly behind the NT-H1 (but within margin of error for the test).
Scythe Thermal Grease
This standard thermal compound comes with every Scythe cooler. I have purchased or reviewed several Scythe coolers over the last few years, so I have several tubes of it, and I just threw it in to see how a very generic thermal compound measures up.
The Scythe thermal grease was almost as easy to apply and clean as the Noctua, but did not perform as well. Of the five compounds I tested, this had the worst performance, though it did still beat out the ID Graphite thermal pads.
For testing, I ran the OCCT small data set, extreme, stable load stress test for 30 minutes on an open test bed. I found that with the overclock and voltage settings I had applied to an AMD Ryzen 7 3800x, this test put a consistent thermal package load of between 128 and 129 watts on the CPU.
In addition to CPU package temperature, I also monitored the coolant temperature, as that would give an indication of which material was transferring the heat away from the CPU with more efficiency. A higher coolant temperature would indicate better transfer. This theory was confirmed by the results, as the compounds with the lowest CPU temperatures also had the highest coolant temperature.
Test System Specs:
- AMD Ryzen 7 3800x [@4.3 Ghz, 1.35 volts (1.275 Vdroop under load)]
- Optimus Foundation AM4 Waterblock
- EK Cool Stream PE 360 Radiator
- Barrow Boxfish Reservoir with coolant temp sensor and display
- Swiftech MCP50x water pump (@2500 rpm)
- Three EK-X3M 120 fans (@ 1500rpm)
- ASRock X570M Pro4 Motherboard
- 16 GB (2×8) G Skill Trident Z 3333 (@3600) Memory
- Zotac GTX 1650
- Sabrent Rocket 4 500 GB NVMe SSD
All tests conducted at a controlled ambient temperature of 24.5° C.
One of the compounds tested (Ceramique 2) required a minimum 25 hour cure time, with several heat cycles over that span. To put all the compounds on equal footing, all the compounds tested were allowed to cure for 25 hours, and Cinebench R23 was run at least 7 times at intervals over the 25 hours.
As I previously mentioned, while these tests were not conducted in a controlled laboratory environment, I took all possible steps to insure equal conditions during the tests, including monitoring and maintaining 24.5° C at the bench for the duration of the test.
In the end, what have I learned from all this? I expected around a 5 degree difference from the best performer to the worst, and the difference was actually less than that at 4.4° C. I was able to confirm that there is no drawback to using the Innovation Cooling ID Graphite Thermal Pads for testing.
While they do not perform quite as well as a standard compound, the difference is very small. They also allow for complete consistency from test to test without concern if compound applications introduced a variable. I was also happy to see that multiple uses of the pad did not seem to harm the performance. For most purposes, I would still recommend using a standard thermal compound, but you certainly could use one of these graphite thermal pads without concern.
If you are one of those mad scientists who like to move your power sliders all the way to the right and push your PC to the limits, then I would certainly go with the best-performing compound you can get, and not use something like these pads. For certain use cases, like needing a consistent and reliable base for testing when you’re changing coolers often, then these pads make a lot of sense.