No. Noctua coolers possess an extremely reliable SecuFirm™ mounting system. Thanks to the screw connection with the backplate on the rear side of the motherboard, the exceedance of the weight recommendations by Intel and AMD common among high-end coolers is completely unobjectionable.

Noctua coolers weren't originally designed for passive operation although the cooling performance is sufficient in many cases. Unfortunately, it is impossible for us to indicate whether a certain CPU can or cannot be cooled without a fan, as this depends on various factors such as case temperature, case ventilation or the orientation of the socket. Generally, we recommend using a fan in the first instance and then lowering RPM while constantly monitoring the temperature.

Please first loosen one of the two thumbscrews that hold the mounting bars in place. Then untighten the second screw so that you can swivel the mounting bar and open the socket. Depending on the position of the mounting bars, you may need to swivel both.

No. As most AMD mainboards come with a stock backplate preinstalled, the cooler doesn’t include an AMD backplate. If your mainboard doesn’t have a backplate, please contact our support team at support@noctua.at.

Some mainboards display a CPU fan error when the fan speed falls below a certain threshold, e.g. 900rpm. If you want to run the fan below this speed (e.g. using the supplied U.L.N.A.), please adjust the threshold in the BIOS or disable CPU fan error warning.

Noctua CPU coolers are shipped with a plastic protection cover at the bottom side of the cooler, which protects the contact surface against scratches and other damage.

The base of Noctua CPU coolers is made from hard-wearing nickel plated copper, so there's nothing to worry about if you buy a Noctua cooler from second hand or if your cooler didn't come with a protection cover. Please proceed with the installation as explained in the manual.

You can simply remove the cover as shown below:

Dust: Fans and heatsinks inside computer cases tend to accumulate dust over longer periods of usage. In order to maintain maximum performance, please clean your fan and heatsink regularly. For cleaning, please first remove the fan from the heatsink and clean it using a duster, slightly moist tissue or canned air. Please be careful not to use too much force in order to prevent any damage to the fan. Please do not use a vacuum cleaner as this may apply excessive force to the fan and do not put the fan under running water as water residues inside the motor may lead to short circuits. Please also note that the fan is not designed to be taken apart by the user. Removing the impeller from the frame will break the sealing of the bearing and results in a loss of warranty. Before reinstalling the fan, clean the heatsink itself with a duster or vacuum cleaner. Do not use water to clean the cooler. Finally put the fan back on and connect it to your motherboard fan header or fan controller.
Thermal paste residues: Whenever you take off the heatsink from the CPU, we recommend to clean the CPU as well as the base of the cooler before re-applying thermal paste and re-installing the cooler. You can either just wipe the base and the CPU clean with a dry, lint-free tissue or, for more thorough cleaning, use a lint-free tissue moistened with either a mild solution of washing-up liquid or isopropyl alcohol. Do not put the cooler or CPU under running water. Note that both the base of the heatsink and the CPU should be dry, free from residues of thermal compound and free from grease before re-applying thermal paste and re-installing the cooler.

The NH-U12DX i4 and NH-U9DX i4 include mounting hardware for LGA2011 Narrow ILM. Other models are mechanically incompatible with the Narrow ILM mounting system and can thus not be used on this platform.

Please kindly note that the NT-H1 thermal compound supplied with our coolers is a consumable item so our manufacturer’s warranty does not include providing additional thermal compound. You can purchase NT-H1 from our resellers.

Please refer to the installation manual and our video installation guides for detailed instructions on how to install the cooler.

As it is not possible to reliably calculate or control the forces that act upon a system during transport (e.g. in shipping), we generally recommend, for safety reasons, taking off coolers with a total weight of more than 700g (incl. fan). If the weight of the heatsink without fan is below 700g and the total weight including fan is above 700g, we recommend taking off the fan from the heatsink in order to reduce the weight below 700g. Noctua cannot be held responsible for any damage that may arise due to excessive stress during transport if you keep the heatsink installed.

