Your CPU is probably the most critical component of your computer. Unlike the GPU, the processor is used for literally any task you might perform on your PC. From reading the latest Reddit posts to editing 4K videos, the CPU is the component that takes the brunt of the work. And when it works, the CPU heats up, and your processor can overheat if your cooling setup isn’t up to the task.
But what are the safe CPU temperatures, and how can you monitor your CPU’s thermals? Read on if you want to learn more about safe CPU temperatures for Intel & AMD CPUs, how to monitor CPU thermals, and what to do if your CPU overheats. Our guide will try to answer these and more questions, and by the end of it, you should learn enough about CPU thermals to know when your CPU is running too hot and what to do to cool it down.
If you happen to find out your CPUs is overheating you can try undervolting it to lower temperatures. Read our CPU & GPU undervolting guide to learn all about it. To learn how to recognize when your CPU is overheating read our CPU overheating guide. Finally, our list of the best air coolers for Ryzen CPUs has a ton of excellent picks that can keep your CPU cool while being whisper quiet.
How to monitor your CPU thermals
HWiNFO64
Before we talk about the safe CPU temperature range and what to do if your CPU gets hotter than what’s deemed safe, let’s show you how to monitor CPU thermals. There are a number of quality system monitoring apps, but our favorite is HWiNFO64.
This lightweight app is easy to install and set up and can show readings from every thermal sensor on your PC. Here’s the download link, and below you can see a screenshot of our CPU’s thermal readings when idle and under heavy load during Prime95 torture tests.
As you can see, our Ryzen 5600X has four different thermal readings. First of all, you won’t find readings for each individual core on Ryzen CPUs because they don’t support this feature. When it comes to Ryzen processors, you can read the following data:
- Motherboard CPU temperature: this is a thermal couple installed next to the CPU die on your motherboard. It shows relatively accurate readings but is not the best data point to use.
- CPU (Tctl/Tdie): This data point shows the highest temperature measured by any thermal sensor inside the CPU die. When it comes to accuracy, this is probably the most accurate reading and one you should focus on when looking for your CPU thermals.
- CPU Die (average): This is the average temperature of all thermal sensors inside the CPU die. Not as handy as the CPU (Tctl/Tdie) reading, but still pretty useful.
- CPU CCD1 (Tdie): This data point shows you the temperature of an individual CCD (Core complex die, core chiplet) on AMD CPUs. Since the 5600X is a six-core part and since each CCD can host up to eight working CPU cores, we only have reading for one CCD. If you own a Ryzen 9 5900X or the 5950X, you’d have reading for CCD1 and CCD2 since those two CPUs feature two core chiplets.
As for the Intel CPUs, HWiNFO can show the temperatures of each individual core, which should be your focus point. When monitoring thermals of Intel CPUs, always look for the hottest core and let that be your primary data point regarding CPU thermals. That’s the same as using the CPU (Tctl/Tdie) reading on AMD CPUs, which shows the hottest point on the processor die.
RTSS + MSI Afterburner
HWiNFO64 is an excellent app for thermal monitoring, but you can’t simply alt-tab all the time while playing a game or using any other full-screen app. You can show the info on the second screen, but not all of us have multimonitor setups. So, the best thing you can do to monitor your CPU thermals while in-game is an app called Riva Tuner Statistic Server (RTSS). This is a small but powerful application that comes with OSD (on-screen display) feature that can be shown in any game.
The best way to download RTSS is to download the app alongside another great program, MSI Afterburner. MSI Afterburner is the best monitoring and tweaking app for your graphics card, so we recommend getting it in any case. RTSS is bundled in the MSI Afterburner installation package, and once you install both apps, you can open RTSS, turn on the OSD, and then pick which data points to show in the OSD in HWiNFO and MSI Afterburner.
Just open the HWiNFO settings page and then select which data points to show in RivaTuner. Our advice is that, for AMD systems, pick either your motherboard’s CPU thermal reading or the CPU (Tctl/Tdie) value. As for Intel systems, it’s best to select each core since you want to know your CPU’s hottest point. Finally, before you finish the setup, go to the MSI Afterburner settings->On-Screen Display, and assign keys for showing and hiding the OSD.
When you do all this, you should be ready to monitor your CPU thermals while gaming and running full-screen apps. You can either use HWiNFO (for windowed apps, such as Prime95) or the RTSS OSD (for full-screen apps, such as games). Now, let’s talk about how you can stress test your CPU and check its maximum thermals.
