I've been working to refine a method I developed for calibrating MIDI volume controllers, and I made a discovery Saturday night that I'm pretty excited about. I wanted to share it here in case I stumbled on something new that might actually be useful.
Basically, my method involves using a volume meter to determine the volume controller values that should produce realistic dynamic levels and instrument balances. The premise behind this method is that handling dynamics as much as possible on the MIDI side could reduce (and drastically, I hope) the amount of time and effort that would be required in the audio mixing process.
After I had to replace my sound card recently, I found that all the calibrations I performed earlier had to be redone because my new card (an EMU 1212M PCIe) posesses a wider dynamic range response than the Turtle Beach Montego it replaced. I recalibrated all the strings, and I got much better results. However, I hit a major snag when I started recalibrating the woodwinds. Here it is in a nutshell -- even though they were matching up on Sonar's volume meter, there were huge variations in loudness. Some frustration started running through my mind, so I decided to consult Wikipedia and see if I could find some sort of answer. And I found it in the following quote from the Wiki article on loudness monitoring: "Traditional methods of measuring signal levels such as the Peak programme meter, and VU meter do not give the subjectively valid measure of loudness." Bingo. This confirmed my suspicion that DAW volume meters measure signal strength, but they tell us little about the volume of the actual sounds we'll hear. That nugget of enlightenment pointed me in the direction of trying to find a loudness meter I could use in conjunction with the volume meters in Sonar. Found one I did, and it's a nifty software utility you can download for free, called the ORBAN Loudness Meter. If you'd like to check this out for yourself, just point your browser to http://www.orban.com/meter/.
Once I installed the ORBAN meter, I immediately started using it. The tale it told about the volume level problems I was having is what got me excited, because I immediately recognized the potential for a significant improvement in the realism of my calibration results. My original concept of using an objective measuring device was correct, I was simply wrong on which device to use. Now with the ORBAN meter I seem to be going in a much better direction. If you'd like to see more about my findings, click here for a "mini tutorial" I completed earlier this evening.
For those of you who are already masters in the art of audio mixing (like DPDan), maybe the ORBAN meter could still have some use. But for those of us who aren't, my real hope with my ORBAN meter discovery is that it will open up a lot of possibilities to help anyone struggling to solve dynamic level and instrument balance problems. As to my recalibrations, I have a bunch more to get through, but as soon as my new MIDI volume controller tables are complete, I'll be happy to share with whoever wants them. Can't make any promises about when that will happen, but it should be soon.
Now, I will close out here by anticipating some scepticism about what I'm trying to do. What I'm not trying to do is make the claim that a purely scientific approach can replace the art of audio mixing. I don't think such a thing is possible. Instead, what I'm after is something that I hope will make volume adjustments in the audio mix process a matter for shaping the subtler aspects of an electronic performance, rather than having to fix major dynamic level and balance problems. If what I'm doing answers some needs on the part of other forum members now or in the future, I'll be happy for the opportunity to help.