Our SecuFirm2™ mounting systems undergo comprehensive compatibility tests before being approved for new platforms. We have neither detected any problem with regard to the Intel LGA1151 platform (“Skylake”) during these tests nor have we received any reports from our customers, sales partners and system integrators that would indicate any possible issue. Our SecuFirm2™ mounting systems (with the exception of some compact L-type models) rely on coil springs to create the required contact pressure, which gives a certain flexibility both with regards to tolerances in stackup height and shocks or other forces. Compared to spring-less mounting systems, which exclusively rely on the bending of the mounting brackets in order to create contact pressure, this allows to reduce the mechanical stress on CPU, socket and motherboard and thus helps to prevent possible damage from excessive forces. However, as it is not possible to reliably calculate or control the forces that act upon a system during transport (e.g. in shipping), we generally recommend, due to safety reasons, to take off coolers with a total weight of more than 700g (incl. fan).

Please refer to our TDP guide in order to select the cooler that offers the best cooling performance at a given height.

In accordance to the International System of Units (SI system), Noctua specifies volumetric flow rate (airflow) using the SI unit of cubic metres per hour (m³/h). The SI unit can easily be converted to cubic feet per minute (CFM): One m³/h is ~0.589 CFM and one CFM is ~1.699 m³/h. For convenient conversion, online tools such as www.convertunits.com can be used.

The specified minimum speed refers to the speed at 20% PWM duty cycle. There can be several reasons why the fan doesn‘t go as low as indicated:

1. Many mainboards do not go below 40%, 50% or even 60% PWM duty cycle on case fan headers. Please refer to your mainboard manual to verify whether the fan header actually goes down to 20% duty cycle. If that's not the case, in some cases you can use the fan speed control software supplied with your mainboard to overcome these limitations and reduce the fan speed even further. However, 3rd party tools like SpeedFan offer greater flexibility and better options to work around the limitations imposed by the mainboard vendors.
   
   
2. Some mainboards feature 4-pin fan headers that actually don‘t use a PWM signal on Pin 4 to control the fan speed but rather reduce the voltage on Pin 2 (like a standard 3-pin fan header). As the fan speed at minimum voltage is usually higher than the speed at 20% PWM duty cycle, the fan can not reach as low minimum speeds under voltage control as under PWM control. Please refer to your mainboard manual to check whether or not your mainboard has 4-pin fan headers that control the fan speed by reducing voltage on Pin 2 rather than by changing the PWM duty cycle on Pin 4:
   
   
   
   Note that mainboard manufacturers use different terms to indicate that Pin 4 is not being used for PWM control (e.g. “+5V”, “VCC” or “NC”), but if one of these terms is used, you can be sure that the fan header does not support PWM. If Pin 4 is described as “Speed Control” or “PWM” or the like, you can be sure that the fan header supports PWM.
   
   Unfortunately, the description of Pin 2 is not always a clear indication as some manufacturers use terms such as “Fan PWR” or “Power” for both types of fan headers. However, if Pin 2 is described as “Speed Control”, you can also be sure that the fan header does not support PWM based speed control. If Pin 2 is described as “+12V”, this is a clear indication that the fan header supports PWM.
   
   Please also note that in some cases, the descriptions of the pin layouts in the mainboard manuals may not be correct and some models actually allow you to switch the fan headers from voltage control mode to PWM control mode in the BIOS even though the pin descriptions do not indicate PWM support. We thus recommend to look for these options in the BIOS before taking other measures. In case of doubt, please contact your mainboard manufacturer.
   
   If your mainboard features other 4-pin fan headers that use PWM for speed control, you can run multiple fans from these headers using Y-split cables. Make sure not to exceed the specified maximum power draw of the fan headers (usually 10-12W) though.

