Members veracohr Posted July 11, 2014 Members Share Posted July 11, 2014 Has anyone ever done any perception testing on audio gear distortion figures? As in: attempting to hear differences between Mixer A and Mixer B, which have different distortion specs? I'm starting to design a line mixer for myself, so I can have exactly what I want and none of what I don't need, and I find myself getting caught up in distortion and noise figures specified in opamps and other ICs. My inclination is to go with the lowest numbers possible, but there must be some point of diminishing returns right? How much can one hear a true difference between .005% distortion and .01%? They're both pretty bleeping low! The lower distortion and noise figures I go, the higher the cost becomes. The past couple years I've been so immersed in my EET schooling that I've drifted away from real-world audio. I fear I've become more number-centric than practical. Link to comment Share on other sites More sharing options...
CMS Author MikeRivers Posted July 11, 2014 CMS Author Share Posted July 11, 2014 Different kinds of distortion are perceived differently. At an AES panel entitled "Lies, Damn Lies, and Specifications" I played an example played through two signal paths,, each of which measured 5% THD, which is always the first and often the only distortion spec you'll see published. The one with a bit of clipping (primarily 2nd and 3rd harmnics) was obviously distorted but not unpleasant to listen to. The one with crossover distortion was quite nasty. Intermodulation distortion is another can of worms. If the modulation source is related to the program source, it may be hard to notice, but if it's not related, like for instance a stray oscillation (bad digital clock jitter is an example), a lttle goes a long way. Test your design thoroughly to see what's actually happening when you put signals through it. Link to comment Share on other sites More sharing options...
Members 1001gear Posted July 11, 2014 Members Share Posted July 11, 2014 Hi, just wondering...THD is any harmonic deviation (?) and is this what a device's sound and coloration are? Link to comment Share on other sites More sharing options...
CMS Author MikeRivers Posted July 11, 2014 CMS Author Share Posted July 11, 2014 Hi, just wondering...THD is any harmonic deviation (?) THD is usually measured by putting in a single frequency test signal (a sine wave) at the frequency you're measuring (you should test several over the audio range), filtering that frequency out at the output, and measuring what's left. If there was no distortion, you should have nothing left. A true measurement of harmonic distortion is made by measuring and summing only the amplitudes of frequencies that are harmonics of your test signal, excluding anything else present in the output that you didn't put into the input. Usually there are some things coming out that don't go in like hum and noise. Since anything you get out that you didn't intend to is considered noise, we usually measure THD+N (total harmonic distortion plus noise) since that crud contributes to the total number that we can measure. It's useful to look at the spectrum of what comes out so you can see what harmonics are present, and also see an unrelated frequency (like maybe 60 or 120 Hz) that's contributing to the THD number. and is this what a device's sound and coloration are? "Sound" and "coloration" are non-specific terms, but, yes, harmonic distortion is one of the main contributors to what, to our ears, sounds "colored" or has a particular "sound" If we like it, we call it "coloration." If we don't like it, we call it what it is - "distortion." Link to comment Share on other sites More sharing options...
Members 1001gear Posted July 12, 2014 Members Share Posted July 12, 2014 Question answered. Thanks. Link to comment Share on other sites More sharing options...
Members blue2blue Posted July 13, 2014 Members Share Posted July 13, 2014 Mike's a real good explainer. Link to comment Share on other sites More sharing options...
CMS Author MikeRivers Posted July 13, 2014 CMS Author Share Posted July 13, 2014 Mike's a real good explainer. And without Auto Tune, too. Link to comment Share on other sites More sharing options...
Members 1001gear Posted July 14, 2014 Members Share Posted July 14, 2014 ...and no discernible distortion. Link to comment Share on other sites More sharing options...
Members veracohr Posted July 15, 2014 Author Members Share Posted July 15, 2014 Thanks for the replies folks. I think my concerns were momentary, and I'll chalk them up to my newly-invented phrase 'DAS': Distortion Avoidance Syndrome. With a nod to whoever made up the term 'GAS'. Or I could call it 'TAS': Temporary Audiophile Syndrome. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.