Threadripper Cooling Comparison - Noctua CPU Cooler Differences
### Article: Noctua CPU Coolers Comparison for AMD’s TR4 Threadripper Socket
#### Introduction
Hey guys, Jarrod here, and today we’re going to check out three different CPU coolers from Noctua that support AMD's TR4 Threadripper socket. We’ll be comparing their performance in terms of temperatures and overall system volume at both stock clock speeds and while overclocked.
I originally bought the largest model available, the NH-U14s, for my 16-core AMD Ryzen 1950X CPU. However, I wondered if it’s possible to save a little money by opting for a smaller and cheaper cooler that might still deliver similar performance. So, I reached out to Noctua and asked if they’d be willing to provide their other two Threadripper coolers—the NH-U12s and NH-U9—that also cover the full TR4 socket. Thankfully, they agreed, and we’re here with all three coolers for a comprehensive comparison.
#### The Main Differences Between the Coolers
The first obvious factor we’ll consider is the size of the heatsink. The NH-U14s that I purchased initially is the largest among the three, followed by the NH-U12s, and then the NH-U9 being the smallest.
One thing worth noting is that while the NH-U14s covers the first PCIe slot on my ASUS ROG Zenith II Extreme motherboard by default, it does have the ability to lay off-set the heatsink by three or six millimeters if needed. Fortunately, I didn’t encounter any coverage issues with the smaller coolers during testing.
Here’s a quick breakdown of the specifications:
- **NH-U14s**: Comes with a 140mm fan and weighs 1,030 grams.
- **NH-U12s**: Features a 120mm fan and weighs 870 grams.
- **NH-U9**: Includes two small 92mm fans and actually weighs slightly more at 895 grams, owing to the extra fan.
What makes all these coolers special is that they provide full coverage of the Threadripper IHS (Integrated Heat Spreader), which should result in better cooling compared to other coolers that use the ASA tech bracket.
#### Testing Methodology
For this comparison, I’ll be testing all three Noctua coolers with my AMD 16-core Ryzen 1950X Threadripper CPU at both idle and under full load, using stock clock speeds and overclocked settings.
To simulate a realistic workload, I chose Cinebench R23, as research suggests it’s more representative of real-world performance compared to something like Prime95. All tests were conducted in an ambient room temperature of 20 degrees Celsius, with temperatures allowed to stabilize for half an hour before each test.
#### Temperature Performance
Let’s start by discussing the idle temperatures. As expected, the larger NH-U14s cooler performed the best at idle, both at stock speeds and when overclocked. However, there was only a small difference of one to two degrees between the coolers during idle.
Under full load at stock speeds, the differences became slightly more noticeable, but they were still fairly close. When we pushed the CPU to an overclock of 3.9 GHz across all 16 cores, we began to see a larger disparity in performance. The NH-U14s outperformed the NH-U12s by almost 8 degrees Celsius under load. Interestingly, the smaller NH-U9 didn’t fare much worse than the NH-U12s, which I found surprising given its size. It seems that the extra fan on the NH-U9 helps compensate for its compact design.
#### Noise Levels
Noise is another critical factor when selecting a CPU cooler. For this test, I measured the overall system noise rather than just the CPU fans. The case fans ramp up and down automatically based on CPU temperatures, so it should provide a fairly real-world example of what you can expect.
At idle, the coolers performed similarly in terms of volume, with only slight variations when overclocked. Interestingly, I found that the larger NH-U14s was slightly louder at idle but quieter under full load compared to its smaller counterparts. Given that I’ve run much lower-powered laptops that easily reached 55 decibels, I’m impressed by how quiet these coolers are, even with the demanding task of cooling a 16-core Threadripper CPU.
#### Final Thoughts and Recommendations
It’s important to note that there are other Threadripper chips with fewer cores that might produce less heat. For example, the 12-core Ryzen 1920X or 8-core Ryzen 1900X would likely run cooler than the 16-core Ryzen 1950X I tested here. However, since I don’t currently have those lower-tier Threadripper CPUs to compare against, we’ll focus on the cooling differences with the top-end chip, which should provide a worst-case scenario for these coolers.
Based on my testing, running at stock speeds, any of the three Noctua coolers would handle the Ryzen 1950X without issues, both at idle and under full load. However, when overclocking is introduced, the smaller coolers start to struggle a bit more. If you’re planning on buying a Threadripper and overclocking it, I’d highly recommend going with the larger NH-U14s cooler if you have the physical space in your case.
I initially planned to use this air cooler as a temporary solution until more all-in-one coolers became available that cover the full Threadripper socket. But after my testing, I’m more than happy with the results of these Noctua coolers and will be sticking with the NH-U14s for some time to come.
#### Conclusion
If you’re interested in seeing the full benchmarks from this Threadripper system, make sure to subscribe to my channel for a separate video on that topic. In the comments below, let me know which cooling solution you’ll be using for your Threadripper setup and don’t forget to leave a like if you found this information useful!
Thanks for watching, and don’t forget to subscribe for more tech videos like this one.