Some mainboards feature 4-pin fan headers that actually don‘t use a PWM signal on Pin 4 to control the fan speed but rather reduce the voltage on Pin 2 (like a standard 3-pin fan header). In this case, it may occur that the mainboard reduces the voltage so much that the fan stops. Please refer to your mainboard manual to check whether or not your mainboard has 4-pin fan headers that control the fan speed by reducing voltage on Pin 2 rather than by changing the PWM duty cycle on Pin 4:

Note that mainboard manufacturers use different terms to indicate that Pin 4 is not being used for PWM control (e.g. “+5V”, “VCC” or “NC”), but if one of these terms is used, you can be sure that the fan header does not support PWM. If Pin 4 is described as “Speed Control” or “PWM” or the like, you can be sure that the fan header supports PWM.

Unfortunately, the description of Pin 2 is not always a clear indication as some manufacturers use terms such as “Fan PWR” or “Power” for both types of fan headers. However, if Pin 2 is described as “Speed Control”, you can also be sure that the fan header does not support PWM based speed control. If Pin 2 is described as “+12V”, this is a clear indication that the fan header supports PWM.

Please also note that in some cases, the descriptions of the pin layouts in the mainboard manuals may not be correct and some models actually allow you to switch the fan headers from voltage control mode to PWM control mode in the BIOS even though the pin descriptions do not indicate PWM support. We thus recommend to look for these options in the BIOS before taking other measures. In case of doubt, please contact your mainboard manufacturer.

To resolve the issue, you can:

1. Choose a higher fan speed profile in the BIOS (e.g. „normal“ instead of „silent“, etc.) or deactivate automatic fan speed control and use a Low-Noise Adaptor instead.
   
   

2. Use the mainboard‘s fan speed control software or 3rd party tools like Speedfan to regulate the fan speed. Unlike BIOS based fan speed control, the supplied software usually checks whether the fan has stopped and increases voltage accordingly or at least offer more options to set up the fan properly.
   
   

3. If your mainboard features other 4-pin fan headers that use PWM for speed control, you can run multiple fans from these headers using Y-split cables. Make sure not to exceed the specified maximum power draw of the fan headers (usually 10-12W) though.

No. Depending on the orientation of the cooler, 1-2mm at the outer edge of the heatspreader may not be covered by the heatsink base. As the heat is concentrated at the center of the CPU and the outermost parts are irrelevant for heat-transfer, this is no problem at all.

The BIOS of many motherboard models from Supermicro expects a certain minimum fan speed (usually 600 or 700rpm), which is higher than the minimum fan speed that can be achieved with Noctua's PWM fan models. The BIOS may interpret the low minimum RPM of Noctua fans as a fan error and thus try to run the fan at 100% for a short period before going back to automatic control. This can result in oscillating fan speeds and fan speed warnings in the BIOS or the fan management console.

Unfortunately there is usually no option to set a lower minimum value in the BIOS that would suit the low minimum speed of Noctua PWM fans. The issue can thus only be resolved by either disabling automatic fan speed control in BIOS, which will cause the fans to run at a constant speed, by using 3rd party tools like IPMItool or IPMIutil, or by requesting a modified BIOS file from Supermicro, which takes the minimum fan speed of Noctua fans into account (charges might apply). Please note that any modifications performed with 3rd party tools are done at your own risk and that Noctua can not be held responsible for any possible issues.

Systems from Acer, Apple, Dell, HP, Lenovo or other major brands often use motherboards which differ slightly from the specifications issued by Intel and AMD. While those changes are usually subtle, they can lead to compatibility issues with coolers that were built to comply with these specifications.

Even in case the cooler is mechanically compatible and can be installed, other issues can occur, e.g. proprietary fan connectors, BIOS errors due to a low fan speed, shutdowns, etc. Some of these problems can be avoided with some technical knowledge, but especially BIOS related issues can often not be resolved.

Due to the large number of possible issues that cannot be resolved with different mounting parts alone, Noctua does not officially support systems from Acer, Apple, Dell, HP, Lenovo or other major brands.

