TR Forums

I understand somewhat what SNR is, but I'm confused as to what it applies to. In order to get the advertised SNR, does the audio being played back need to be recorded and saved as a file with that SNR as well, or does any file play back with the advertised SNR? I'm using an ASUS Xonar Essence ST with an SNR of 127 decibles, and Klipsch ProMedia THX 2.1 computer speakers. Not sure what the SNR is of the speakers.

Signal to noise ratio is the ratio of signal (what is meant to be heard, like music) to noise (interference from things like the power supply or your microwave oven that mess up the audio signal). If the signal (your CD track, for example) was recorded poorly, there will already be noise and distortion in the recording that cannot be fixed by good player/receiver/speakers. Typically, though, professional audio recordings are quite good and it is usually on the playback end that noise gets introduced. That's where a high quality playback device (like the Xonar) comes in. SNR is one of many measurements that can determine how clean a recording is played back, but it's a popular one.

Theoretically, every time you play a sound on your Xonar it's being played back with the specified SNR from the Xonar itself. Once the signal exits the sound card, you're at the mercy of the amplifier and speakers.

SNR applies to sound, but it also applies to any field where one wishes to measure the ratio of some thing that is desired to some kind of interference.

EDIT: for a good example of the production pipeline, think of a football player during a game. At every stage of the game, his jersey is going to get mud slung onto it. The amount of mud that gets slung onto his jersey is akin to the SNR -- the more mud on his jersey, the lower the SNR, i.e. what you intend to be there is blotted out by interference. If there's enough mud slung during the game you won't see his jersey by the end; it'll be all covered in mud. Just like mud gets slung around from the start of the game to the finish, the same is true with signal noise. Noise can be introduced at the recording stage, encoding, decoding, transport, playback, etc. stages. If there's enough noise at any one step or combination of steps, the sound quality gets really bad. The Xonar's high SNR means that it is not going to sling much mud onto your music.

BlaaaAAAAHhhhHBVpoi a;lskkj! Fjkeooskd! vjkds;lak*(kjf2@33 50kdjf

THere, I've just raised the SNR of this thread by 33%.

FireGryphon's got it. Just to boil that down a bit...

S/N ratio tells you how much noise a given component will introduce. A S/N ratio of x dB means that any noise introduced by the component will be that many dB quieter than the desired signal. Typically this is quoted at maximum output level; at lower volume levels the S/N ratio may or may not be worse, depending on how late in the component's signal processing chain the noise is being introduced.

The S/N ratio of your soundcard only tells you how much noise the soundcard may introduce. It says nothing about the S/N ratio of the source material you're playing through the soundcard (e.g. audio which was originally recorded on analog tape will typically have detectable levels of tape hiss, and devices connected to the soundcard via an analog connection may pick up hum or other interference via their connecting cables), or any electronics after the soundcard in the signal chain (e.g. noise introduced by the electronics in amplified speakers). The end-to-end S/N ratio will likely be worse than the S/N spec for the soundcard alone, because the noise adds up.

drsauced got it backwards. He actually lowered the S/N ratio by 33% (but now it is only 25%).

Well, good thing I previewed before posting, because JBI hit all the points I was about to.