[SOLVED] 12700K - TDP and PL2 settings

[Hard_ware]

Intention to buy an Intel 12700K CPU and mainboard MSI Z690 tomahawk DDR4 Wifi.

The CPU power ist needed for around 1-2 days every other 14 days.
Otherwise it is very important for me to have a very, very quiet PC with fans barely noticible. Target is 700 rpm for silent wing 3 fans in case and Dark Rock 4.

1. Idea: I wonder if PL2, PL1 and TDP can be reduced in Bios settings for MSI Z690 tomahawk board to around 100W to prevent draw spikes?
2. Other Idea if idea 1 is not possible: Set a 12700 non K (normally PL2 of 180W and 56s) to TDP 65W and PL2 to 100W for these 56s?

 

[Karadjgne]

Sound is just vibrations of frequency. Everything that deals with a frequency inside a pc make sound. With that sound you get harmonics and transmission. It's a complicated mess, but what it boils down to is that hard surfaces will refract, amplify, transmit frequency and harmonics. Especially case panels.

It's a fine edge to walk, having little enough inside a case to keep good airflow, yet at the same time have enough inside the case to create opposition frequencies that cancel out. That's the job of baffling. Not to absorb sound, but refract it in opposition.

Silent cases use a thin layer of foam rubber on the panel surfaces, that's a damper on transmitted vibrations, like putting a hand on a drum face. But the frequencies still exist on other vectors and will eventually head straight out places like fan ports or affect things like hdd cages, ssd surfaces or other hard surfaces.

If you think about a subwoofer, just the speaker itself sitting on a table, and you put a 60Hz waveform into it, you get almost nothing out. As the sinewave causes the speaker to go from positive to negative position, that generated sound refracts from the table and you get the positive bouncing off the negative frequency, which totals zero, cancels itself out. Stick that sub in a sheet of wood and the 2 sides of the frequency are seperated and you get a ton of sound instead.

Slap a couple of small foam pyramids around the inside of the pc, those baffles will change the vector of the frequency, until it cancels itself out, and takes the harmonics with it.

You'd have to experiment with placement because a positive hitting a positive is an amplification, not a cancelation. The exact distance from source determines whether or not you cancel the sound or amplify it. Even as little as 1/2" change can make a difference.

 

[Hard_ware]

Sound is just vibrations of frequency. Everything that deals with a frequency inside a pc make sound. With that sound you get harmonics and transmission. It's a complicated mess, but what it boils down to is that hard surfaces will refract, amplify, transmit frequency and harmonics. Especially case panels.

It's a fine edge to walk, having little enough inside a case to keep good airflow, yet at the same time have enough inside the case to create opposition frequencies that cancel out. That's the job of baffling. Not to absorb sound, but refract it in opposition.

Silent cases use a thin layer of foam rubber on the panel surfaces, that's a damper on transmitted vibrations, like putting a hand on a drum face. But the frequencies still exist on other vectors and will eventually head straight out places like fan ports or affect things like hdd cages, ssd surfaces or other hard surfaces.

If you think about a subwoofer, just the speaker itself sitting on a table, and you put a 60Hz waveform into it, you get almost nothing out. As the sinewave causes the speaker to go from positive to negative position, that generated sound refracts from the table and you get the positive bouncing off the negative frequency, which totals zero, cancels itself out. Stick that sub in a sheet of wood and the 2 sides of the frequency are seperated and you get a ton of sound instead.

Slap a couple of small foam pyramids around the inside of the pc, those baffles will change the vector of the frequency, until it cancels itself out, and takes the harmonics with it.

You'd have to experiment with placement because a positive hitting a positive is an amplification, not a cancelation. The exact distance from source determines whether or not you cancel the sound or amplify it. Even as little as 1/2" change can make a difference.

Thank you very much for the explanation: I will experiment with your suggestions in case my build will not be quiet enough:

To avoid this, my approach is to not produce noise at the source. This will prevent that I have to look at other more cumbersome possibilites to kill noise
With a Dark Rock 4 cooler and 3 silent wings 3 fans for the case spinning at around 700-800 rpm the PC should be barely audible (I have no experience with AIOs - will they be even quieter than silent wings 3 at 700-800rpm?) My experience from my old build is that I my CPU can draw around 80-90W until the fans will be audible. Therefore with better fans the sweet spot for non audible could be at around 100W.