As most of today's PC enthusiasts use high-viscosity thermal compounds, the contact surface of Noctua coolers is optimised for use with this type of pastes. The micro-grooves at the contact surface ensure that high-viscosity thermal compounds are dispersed to a uniform thin layer across the whole contact area and that no air pockets remain between the cooler and the CPU. With a polished, mirror like surface, the risk of uneven dispersion is much higher. As too thick layers of thermal paste and air pockets drastically deteriorate heat transmission, the micro-grooves are is of vital importance to the overall cooling performance of Noctua coolers when used with today's high-viscosity thermal pastes.

Striving for excellence in both product and service quality, we believe that warranty and support should not depend on the customers willingness to provide his data throughout a registration process. We therefore do not offer the possibility to register Noctua products and provide the same 6-year manufacturer’s warranty and the same excellent level of support for all products.

Most Noctua SecuFirm™ mounting parts are cross-compatible, so if you’re replacing one Noctua cooler with another, it is usually possible to keep the original backplate and/or mounting system in place in order to re-use it for the new cooler. If you had your previous Noctua cooler installed on an Intel LGA115x (LGA1156, LGA1155, LGA1151, LGA1150), LGA2011 (LGA2011-0, LGA2011-3) or LGA1366 system and replace it with an NH-U14S, NH-U12S or NH-U9S, you only need to replace the original mounting bars with the NM-IMB3 ones supplied with the new cooler. For all other models, you can keep the entire mounting system in place on the aforementioned Intel sockets. On AMD AM2, AM2+, AM3, AM3+, FM1, FM2, FM2+, you only need to replace the original mounting bars with the ones supplied with the new cooler.

All screws should be tightened gently until they stop without using excessive force. Please do not exceed the following values for maximum tightening torque:

Screw type	Max. torque
NM-SSC1 screws for fixing the fastening brackets to the base of the heatsink	0.5 Nm
NM-ITS1 thumb screws for fixing Intel mounting bars	0.5 Nm
NM-ALS1 screws for fixing AMD mounting bars	0.6 Nm
Spring-loaded screws for fixing the heatsink to the mounting bars	0.6 Nm

The supplied backplate is for LGA115x only and there is no backplate required for installation on LGA20xx (LGA2066, LGA2011-3, LGA2011-0). Please refer to the installation manual or our installation video for how to install the heatsink on LGA20xx.

Older, discontinued models such as the NH-D14, NH-C14, NH-U12P SE2 and NH-U9B SE2 ship with 3-pin fans that can only be controlled by means of voltage adjustment. When these models were introduced in 2009 and 2010, many motherboards supported this type of speed control, whereas today, most motherboards can only control 4-pin fans by means of PWM (Pulse Width Modulation). As indicated on the resepective product pages, these models have been discontinued and superseeded by newer models such as the NH-D15, NH-C14S, NH-U12S and NH-U9S, which, like all current Noctua CPU coolers, support speed control via PWM. If you would like to use PWM speed control, please purchase one of our current cooler models or change the fans for some of our 4-pin PWM fans. Alternatively, you can also use the supplied Low-Noise and Ultra-Low-Noise adaptors in order to reduce the speed of the supplied fans.

Not sure which Noctua product to buy? Our detailed buying guides for fans and CPU coolers help you to choose the model that works best for you.

Most motherboard vendors allow their overclocking-enabled products (e.g. those with Intels X- or Z-series chipsets) to run the processor at increased clock speeds by default, without requiring any user action at all. Since TDP (Thermal Design Power) limits are usually also disabled by default, this leads to the CPU exceeding the rated TDP to a varying degree, depending on the used applications and their workloads. Due to the increased heat output of the CPU, you may see higher CPU temperatures than expected.

