what is SIGNAL TO NOISE?

[Adi Head]

what exactly is signal to noise and how do i check it on lets say a pd150?
what other tests can one perform on a camera/mic to evaluate it's audio quality?

[Mike Rehmus]

Signal to noise is the ratio of the noise of the system with the input shorted and compared to the maximum signal level.

16 bit DV cameras have a theoretical 96 dB of signal to noise. But that uses 0 dB as the maximum. Clearly an unsafe place to be with digital recording. So we run at -12 dB which allows a theoretical 80 dB signal to noise ratio.

Since even the best of the camcorders have noisy electronics, the nominal S/N is usually far below 80 dB.

You need some reasonably sophisticated equipment to make these measurements. And when you get done, the value is only one element that contributes to overall sound quality.

In reality, IIRC, almost any professional microphone will have a better S/N than the 2000 or 150. In the real world, that rarely is an issue. Background noise masks the noise floor in most cameras.

BTW, everything in an audio chain creates some noise. And the noise is additive. So we all go to great lengths to get quiet equipment. And then throw a lot of the advantage out by improperly setting the recording level.

[Adi Head]

thanks for the reply. how exactly do i check signal to noise?

[Mike Rehmus]

You need to obtain test equipment to do that. I'd guess you could rent test equipment for $500 - $1,000 that is specifically designed to test that.

A computer program might be available to make a test. I'd think that a Firewire transfer would transfer the sound's original values. Where you find that, I don't know.

Just out of curiosity, what are you going to do with the information when you get it?

[Adi Head]

i was just wondering if there is some simple way i could test that on sound recording devices. reading some material regardig audio, it seems to be an important factor when evaluating sound recording technology. if it's not something i can measure easily or hear, i guess this information is useless to me for now, except for knowing more what they mean when using the term. thanks mike.

[Adrian Douglas]

Any manufacturer of quality audio and video gear will usually test the S/N ratio and include it in the specs of the piece of equipment. It is usually indicated as a decibel (db) value.

[Mike Rehmus]

And, like the power of amplifiers and the sensitivity of camcorders, it is a value that is very open to interpretation.

Unfortunately, there are no really good and easy ways to make the measurements. It costs a lot of money to buy or rent the test equipment and then you need to know how to operate it. Lots of pitfalls for the undereducated.

The S/N ratio of a single piece of equipment in the audio chain isn't really going to tell you much with regard to predicting how the sound will finally, well, sound. Because every pieced of equipment in the chain, including the cables and the connectors contribute to a worsening of the S/N of the audio. Then the real-world contributs a lot more noise and even predicitons go out the window.

So the best way to test the outcome is to run some trials and see what you like and don't like. Then be prepared to have your opinion of gear and performance levels change as you gain more experience.

Three years ago, I would have told you that a Sennheiser ME66 is a perfectly good microphone for recording narration and recording it on a camera. Now I know that a reasonable studio microphone and a md recorder works and sounds better. I just hope I don't get to the point where I need/want a DAT or hard-disk recorder and a $5,000 studio microphone.