Core i9-13900K Cooling Guide: Testing Intel’s Flagship With Budget Air and Big AIOs
When Intel launched its 12th Generation Alder Lake CPUs, the Core i9-12900K presented cooling challenges in some scenarios due to the increased heat density of the Intel 7 manufacturing process. However, with the launch of Raptor Lake and especially the Core i9-13900K, Intel has bumped up both the core count and clock speeds of their latest flagship processors. As a result, once the power cap is lifted, it can draw over 330W while running Cinebench R23 (about 100W higher power consumption than the i9-12900K), and it’s not easy to cool.
Below we take a look at Intel’s Core i9-13900K and what it takes to cool it. We also test it with more basic air coolers to see what we gain and lose with different levels of cooling.
New test configuration
| CPU | Intel i9-13900K |
| Tested comparison air cooler | DeepCool AG620, dual tower air cooler |
| Thermalright Assassin X120 R SE, single tower air cooler | |
| Thermalright AXP120-X67, SFF Air Cooler | |
| Compare tested AIO coolers | Deep Cool LT720 (360mm) |
| motherboard | Asus TUF Gaming Z690 Plus Wi-Fi DDR5 |
| sheep | Important DDR5-4800 |
| GPUs | Intel ARC A770 LE |
| case | Cooler Master HAF 700 Berserker |
| PSUs | Cooler Master XG Plus 850 Platinum PSU |
In today’s test, we’re testing an entry-level air cooler, a high-end air cooler, and a high-end liquid cooler to see how different levels of cooling affect the Raptor Lake.
Starting with the SFF cooler, we use Thermalright’s AXP120-X67 to test.
We also tested Thermalright’s Assassin X 120 R SE to show more typical low-end cooler results. This is an entry level single tower air cooler that sells for around $20.
I’m testing DeepCool’s AG620 to understand how a high-end air cooler works with the i9-13900K. Previously reviewed AK620.
We also test DeepCool’s recently released LT720 360mm AIO liquid cooler to see how top-of-the-line liquid cooling will affect Raptor Lake. The LT series is his AIO successor to his LS series which was a mighty cooler when paired with Intel’s 12900K.
test method
Peak CPU temperatures have been a source of concern in the past, but enthusiasts should learn to accept high temperatures as ‘normal’ while running demanding workloads on Raptor Lake and Ryzen 7000 CPUs. there is.
The latest AMD and Intel CPUs are designed to run fairly hot without problems, up to 95 degrees Celsius for AMD Ryzen 7000 CPUs and 100 degrees Celsius for Intel Raptor Lake CPUs. Similar behavior has been standard in laptops for years due to cooling limitations.
Additionally, Intel’s i9-13900K supports Adaptive Boost Technology (ABT), which allows Core i9 processors to dynamically boost to higher all-core frequencies based on available thermal headroom and electrical conditions. increase. This allows all cores to operate at rated single-core speeds of up to 5.8 GHz with the required amount of heat dissipation. This feature works in an aggressive hot-seeking manner. If the chip detects that it is operating below the 100°C threshold, it will increase performance and power consumption until it hits the safe 100°C limit, maintaining higher clocks (and providing better performance). increase).
With previous generation CPUs for coolers, it was fairly easy to keep the processor well below TJ max (the maximum temperature the CPU can sustain without throttling) under demanding workloads, but the current generation extreme cooling (or enable power capping) on CPUs.
When we started testing coolers with Intel’s i9-12900K we found that many that cooled well enough i9-10900K struggled in some scenarios when paired with Alder Lake CPUs Raptor Lake Cooling is even more difficult in these situations.
The increased cooling challenges with Raptor Lake meant that we had to change some of our testing methods.Some coolers were able to pass the Cinebench R23 multicore test with Intel’s 12th Gen i9-12900K, but only the most powerful models were able to pass that test. liquid cooler and all In this scenario, the air cooler I tested “failed” because the CPU reached TJ max (maximum temperature before throttling).
On Raptor Lake’s 13900K, not a single cooler has been able to keep the CPU below its max TJ in this test. As mentioned above, that’s because the Raptor Lake flagship is designed to dial up power. Until then reach that temperature. Instead, compare performance by comparing total benchmark scores and sustained clock speeds.
Asus’ TUF Gaming Z690 Gaming Plus WIFI motherboard is used to test Intel’s i9-13900K CPU. Cooler Master’s HAF 700 Berserker Case fans are limited to 35% speed. The motherboard’s default fan curve is used for the CPU cooler fan.
In addition to testing Cinebench with no power cap applied, we also show the results of limiting the CPU power consumption to a more reasonable 200W. We also show the results at 125W for those who prefer very quiet cooling at the expense of some performance. For both of these results, we show the traditional delta result versus ambient temperature.
LGA1700 socket bend
Keep in mind that besides the CPU cooler, there are many other factors that affect cooling performance, such as the case you use and the fans attached to the case. Your system’s motherboard can also affect this. suffer from bendingThe result is poor cooler contact with the CPU.
To prevent bending from affecting the cooling results, Thermalright’s LGA 1700 contact frame was attached to the test rig. This means that if the motherboard is subjected to bending, the thermal results will be worse than those shown below. Not all motherboards are affected by this issue. We tested the Raptor Lake CPUs on two motherboards. One of them showed a significant thermal improvement after installing his Thermalright’s LGA1700 contact frame, while the other motherboard had no difference in temperature at all!