In order to find out if your motherboard is overclocking your processor by default, please enter the UEFI BIOS and select the “advanced” or “overclocking” menu. There you should be able to find options such as “MultiCore Enhancement” (options: enabled/disabled), “CPU Ratio Apply Mode” (options: all/per core) or similar. To disable the automatic overclocking, adjust the settings either to “disabled” or “per core” and make sure that the individual multipliers match the original specifications.

In doubt, please contact your motherboard vendor for detailed instructions on how to disable this feature.

Intel's unlocked CPUs (K, X and C suffix) can dissipate more heat than indicated by the TDP specification if

1. the TDP limits are extended or disabled in the motherboards' BIOS.
2. the motherboard applies automatic overclocking by default, e.g. by raising the supply voltage of the CPU and using higher Turbo-Mode multipliers.
3. some software creates untypical loads, e.g. Prime95 with AVX2 support and a) and/or b) apply.

This can lead to temperature issues, especially when using smaller coolers or compact cases.

The actual power draw of the processor can be monitored with software provided by the motherboard vendor or with 3rd party tools like HWInfo or HWMonitor.

If you encounter temperature issues (>90°C) and notice a higher than specified power draw, please ensure that no automatic overclocking is applied and limit the TDP to the specified value by choosing appropriate BIOS settings.

For Kaby Lake CPUs, it may also help to lower the CPU clock speed for applications that heavily use the AVX instruction set, which can lead to higher loads and power draw. This option is usually referred to as “AVX offset” and makes it possible to lower the multiplier specifically for AVX based applications without reducing performance when using other instruction sets. Depending on the quality of the CPU and the programs being used, a reduction of 2-3 steps usually gives very good results.

Please contact your motherboard vendor for details if you have trouble finding the appropriate settings in the BIOS.

All our TDP recommendations are based on thorough testing with the default values specified by Intel using popular applications such as Asus Realbench and prime95. Please note, however, that prime95 creates a particularly high load that goes beyond typical application scenarios and this leads to elevated temperatures. We thus recommend using other programs such as Realbench for checking the stability and temperatures of the CPU in realistic scenarios.

Due to the heatsink mounting mechanism being identical on LGA2011 and LGA2066, Noctua’s SecuFirm2™ mounting systems for LGA2011 also support Intel’s LGA2066 processors. Most current Noctua coolers already include SecuFirm2™ mounting systems for LGA2011 and can thus be used on LGA2066 motherboards without any upgrades or modifications.

The following models include a mounting-kit for socket LGA2011/2066 and are thus compatible out of the box:

NH-C14S
NH-D14 SE2011
NH-D15
NH-D15 chromax.black
NH-D15S
NH-D9L
NH-D9DX i4 3U
NH-L12
NH-L12S
NH-L9x65
NH-U12A
NH-U12S
NH-U12S chromax.black
NH-U12DX i4
NH-U14S
NH-U9DX i4
NH-U9S

The following models can be made compatible with the LGA2011/LGA2066 sockets free of charge using the NM-i20xx upgrade-kit:

NH-C12P
NH-C12P SE14
NH-C14
NH-D14
NH-D15 SE-AM4
NH-U12S SE-AM4
NH-L9x65 SE-AM4
NH-U12
NH-U12F
NH-U12P
NH-U12P SE1366
NH-U12P SE2
NH-U9
NH-U9B
NH-U9B SE2
NH-U9F

The following models can be made compatible with the LGA2011/LGA2066 sockets using the NM-i20xx upgrade-kit but are not eligible for Noctua‘s free mounting offer, so users have to purchase the kit at local resellers:

NH-U12DO (Note that the A3 version is not compatible!)
NH-U12DX
NH-U12DX 1366
NH-U9DX 1366
NH-U9DO (Note that the A3 version is not compatible!)

