Should I be concerned about this?

  • Thread starter Thread starter Middleman
  • Start date Start date
Middleman

Middleman

Professional Amateur
I was doing some background reading on monitors and why they all sound different and came across the concept of measuring their response with pink noise.

Do sound cards have their own sound shape too? I decided to download a pink noise wave file out on somebodies site. I plugged this into Sonar along with a spectrum analyzer plug-in to see what was happening by just making my card play the pink noise and seeing the spectrum curve.

It looked like a straight downward slanting line from left to right about 24db higher around 50HZ than at 15KHz. Does this mean that my Audiophile sound card has a tendancy to emphasize bass frequencies over high frequencies? Should I automatically compensate with EQ to try and get the channel to be flat on all audio files I record?

I am more musician than tech oriented so if anyone has any thoughts on this I would appreciate them.

Possibly I should just keep turning the dials until it sounds good and ignore all this.
 
OK, I'll admit to having a few drinks before signing on tonight...

but, what does your sound card have to do with this analysis? If you are importing a wave file into Sonar, and you are analyzing the file with a plug-in in Sonar, where exactly does your sound card come into play in this equation?

Seems to me the only thing the sound card is doing is converting the digital file to analog and feeding it to your speakers so you can hear it. Your sound card is affecting what you hear, but I don't see how it is affecting the spectal analysis of the wave file.

Maybe I need another drink :)
 
Well I'm not sure...

The card is taking a file, which I assume is an accurate pink noise file, meaning all frequencies are weighted the same. If I put this file in Sonar and put a spectral analysis scope on the track, I should be seeing a flat line of frequencies because they are all supposed to be the same volume.

This is not the case however as the file is weighted to the lower frequencies. So the bass ended response must be coming from either Sonar's interpretation of the data, or I assumed the way the soundcard is reading the data.

OR maybe pink noise is supposed to be weighted on the low end.

One more beer and this will all come clear...
 
Middleman -

I have no expertise on this at all, but it seems to me your problem is rooted in your assumption.

I believe white noise is weighted equally across all frequencies, whereas pink noise decreases in power as frequencies icnrease. According to the below information, the drop is about -3db per octave (IOW, something like what you are seeing in your spectral analyzer :) ).

BTW, I still contend that none of this will tell you a gosh darned thing about your sound card.

Analog pink noise
Pink noise has equal energy per octave instead of equal energy per frequency like white noise. In other words, its energy is equal to 1/f, which describes a -3dB/octave response. This means that each octave of increasing frequency should contain half the power of the preceding one.
 
I believe white noise is weighted equally across all frequencies, whereas pink noise decreases in power as frequencies icnrease. According to the below information, the drop is about -3db per octave
Click to expand...
Correct, sir. You may move on to the next level! :D

But why not use a white noise-file? When I'm testing measuring-equipment I use white noise, and then see if the curve is absolutely straight. I don't know where to get a white noise file, I have a box to produse it...
 
Thanks dachay2tnr

I had pink reversed with white in my thinking and thus the results I see on the analyzer are correct.

My concern is basically that most of the gear I am putting together is under the $300 price point. i.e. Audiphile card, Studiophile mixers, C1 mike Sonar etc. I was just trying to make sure that the signal that comes in from the outside is not being dramatically altered either through the bit interpretation of the Software or the characteristics of the hardware. Otherwise I will be constanly altering the sound with EQ across everything I do.

Thanks for you help. I'm buying the next round when you come to CA.
 
Seems to me that if you want to test your sound card, you need to record a known signal (e.g., white noise or pink noise) and then analyze the recorded signal with your spectral analyzer. Using a pre-recorded wave file is really only testing your spectral analyzer.

As moskus was implying, if you generate white noise and record it, this will give a representation of the signal path to your recording software (mic, preamp, sound card, etc.) and whether the sound is being colored.

Of course, you could simply playback your existing pink noise file and re-record the output. The downside of this approach would be that now you would also be adding your playback speakers and the D/A conversion of your sound card to the equation. Too many variables to my taste.
 
You must log in or register to reply here.
Back
Top