Test Setup & Measurement Methods
Why Do We Test Each Paste In Four Scenarios?
In choosing four test platforms, we incorporated feedback from our valued readers. For instance, you wanted to see us take the cooler mounting pressure into account. We're skipping the LN2-based testing and focusing on scenarios you'll encounter in the real world. For example, we're using popular pre-assembled water coolers that should ensure heat sink temperatures below 60 °C (140 °F), premium aftermarket air coolers with back plates that should demonstrate high mounting pressure, and a run-of-the-mill budget cooler with push-pin mounting (that'll give us limited pressure). Stock coolers like that let the CPU get above 60 °C/140 °F (AMD) and 80 °C/176 °F (Intel).
Depending on viscosity and composition, not all pastes are a good fit for every application, nor are they all well-suited for novices.
The systems we use for CPU measurements haven’t changed since the last time we covered thermal pastes. A quick check showed us that old pastes tested on the latest hardware yield similar results. Moreover, the sensors used in previous-gen CPUs yield more accurate user-facing data than those found in newer processors.
Exact temperatures are measured using a thermal diode under the heat spreader. This is extremely important and essential for an objective evaluation. Using Tcase instead of Tcore is the correct approach to take.
It is even a little exciting that these test components have lasted so long, even though our CPUs, graphics card, and motherboard were all sitting in storage. The only piece of hardware we swapped out for today's update was the power supply.
| Test System One: Closed-Loop Liquid Cooling | |
|---|---|
| Cooler | Corsair H80i |
| Fan | Original H80i fan, powered from an unregulated 7 V supply |
| CPU | AMD FX-8350 |
| Motherboard | Asus 990FX Sabertooth |
| Test System Two: Air Cooler with Back Plate (Screwed On) | |
|---|---|
| Cooler | be quiet! Shadow Rock |
| Fan | Original Shadow Rock fan, speed set to 70% |
| CPU | Intel Core 2 Quad Q6600 (Q0 Stepping) At 2.66 GHz |
| Motherboard | Gigabyte UP45-UD3LR |
| Test System Three: Intel Boxed Cooler (Mounted with Push Pins) | |
|---|---|
| Cooler | Intel Boxed Cooler |
| Fan | Original Intel Fan, Speed set to 80% |
| CPU | Intel Core 2 Duo E6850 |
| Motherboard | Gigabyte UP45-UD3LR |
Testing Thermal Paste On A Graphics Card
This is a somewhat sensitive topic, and for safety reasons we excluded electrically conductive or liquid metal pastes from our testing. Since GPUs don’t have a heat spreader, but allow the cooler’s sink to directly sit on the die, I didn't want anyone to risk a short circuit.
We also used an older graphics card, which was convenient to test with. Its cooler was mounted using four screws and its fan speed could be dialed in to a constant value. Furthermore, we figured that an older card would be more tolerant of higher temperatures. After all, we didn't want a cheap paste to destroy an expensive, current-gen board. Fortunately, the GPU die size and surface temperature are still comparable to modern mid-range cards.
| Test System Four: Graphics Card Test | |
|---|---|
| Cooler | Zalman GPU cooler |
| Fan | Original Zalman fan, speed set to 80% |
| CPU | AMD Radeon HD 4850 |
| Test Environment | Test System 1 (see above) |
Test Cycles, Test Duration & Settings
Because the digital temperature sensors built into modern CPUs only give us uncalibrated Tcore values, as mentioned, we use the old way of measuring die temperature with a thermal diode under the heat spreader. The processors in this story still use soldered-on spreaders, so this value should be fairly accurate. We'll report the difference between Tcase and the room temperature because there are always slight fluctuations in the (air-conditioned) room that could slightly distort the results for Tcase.
For the graphics card, we use the temperature as the GPU reported it. That number isn't influenced by minor changes in room temperature, so long as they stay within 2°C of our 22°C baseline.
| Test Environment | |
|---|---|
| Room Temperature | Approx. 22°C (relatively constant from 21.4 to 22.7°C) |
| CPU Test Results | Reported in °C as an average of temperature differences (Difference between the ambient temperature and the reading of the sensor under the heat spreader) |
| GPU Test Results | Output in °C for the GPU diode |
| Test Cycles CPU | 1x four-hour burn-in, followed by break of at least two hours 4x one-hour measurement, with one-hour breaks Total time at least 16 hours per thermal product and cooler |
| Test Cycles GPU | 1x four-hour burn-in, followed by break of at least two hours 2x one-hour measurement, with 30 minute breaks Total time at least 8.5 hours per thermal product |
Comparison Thermal Pastes
MORE: Best CPU Cooling
MORE: How To Choose A CPU Cooler
MORE: All Cooling Content
I love these kinds of articles and in-depth super tests!! Thank you so much for all your time, effort and hard work, I appreciate it. I'm sure that I'm going to enjoy reading it.
Um, do you guys still have a single page or "printable" view please?
Maybe it's because I've used Artic Silver 5 for so many years, but for me it's the best all-rounder compound there is. Plus it's very cheap. I like it more than the MX-2 and MX-4 compound siblings people usually recommends. But I have to say, the "diamond" compounds are indeed better it seems. I had my doubts, but no more with these tests.
Cheers!
I work with MX-4 almost exclusively. Yeah, it's not $30 a tube, but it's also not "very cheap," are you kidding me? "Very cheap," is the Elmer's glue you sniffed as a kid, repackaged as thermal paste.
I use MX4 specifically because it doesn't have a burn in period and because it lasts FOREVER.
No, it doesn't deteriorate. I've seen reports a decade after the fact showing less than three degrees celsius difference from when it was first applied.
So. Either you're biased because of ignorance, or both Artic's warranty and every long term test done before this has been lying. Gosh, lemme think which is more likely...
Now, is something like MX4 the best thermal paste out there? Of course not. But it IS way better than a lot of the market, super easy to apply and maintenance-free, and very reliable. If you're going to be a snob about your thermal pastes, at least be accurate about it.
Looking at your data Thermal grizzly Kryonaut wins as the best non-metal TIM except in low mounting pressure situations. it doesn't seem to matter as long as you have one of the decent pastes but its obvious there are a few to avoid like the Coolplast20 or Amasan T12 for example.
I'm using TIM since over 15 years, not only for Home PC's, but also in the industry. The major problem of this MX-4 are the long Burn-In time to get a better performance and the fast dry-out issue. As hotter a CPU or GPU works, as worse this grease performs (and is drying out). I does a lot of long-term runs with different products and especially this older products (not only from Arctic) were showing this typical behavior.
If you prefer MX-4, why not? Use it. But please accept, that a test of different products over 4 years can show at the end a completely different picture. :)
I get a lot of hardware (mostly VGA) with MX2- or MX-4 as replacement of the original TIM from other reviewers in rotation. And I have every time to replace this replacement with better (or original) products to get the original performance back. MX-2 on a VGA card is pure pain. Simply try one time another, better products and you will be surprised.
@JamesSneed
I have to take, what's in Germany on the market. All pastes were retail and not sponsored samples from the manufacturer. It was my idea to do this under real conditions. But I think it is possible to organize some stuff also from the US or Asian market.