The following models are not compatible with the LGA2011/LGA2066 sockets and can not be upgraded:

NH-L9a
NH-L9a-AM4
NH-L9a-AM4 chromax.black
NH-L9i
NH-L9i chromax.black
NH-U14S TR4-SP3
NH-U12S TR4-SP3
NH-U9 TR4-SP3
NH-U14S DX-3647
NH-U12S DX-3647
NH-D9 DX-3647 4U
NH-U12DO A3
NH-U9DO A3

Unfortunately, it is not possible to upgrade existing Noctua heatsinks to support the AMD TR4 and SP3 sockets for Ryzen Threadripper (X399) and Epyc CPUs. TR4/SP3 CPUs have much bigger heatspreaders (as compared to e.g. LGA2066 or AM4 processors) and the contact surfaces of standard Noctua heatsinks would cover only about half of these heatspreaders, which would result in insufficient cooling performance. On top of that, the heatpipes of bigger cooler models like the NH-D15 or NH-D15S would completely block the RAM slots on many TR4/SP3 motherboards. For this reason, Noctua has introduced the new NH-U14S TR4-SP3, NH-U12S TR4-SP3 and NH-U9 TR4-SP3 cooler models that feature bigger contact surfaces and have been tailored to fit TR4 and SP3 systems. Please choose these models for AMD Ryzen Threadripper and Epyc systems.

Depending on the chassis, card and motherboard being used, installing PCIe cards in a vertical position using riser cards may cause compatibility issues with CPU coolers. Whether there will be a problem or not depends on the exact position of the card (which is determined by the PC case and/or optional vertical GPU holder bracket), the width of the card as well as the position of the CPU socket on the motherboard. In order to verify that there is sufficient space for the cooler, please measure the distance between the centre of the CPU socket and the top of the PCI card in order to make sure that the cooler fits.

The mounting is identical on all LGA115x sockets (LGA1150, LGA1151, LGA1155, LGA1156). The following coolers are thus compatible with LGA115x out of the box and don’t require any upgrades in order to be used on this platform: NH-C12P SE14, NH-C14, NH-C14S, NH-D15, NH-D15S, NH-D14, NH-L12, NH-L12S, NH-L9i, NH-L9x65, NH-U14S, NH-U12S, NH-U12P SE2, NH-U9S, NH-UD9L, NH-U9B SE2. In addition, all older models which have been upgraded with the NM-I3 kit can also be used on LGA115x without any further upgrades.

The following coolers can be made compatible to LGA115x using the NM-i115x Upgrade-Kit: NH-C12P, NH-D15 SE-AM4, NH-L9x65 SE-AM4, NH-U12, NH-U12F, NH-U12P, NH-U12P SE1366, NH-U12S SE-AM4, NH-U9, NH-U9F, NH-U9B.

The following coolers are mechanically incompatible with the NM-i115x Upgrade-Kit and can not be made compatible with LGA115x: TR4-SP3 coolers, DX-3647 coolers, DO-Series AMD Opteron coolers, NH-L9a, NH-L9a-AM4.

Current processors from Intel, especially the unlocked models with the suffixes C, K, X or XE (e.g. Core i7-8700K, Core i9-7900X oder Core i9-9900K) can dissipate noticeably more heat than indicated by the TDP specification.

While many motherboard models stick to the recommended power levels and clock speeds settings from Intel, there are models where the UEFI BIOS is configured in such a way that:

1. the TDP limits are extended or disabled.
2. the motherboard applies automatic overclocking by default, e.g. by raising the supply voltage of the CPU and using higher Turbo-Mode multipliers.

Depending on the load scenario and CPU configuration settings, the actual power draw can be 1.5-3 times as high the stated TDP value. This can lead to temperature issues, especially when using smaller coolers, compact cases or cases with poor airflow, when some applications create very high loads, e.g. Prime95 with AVX2 support, 3D-rendering or simulation software.

The actual power draw of the processor can be monitored with software provided by the motherboard vendor or with 3rd party tools like HWInfo or HWMonitor.

