Sound Sample rate and bit depth for gaming.

[clickyclick]

Hello,

I was wondering what is the best setting for sound sample rate and bit depth for gaming in control panel. Note that I mostly play CSGO.

My headset is Razer Tiamat 7.1.

Thanks in advance.

 

[hang-the-9]

Leave it at defaults will sound fine. They are usually set to CD quality for defaults.

 

[amik3]

You should set the samplingrate and depth it as high as possible. because (warning, understanding this explaination will require some understanding of accustic engineering):

1: 44.1 khz will distort your treble sound when codecs with variable bitrates are used because the actual samplingrates of the recording may be lower then 44.1 khz. This will generate spurs (possible harmonic distortion) and reduce the sound quality.

2: If your soundcard is able to upsample (creating more samples by interpolating between adjacent samples), you should use it! Shannon's samplings theorem have created a myth that have been abused by popular science. According to Shannon's samplings theorem, at least 44.1 khz allows you to theoretically reconstruct any signals up to 22 khz. This holds for any operation in the digital domain, since the ideal behaviour of (for instance) a fourier transformation of the signal makes it possible to perfectly interpolate samples given any sinusoids. On the other hand, this does not hold for the DAC as it's reconstruction filter will behave non-linearily across the frequency spectrum. A sampling rate of 96 khz for instance, will add more information for the reconstruction, and therefore rely less on the behaviour of the DAC's reconstruction filter. The fallancy here is believing that the rules of the digital domain of sound processing, also aplies to the physical domain.

To keep it short a 44.1 khz samplingrate (without upsampling) will most likely distort your treble/high frequency range of the sound you are listening to. How much distortion, and exactly over which frequency range will depend on physical characteristics of your hardware.

3: For the same reasoning as in reason 2, higher bit-depth will allow higher resolution in your low-midrange (if an approximation to smooth out adjacent samples with the same value is used). This is because the low range signals will have a massive amount of samples for each period (compared to the high range). If a sample can take more possible values, the maximum step (in voltage) between 2 samples will be minimized, and therefore reducing harmonics produced from oscillating behaviour and ringing.