How to stress test your CPU
Now, we have an excellent guide about the best PC stress test tools for your CPU, GPU, and RAM. We recommend checking it and reading the part about CPU stress testing. But if you want a short guide on CPU stress testing, keep on reading.
First of all, when stress testing, you should let the test run for a longer period of time so that your CPU can max out its thermals. We recommend running a test for at least half an hour, but you can freely run tests for longer than that.
A couple of hours during the hottest period of the day is great for nailing down the highest possible temperature your CPU can achieve with its current cooling setup. Overnight tests aren’t that helpful for testing thermals; they’re much better for testing system stability.
When it comes to the best CPU stress tests, check out the following apps:
- Prime95 (use small FFTs for the highest CPU load)
- OCCT
- Blender Benchmark and custom Blender Renders
- Cinebench R23 (set the minimum test duration at at least half an hour)
Aside from using the aforementioned CPU stress tests, you can also play CPU-intensive games to see what you can expect from your CPU during prolonged realistic loads, not stress tests that put the CPU under extreme and unrealistic loads.
Our recommendations include Cyberpunk 2077 with Ray tracing effects or the recently released Marvel’s Spider-Man Remaster for the PC. If you pick Spider-Man, make sure to turn RT effects to the max and set the “Object Range” option for RT reflections to 10.
If you don’t own an RT-capable GPU we recommend Shadow of the Tomb Raider, newer Total War games, Civ VI, or Battlefield 2042 multiplayer on a 128-player map.
Okay, now let’s talk about safe CPU temperatures when it comes to Intel & AMD desktop CPUs as well as safe temperatures for laptop CPUs.
Safe CPU temperatures for Intel CPUs
So, what is a normal CPU temp for Intel CPUs? Well, the short answer is anything below 100 degrees Celsius. Newer Intel processors all have a maximum Tjunction temperature (the highest temperature on a CPU die before the CPU starts throttling) of 100 degrees Celsius.
This means that as long as the hottest point on the CPU die stays below 100 degrees, the CPU will work at full power. Once any point on the CPU die reaches above 100 degrees, the CPU will start slowing down in order to cool itself down. In other words, it will start to thermal throttle.
Now, while anything below 100 degrees Celsius is technically safe, most CPUs should operate at thermals that aren’t higher than 85-90 degrees Celsius. As long as your case and cooling setup are up to the task, of course. When it comes to CPU idle temp, it shouldn’t go over 50 degrees Celsius. But the issue here is that there are countless possible combinations of the CPU, CPU cooler, case, case fan setup, etc.
Just look at the screenshot below, taken from Gamers Nexus’ review of the new Lian Li Lancool III case. As you can see, the CPU temperature can vary by more than 20 degrees Celsius depending on the case you’re using. In other words, just one part of your setup can lower or increase your CPU thermals by more than 20 degrees!
Newer Intel CPUs can also greatly vary in their power consumption and heat generation depending on the task at hand, whether you have removed their power limits, overclocked them, etc. For instance, if you own a Core i5-12600K processor cooled by a decent dual heatsink air CPU cooler with the setup housed inside a case with solid airflow, your thermals shouldn’t go much over 70 degrees Celsius even during the most gruesome torture tests.
But if you overclock the same CPU to 5GHz+ and run the same torture test with the same cooling setup, you might get thermals higher than 90 degrees Celsius. Now, this might seem too high and could be worrying, but the 12600K consumes a ton of power when overclocked and under 100 percent load – almost 320W when overclocked during the Prime95 stress test, which is insane – so anything below 100 degrees Celsius during a hardcore CPU torture test should be fine.
On the other hand, the 12900K runs pretty cool during gaming loads, even when coupled with a 280mm AIO. But run any CPU torture test with the same setup, and chances are the 12900K will start to thermal throttle. The reason is that, under maximum load, even 360mm AIOs can have issues keeping the 12900K at or below 95 degrees Celsius.
What we’re trying to say is that when it comes to your CPU thermals, you always have to take context into account.
A highly overclocked 12900K will go over 90 degrees without issues even when paired with a large AIO during CPU torture tests. You might get a case with better airflow, install extra case fans, or even change the CPU cooler, but the CPU thermals won’t drop because the 12900K uses a ton of power and is extremely hard to cool down, even with the most potent AIOs on the market.
On the flip side, if you own a Core i5-12600 chip that runs at lower power and doesn’t support overlock, a working temperature close to 90 degrees Celsius would definitely be worrying. In that case, even swapping one part of the setup – such as getting a new CPU cooler – might lower CPU thermals considerably.
