Golden or Leather Ears?
Based on a post on the theater sound list I decided to undertake a bit of an experiement, though hardly scientific. Which is my point in my response. Many times, if not the majority of the time what we do is more a factor of the application rather than the science we use to do this sound thing. While we need to have a good understanding of the mechanics of audio, we really need to know how they fit into the big picture.
The discussion was regarding the parallel spliting of microphones with regards to sound reinforcment applications. Jim Brown and Bill Whitlock presented AES 118 in Barcelona “A better approach to passive microphone splitting” (paper 6338) Full text is available at http://www.audiosystemsgroup.com/AESPaperSplittersASGWeb.pdf
While the research presents compelling arguments and data for reasons not to Y split mics it fails to address practical and business concerns and offer a comparison of real world parameters aside from what is very comprehensive test data. In other words, plugging a mic in and using it in an app without all the test gear and using perhaps the most important audio test gear, human ears. My point is that even with the science, in a real world SR application one will be challenged to tell in the majority of cases which mic is Y split and which mic is transformer split. The paper does an excellent job of explaining the phenomenon though it doesn’t address the use of lower cost transformers when used degrade or change the audio more than a Y split would, particularly on the low end with high amplitude apps. To me that’s been much more noticable than Y spliting. Also, the current breed of active mic splitters add thier own signature to the sound, offsetting many times the subtle differences in esoteric mic pre amps. In practice I’ve found the results of using active mic spllitters and inexpensive transformers to impact my app more than the results of Y split mics. Given the same level of other equipment, talent, environment, etc.
I had hoped to make a more comprehensive comparison but due to time and other restrictions, I won’t be able to at this point. This weekend I trade my polyester blend polo shirts from the swanky dive on the Strip for some tee shirts with a picture of a yellow sun at another dive down the street where they do the same sort of thing. This time, though, instead of flying 90lb french chicks through the air then dunking them in water, we’ll turn the beach into a forrest and then into a mountain and fly 90lb asian chicks through the air but with no real water to dunk them into.
Here’s what I did for the mic split comparison. Using a PM1D V2 with the new mic preamp cards, supposedly based on the XL4, the LMY4-MLF or as we call them, the MILF cards. I used a Neuman KMS-104 with 100′ cable and split it with a standard Y into two channels of the PM1D. I routed the preamp out (pre HPF) of each channel out of a DIO MY4-AT (Lightpipe) into a Digidesign 192 I/O feeding a Pro Tools HD2 system @ 24 bit/44.1kHz. One file is the direct, unsplit sound of the mic. On the second pass two mono tracks recorded the indentical source concurrently in real time and the mic in the second track is used. The most difficult thing for me was keeping the mic technique consistent between tests.
I’ve uploaded the files at http://roaddog.com/video/split_test/file1.wav and http://roaddog.com/video/split_test/file2.wav . I’ll leave it to our gentle readers to decide which is which or even if so inclined to run a transfer function test to see any differences or if that test matches the differences in what one preceives between the two files.
EDIT: I’ve opened comments for a while so if you have anything to say, post it.
Dave