If you encounter temperature issues (>90°C) and notice a higher than specified power draw, please ensure that no automatic overclocking is applied. You might also want to limit the TDP to either the specified value by Intel or a value that is 5-10W below the value previously read out from the CPU. This can be done in BIOS by setting the “Long Term Duration Power” manually instead of “Auto”.

Furthermore, it may also help to lower the CPU clock speed for applications that heavily use the AVX instruction set, which can lead to higher loads and power draw. This option is usually referred to as “AVX offset” and makes it possible to lower the multiplier specifically for AVX based applications without reducing performance when using other instruction sets. Depending on the quality of the CPU and the programs being used, a reduction of 2-3 steps usually gives very good results.

The option to set an AVX-offset value is usually available for the following CPU generations:

• LGA1151:7th, 8th and 9th generation (Kaby Lake, Coffee Lake)
• LGA2011-3: 6th generation (Broadwell-E)
• LGA2066: 7th and 8th generation (Skylake-X)

Please contact your motherboard vendor for details if you have trouble finding the appropriate settings in the BIOS.

All our TDP recommendations are based on thorough testing with the default values specified by Intel using popular applications such as Asus Realbench and prime95. Please note, however, that prime95 creates a particularly high load that goes beyond most typical application scenarios and this leads to elevated temperatures. We thus recommend using other programs such as Realbench for checking the stability and temperatures of the CPU in realistic scenarios.

Please also note that CPUs that use thermal paste instead of soldering between the DIE and the heatspreader generally run hotter. Due to the thermal bottleneck created by the thermal paste, temperatures can be high even at the specified TDP.

The latest Ryzen 3000 processors (except APU models) differ from previous Ryzen generations in that they are no longer based on a single, large chip but use a multi-chip approach with smaller chips instead. Depending on the exact model there can be one (6 and 8-core models) or two (12 and 16-core models) actual CPU-Dies (CCD) on the package. Each processor also uses an I/O-Die (IOD), which contains things like the memory controller, PCIe controller, connections to the motherboard chipset and other functions.

Because of this design change and the switch to a smaller 7nm manufacturing process, the heat distribution of the overall processor is much different from older 14nm and 12nm based single-chip Ryzen processors with a similar power draw.

Depending on the exact CPU model, its specified TDP value and possibly extended power limits (precision boost overdrive), a single CPU-die can create a heatload of up to 130W easily, whereas the I/O-die usually creates a heatload of about 10W. Due to the small size of the CPU-die, the heat density (W/mm²) of this chip is very high. For example, a 120W heatload at a chip-size of 74mm² results in a heat-density of 1.62W/mm², whereas the same heatload on an older Ryzen processor with a chip-size of 212mm² gives a heat-density of just 0.57W/mm².

This large difference in heat-density is the reason why newer Ryzen 3000 processors become much warmer at similar heatloads than their predecessors.

Furthermore, Ryzen 3000 CPUs are using the rated temperature headroom (up to 95°C) quite aggressively in order to reach higher boost clocks. As a result, it is absolutely no problem and not alarming if the processor runs into this temperature limit. The clock speed and supply voltage will be adjusted automatically by the processor itself in order to remain within AMD’s specifications and to prevent overheating.

Due to the higher heat density, higher thermal limits and more aggressive boost clock usage, it is perfectly normal that Ryzen 3000 CPUs are reaching higher temperatures than previous generation Ryzen CPUs with the same TDP rating. Higher CPU temperatures are normal for Ryzen 3000 processors and not a sign of that there is anything wrong with the CPU cooler.

Noctua has conducted internal tests with various coolers and CPU samples to find out whether or not the orientation of the heatpipe layout has an effect on cooling performance when used with Ryzen 3000 CPUs. The results have shown no evidence that this is indeed the case. However, as there can be minute variances in surface shape both with the CPU heatspreader and with the base of the cooler, it is possible that one orientation has a better fit than the other and hence gives slightly different results. A different orientation of the cooler also changes the airflow within the chassis though, and this can cause different temperatures as well.