Also, there’s the question of CPU idle temp. Many people consider anything over 40 degrees Celsius as too high, but from personal experience, you shouldn’t worry until CPU idle temps surpass 50 degrees Celsius. For instance, our Ryzen 5600X runs at 46-48 degrees when idle. It’s because we tweaked the CPU cooler to slow down to the lowest possible fan RPM when the CPU is idling to stay as quiet as possible.
So, the safe CPU temperature range for Intel CPUs is relative to the CPU you have, its TDP, and your cooling setup. But if we had to give one answer that covers most modern Intel CPUs, it would be that you shouldn’t worry as long as your thermals don’t go over 80-85 degrees while gaming or 90 degrees while torture testing the CPU or using it for heavy workloads such as 4K video editing, CPU rendering, extended code compiling tasks, etc.
When it comes to overclocked and high-end Intel processors such as the Core i7-12700K and the 12900K, you should lower your expectations. In gaming, it’s okay if these CPUs run up to around 80-85 degrees Celsius. In torture tests, you should expect thermals higher than 90 degrees Celsius, especially for the 12900K. But, as long as even a single core isn’t thermal throttling, you should be fine.
Safe CPU Temperatures for AMD CPUs
AMD CPUs are much better optimized when it comes to power consumption. You’ve got Ryzen CPUs running either at 65W or 105W, and they tend to not go over those numbers even during the most extreme torture tests, at least with default power settings. This makes them much easier to cool down than Intel CPUs.
For instance, you can use a quality mid-range dual-heatsink CPU cooler such as the DeepCool AK620 even for the flagship Ryzen 9 5950X CPU. And even during the highest loads, the 5950X shouldn’t go much over 80 degrees Celsius when coupled with this cooler.
Also, Ryzen processors tend to boost their clocks as long as they have thermal headroom. This isn’t that noticeable at default settings since you’ve got a hard-imposed power consumption limit – either 65W or 105W.
However, if you try to manually overclock them or use PBO (Precision Boost Overdrive), which is a sort of an auto-OC option for AMD CPUs, you’ll notice that they can get pretty hot while trying to boost their frequencies as high as possible. It’s because now, the CPUs aren’t hard limited to 65W or 105W.
AMD actually confirmed that Zen 3 processors (Ryzen 5000 family) hitting 90 degrees or 95 degrees Celsius (in the case of the 5600X) is completely fine. In other words, if you’re seeing your AMD Zen 3 CPU working at or near 90 degrees Celsius, you shouldn’t be worried. But, of course, you shouldn’t forget about context.
If you have a Ryzen 5600X housed inside a high airflow case and paired with, let’s say, a quality 240mm AIO, thermals that get near 90 degrees Celsius should be worrying. Even if you use PBO.
On the flip side, if you own a 5800X, which is the hardest Ryzen 5000 CPU to cool because it features only one CCD filled with eight cores – the smaller the heat dissipation surface, the harder a CPU is to cool down – temperatures around 80 degrees Celsius are okay, even if you pair it with a decent dual-tower air cooler or a 240mm AIO.
Our CPU, the 5600X, is paired with a Scythe Mugen 5 Rev. B heatsink. This is one of the best single-tower air coolers on the market. And while our CPU’s idle temp is almost 50 degrees Celsius on a warm summer day the CPU will barely reach 72-73 degrees Celsius, even during an hour-long Prime95 torture test.
However, if we turn on PBO, even a minute or so in Prime95 will raise our CPU temperature to over 85 degrees Celsius. And while some might balk at that temperature, in reality, that’s completely fine since we’re using PBO, and Zen 3 CPUs are designed to boost as long as they have thermal headroom.
Below you can find a slide from AMD that answers the question of safe temperatures for Ryzen CPUs excellently. You have a list of safe thermals, depending on the cooler you’re using. The list is pretty good and faithful to real-life data. The only thing we’d like to add is that, in case you’re using a high-end cooler, don’t freak out if you see thermals higher than 80 degrees Celsius.
Despite AMD claiming that the max temperature shouldn’t break the 80 degrees limit, you should expect higher thermals depending on your setup and use scenarios. For instance, you’re rocking an air cooler or a 240mm AIO, living in a hot climate, and using a high-end Ryzen 5000 CPU all day long for stuff like CPU rendering or 4K video editing. In that scenario CPU thermals higher than 80 degrees Celsius are completely fine.
