Jeff Evans
Nice articles and both great. Like the Rane article as well. Bill that Rane-Gain is amazing to say the least. I cannot believe it is so simple. If I ever see a second hand one I will snap it up without fail.
The Gain-Brain was the brain child of a Syn-Aud-Con member (can't remember if it was Pat or one of the other geniuses that hangs out there.) It isvery easy to build, my first one was a home brew contraption, but the Rane version is much more compact!
Brief aside - Syn-Aud-Con has been the source of a number of great ideas over the years. They were the first to champion pressure zone microphones, the LEDE(tm) design concept, TEF(tm) which makes LEDE possible, and the infamous Hummer, which was introduced by the late Neil Muncy, and put into product form by Middle Atlantic (it, however, is not as compact as my home brew version, I really need to fix that!)
If you are serious about the engineering side of audio I can't recommend their training or their forum highly enough. It is truly the best and the brightest of our field.
Jeff EvansI believe all this gain staging works in the studio as well. Firstly the output level may vary when any mixer is showing 0 dB VU on its output. Might be 0.775V or +4 dBu at 1.23V or -10 dBV.
Yes and no. The process outlined in those articles is designed optimize a presentation system (touring or installed). It requires some additional thought to use it in a recording studio, whether it is tape (what's tape) or computer based. It isn't that the process itself doesn't work, it is that the objectives are different.
In a live sound setting we need maximum gain before feedback and clipping - and it is difficult to prioritize one over the other. Sadly they are not directly related.
In a recording system we don't have to worry about feedback, and we are generally a little more diligent with respect to the competing requirements of maximum S/N and minimum non-linearity.
The nominal or reference level makes no difference whatsoever, and that trips people up all the time. That reference level is just that, a reference, and it isn't even tied to loudness (which we measure, more or less, with VU, not dB).
The other key difference, and this is important, is that in a live sound setting we will have a target sound pressure level that is based on local ordinances and the source and the audience. While it isn't exactly unscientific, it is plagued by lots of variables! In a studio we tend to choose a target sound pressure level that allows us to hear the details, and feel the impact, and prevents fatigue. A very different set of criteria!
Jeff EvansIf you are prepared to put a VU meter over a mixer output then it too can help you line up input signals. <snip>
A VU meter is designed to represent perceived loudness. 0VU is set to match the reference level of the system (or part of the system) that is being measured. It is not a deciBel meter, and that confuses folks no end. They are related, but it is helpful to remember that they are not the same thing.
Jeff EvansIf the parts interweave and don't play at the same time, the faders can all be close to unity. Once parts start overlapping more then our channel faders have to come down to compensate. And form our mix.
This is the difference between incoherent and coherent systems. Most music is incoherent (well, my writing is often incoherent in many ways<G>). In this case we are referring to how the signals add together. If I have two sine waves and I add them together the sum will be 3 dB greater (power) or 6 dB greater (level). This is the basis of an old trick where one would measure one channel, then add the second channel but watch a VU meter (or dB meter, but they hadn't been invented yet when I learned this trick) and increase the output of the second channel until it read 3 or 6 dB hotter. Stereo tape decks are often aligned (preliminary alignment) using this trick.
If you use noise or music as the source for this test you will get varying results. When I design a summing stage I use sine waves, not terribly musical, but a good worst case.
Jeff EvansIt flows into DAW's as well. Simply monitor incoming rms levels with a VU meter during tracking and keep all those levels constant. Put a VU on a buss and you can get a great buss mix to just hit 0 dB VU as well.
It is important to remember that a VU meter does not measure RMS levels. It is close, often close enough, but a true VU meter does not measure RMS levels.
Jeff EvansAnd then the final mix also ends up at around 0 dB VU. Pick your 0 dB VU ref accordingly. -14 or -20 being two good choices.
And that is probably the biggest source of confusion!
VU is sorta/kinda related to RMS.
dBFS is a peak measurement, in spite of the fact that the decibel is defined as an RMS measurement.
And dB-SPL is a level measurement, but not truly representative of the analogous electrical levels.
What a mess, I could write a book, well, I could if I was smart enough!!!