SFF.Network L91 Tests 13 Low Profile Coolers on the AM5 7600X and 7700X

Image Credit – L91

 
L91, seen above holding our hopes and dreams in his hands, has put together an excellent look at 13 different low-profile coolers for AM5. Of special interest in the tests is the inclusion of several often forgot Thermalright coolers, as well as the classic Scythe Big Shuriken. L91 also includes the variations of core clocks between the coolers. Plus he has this awesome test bench.
 ...

Continue reading...

 

[scatterforce]

Test conditions are an open-air bench with no GPU attached. That does bias the data a bit, but there are still some surprising takeaways:

Thermalright AXP-120 x67 trades blows with the Noctua L12s. You can't compare fan speeds as they are just % of the PWM signal.

Thermalright AXP-90 x53 Full (copper version) Black performs on par vs the Scythe Big Shuriken 3 Rev. B. I did not expect this, as Scythe has performed better in the past.

AXP-90 x47 Full Copper performed better than ID cooling IS-55. The video also stated that the Aluminum version of the x53 is equivalent to the copper version of the x47, so that would place the 53mm cooler with a 90mm fan ahead of a 57mm cooler with a 120mm fan.

The Blackridge really isn't suited for AM5.

All of the AXP-90 designs appear to be punching above their "height" class. I can see the Nocuta still reigning supreme depending on case configuration.

 

[Revenant]

Test conditions are an open-air bench with no GPU attached. That does bias the data a bit, but there are still some surprising takeaways:

Thermalright AXP-120 x67 trades blows with the Noctua L12s. You can't compare fan speeds as they are just % of the PWM signal.

Thermalright AXP-90 x53 Full (copper version) Black performs on par vs the Scythe Big Shuriken 3 Rev. B. I did not expect this, as Scythe has performed better in the past.

AXP-90 x47 Full Copper performed better than ID cooling IS-55. The video also stated that the Aluminum version of the x53 is equivalent to the copper version of the x47, so that would place the 53mm cooler with a 90mm fan ahead of a 57mm cooler with a 120mm fan.

The Blackridge really isn't suited for AM5.

All of the AXP-90 designs appear to be punching above their "height" class. I can see the Nocuta still reigning supreme depending on case configuration.

Pretty good takeaways from that.

As far as bias goes, it's unavoidable either way. You can test in a case, but which one? Small? Large? Mesh? They all add bias. You can test with a GPU, but again, which one? Flow through? How much wattage?

It really comes down on the user to match their CPU cooler with their CPU, usage, and the case. Luckily, we're all here to help.

 

[ScottSimply]

L91 is also in the SFF.Network and SFF Gurus Discords and is rather active in the latter.

 

[L91]

Test conditions are an open-air bench with no GPU attached. That does bias the data a bit, but there are still some surprising takeaways:

Thermalright AXP-120 x67 trades blows with the Noctua L12s. You can't compare fan speeds as they are just % of the PWM signal.

Thermalright AXP-90 x53 Full (copper version) Black performs on par vs the Scythe Big Shuriken 3 Rev. B. I did not expect this, as Scythe has performed better in the past.

AXP-90 x47 Full Copper performed better than ID cooling IS-55. The video also stated that the Aluminum version of the x53 is equivalent to the copper version of the x47, so that would place the 53mm cooler with a 90mm fan ahead of a 57mm cooler with a 120mm fan.

The Blackridge really isn't suited for AM5.

All of the AXP-90 designs appear to be punching above their "height" class. I can see the Nocuta still reigning supreme depending on case configuration.

As for cases, open testbench provides least bias and not every cooler tested will be compatible with every case. Open air provides best case for everyone.

I chose not to test with GPU installed due to the inclusion of the Alpenfohn BR, which doesn't support GPU installed direcly in PCIe slot. Having GPU installed directly in PSU slot increases temperatures, so having one installed vs without would make BR data pointless. There is also no best solution either, since some cases are reference and some are sandwich layout. The increase in temperatures is also not directly caused by GPU itself, but the blocking of air surrounding the coolers. So by removing the GPU, is one less variable to consider.

If you have any questions, feel free to hit me up.

 

[L91]

Continue reading...

Thanks for the shout out

 

[hereforthefeast]

Not surprised to see these results, but nevertheless thank you for a proper comparison!

AXP90 has always been the cooler to beat at 47mm, and from my own testing the newer AXP120 performs really well due to its 6 heat pipes vs the 4 of the NH-L12.

 

[scatterforce]

As for cases, open testbench provides least bias and not every cooler tested will be compatible with every case. Open air provides best case for everyone.

I chose not to test with GPU installed due to the inclusion of the Alpenfohn BR, which doesn't support GPU installed direcly in PCIe slot. Having GPU installed directly in PSU slot increases temperatures, so having one installed vs without would make BR data pointless. There is also no best solution either, since some cases are reference and some are sandwich layout. The increase in temperatures is also not directly caused by GPU itself, but the blocking of air surrounding the coolers. So by removing the GPU, is one less variable to consider.

If you have any questions, feel free to hit me up.

Thanks for the response and I agree with your assessment. As Revenant said, every test configuration has bias.

I have both the Blackridge and the AXP-90 x47 full. At present, the Blackridge performs marginally better in my case with AM4, but I'm able to use a 120 under the fins with VLP ram.

Thank you for the data.

 

[Revenant]

Thanks for the shout out

Any time!