Okay, to summarize. When it comes to Ryzen 5000 processors, anything below 90 degrees Celsius should be fine. Going over the 90-degree Celsius mark should be worrying, especially if you have a quality case, lots of case fans, and a quality CPU cooler. Older Ryzen processors run cooler but not by much. If you own an older Ryzen processor, we’d say that anything lower than 85 degrees Celsius should be fine.
Safe CPU temperatures for laptop CPUs
Up until now, we’ve talked about desktop CPUs. In the world of convertibles, things are different. Many powerful gaming laptops can constantly run their CPUs at the edge of thermal throttling. Other, less powerful laptops can also have their CPUs always working at temperatures at or over 80 degrees Celsius.
For instance, the Scar 17 SE, the most powerful gaming laptop from ASUS in 2022, equipped with top-of-the-line Core i9-12950HX, has its CPU running at almost 90 degrees Celsius when in high-performance modes. The MSI Titan GT77, on the other hand, has a bit weaker – on paper at least – but still quite powerful Core i9-12900HX that runs at less than 75 degrees Celsius during torture tests in high-perf modes.
Now, you might think that the ASUS laptop has a more powerful CPU that needs more power and thus, runs warmer, but the reality is different. The Core i9-12900HX in the MSI Titan GT77 uses more power (75 watts) than the ASUS device (about 65 watts at max load) while being cooler, which tells us that the MSI laptop has a much better cooling setup.
This example shows the most significant issue when talking about laptop CPU temperature range. In the world of laptops, one CPU model can work at many different TDPs depending on the exact laptop model the CPU is used in. For instance, you can have a thin and light laptop with its CPU running at 70 degrees Celsius under load. Then you have a powerful gaming machine with the same CPU. And in the latter case, the same CPU can run much hotter because it has a higher TDP and can use much more power.
So, when talking about safe CPU temperatures in laptops, things get really messy really fast. You might think that your CPU runs too hot, but when you check out reviews, you might find that the constant near-thermal throttle behavior is completely fine.
Also, some cheaper gaming laptops have weaker cooling setups despite rocking powerful CPUs. This could lead to these CPUs constantly working at temperatures higher than 90 degrees Celsius. This might look unsafe but is a planned behavior by the manufacturer.
If we had to give you a general rule of thumb when it comes to CPU thermals in laptops, don’t worry too much as long as your CPU isn’t working at 90 degrees Celsius or higher during heavy gaming sessions, torture tests, or heavy CPU loads such as CPU rendering. If you notice temperatures higher than 90 degrees, check reviews of your laptop model to see whether that’s normal behavior or not.
If you cannot find reviews, look for user reports on forums and sites such as Reddit. And if your investigation ends with a conclusion that your laptop CPU is running too hot, we have a great guide on how to reduce CPU temperature in your laptop for you to check out.
Variables that can affect your CPU temperature
Now, let’s talk about different causes that can potentially lead to your CPU overheating and potential solutions.
Ambient temperature – this is a variable that slips the mind of many users. Don’t forget that all those professional hardware reviewers keep their ambient temperature at or around 20 degrees Celsius.
So if you see that a reviewer achieved a max CPU temperature of 80 degrees while you’re getting near 90 degrees and you have the same cooling setup but live in a hot climate with outside temperatures surpassing 35 degrees Celsius, the ambient temperature might be the culprit behind your CPU’s higher than expected thermals.
Also, if you notice that your CPU runs near 90 degrees Celsius during the hottest summer days but is fine during the rest of the year, the cause for higher-than-average thermals is pretty clear.
Improperly mounted CPU cooler – A CPU cooler that’s improperly mounted is probably the most frequent cause of higher-than-normal CPU thermals. You might’ve applied too little pressure when mounting it, or you may used too little thermal paste. Hell, the situation where users remove their CPU cooler and find the plastic label still glued to the heat spreader happens very frequently.
So, before you swap any component in your setup to try to lower your CPU temperature, dismount your cooler and reapply the thermal paste or simply check whether you forgot to remove the plastic label. This is especially important if you happen to own a prebuilt gaming PC since those often have less-than-ideally installed CPU coolers.
Weak CPU cooler – It might happen that your CPU cooler isn’t up to the task. All those CPU reviews usually include overkill cooling setups consisting of massive 360mm AIOs. Also, many hardware reviewers use open benches and not closed cases, meaning that they always have the best possible airflow setup.
So, if you see much higher values for your CPU in your setup versus an online review, maybe you don’t have as robust a CPU cooler. Further, as long as your CPU thermals are higher than you see online but still in the safe range, you shouldn’t worry.
Another potential issue is that box coolers some CPUs are shipping with are only good enough for default power settings. If you decide to use PBO with AMD CPUs or remove power limits for Intel CPUs, those box coolers might prove too weak for the new setup.
For instance, if you own a CPU such as the Core i9-12900, a low-power version of the 12900K, the CPU might work great at default settings even with the meek box cooler it came with.
However, if you decide to remove power limits in the BIOS, the box cooler won’t be enough to keep the CPU under control. It might keep the CPU at or near 90 degrees Celsius, but it will prevent it from reaching its max boost clocks. It also might get uncomfortably loud since it has to work at 100 percent RPM all the time now that your CPU is using much more power.
In situations like that, you might want to replace the default cooler with something beefier. The same can be said if you happen to have a decent cooler coupled with a non-overclocked K-series Intel CPU or AMD CPU running without PBO. Once you overlock the CPU or turn on PBO, thermals might go over the comfort zone, and replacing the cooler might be a good idea.
Poorly ventilated case – As you already saw, different enclosures can increase CPU temperature as high as 20 degrees Celsius! That’s a massive increase, and a poorly ventilated case can keep your CPU running too hot even if the rest of the setup is fine.
If you have a case with a solid front panel or a chassis that doesn’t have air intakes at the sides, top, or rear, it might seriously negatively affect your CPU temperatures. If you’re seeing higher than usual thermals for both your CPU and GPU and you have already reseated the CPU cooler, it might be that your case is simply too poorly ventilated for your CPU to work at optimal temperatures.
If you decide to replace the enclosure, pick one that has great airflow and is easy to work in. Our guide for the best PC cases for airflow has a number of excellent pick.
Suboptimal case fan setup – Another potential reason for your CPU overheating is a poor case fan setup. You might not even have a single case fan inside your housing. If you don’t have any case fan installed but own a case with more than decent cooling potential, hook up some case fans and see whether you’ll see the difference in CPU thermals.
There are a couple of simple case fan setups that should improve airflow in most cases. If you have an enclosure with a front air intake or dual intakes at the sides, mount two fans at the front and use them to push air into the chassis.
At the same time, install one case fan on the rear side and set it to pull the air from the case. Finally, if your chassis has a top side with air intakes, you might also install one or two case fans up there, both set to pull the air from the case.
Users rocking housings with solid front panels might try installing one fan on the rear of the case and one fan on the top, both set in pull mode. This way, they will create a negative pressure setup, pushing air inside the enclosure through openings on the back and the bottom of the case.
If you have improved your case fan setup and still see the same thermals, maybe it’s time to replace the whole case.
Safe CPU temperatures – conclusion
As you can see, different CPUs vary significantly in their power consumption, and this single variable massively affects their temperatures. Next, CPU thermals can be noticeably affected by a poorly mounted CPU cooler, a poorly ventilated case, a less than ideal case fan setup, or by the simple fact that your CPU cooler isn’t up to the task.
When it comes to the safe CPU temperature ranges, they are different for different CPUs. In general, if the CPU is under high load, we can say that the following thermals are considered safe:
- AMD Ryzen 5000 CPUs: anything lower than 90-95 degrees Celsius is safe. The sweet spot should be between 65-85 degrees Celsius, depending on your CPU and the cooling setup.
- AMD Ryzen 1000, 2000, and 3000 CPUs: temperatures lower than 85-90 degrees Celsius are safe. Optimal values are between 60-80
- Non-K Intel CPUs: Anything lower than 100 degrees Celsius is safe. Optimal temperatures are between 60-85 degrees Celsius, depending on the CPU and the cooling setup.
- Newer K-series Intel CPUs: Anything lower than 100 degrees Celsius is considered safe. If you rock a non-overclocked CPU, optimal thermals should hover between 65-90 degrees Celsius, depending on the CPU and the cooling setup. In the case of the Core i9-12900K, as long as the CPU isn’t throttling, you’re fine. When it comes to overclocked i5s and i7s, we’d be satisfied with thermals that hover at or are lower than 90 degrees Celsius.
- Laptop CPUs – This can vary greatly depending on the specific laptop model. However, if you’re seeing thermals lower than 90 degrees Celsius, you should be fine. If you’re rocking a thin and light laptop with a low-power CPU, anything under 80 degrees should be fine. On the other hand, if you own a beastly gaming laptop with a high-end CPU that uses lots of power, that CPU can constantly work at the thermal throttling limit and be fine.