www.neumann.com
Forums:
Email:









Forums / Archive / Thread
In favor of coloration...
Author: Posted by dan.dan
Date: 03/16/2004
Hi All,
I was just reading a detailed post by Mr. Sikkema about his observations of coloration in different microphone preamps through some A/B tests, etc.
One thing he mentioned was that transformer based, discrete preamps color the lows, resulting in a perception of 'warmth' and tend to also be tilted toward the highs, emphasizing sibilance to some degree, and that most people prefer those traits.
Without arguing whether specific preamps do or don't add such coloration, I thought I would raise the topic of coloration in general.
One way to look at audio processing is that it should be as faithful as possible (high fidelity), approaching, as closely as it can, the ideal of a unity transfer function.
Another way to look at it, the artistic way, says that whatever sounds the best, subjectively, is what ought to be used, regardless of the specs.
I think there is a whole continuum between these two extremes as well, and I personally think both perspectives are equally valid.
As I've pondered this subject, though, it has occurred to me that we can identify certain specific reasons why, in some situations, coloration is really desirable, and even necessary. Doing so serves not only to validate our desire for 'gourmet' type audio devices (preamps, vintage microphones, etc.), but also to help us understand why coloration is more appropriate for some chores than others.
So I thought I would share a couple of items from my own growing list of why coloration is necessary in recording, hoping to spark intelligent discussion and constructive debate. These are all just my opinions and observations.
1. Mic proximity: In cases where the mic is placed closer to the source than any listener would ever be, the sound that the mic 'hears' will not have a 'natural' quality.
Perhaps the best example is a vocal. With a vocal mic a few inches from a singer's mouth, the balance of transients and sibilance vs. vocal tone is seriously skewed, and highly accurate mics will sound 'choppy' and 'harsh' because they capture what's really going on there. Dynamic mics or large diaphragm condensers are preferred because their more sluggish time behavior serves to de-emphasize 'sharp' transients, giving a more smooth, natural sound. This also applies to close miking drums, plucked acoustical instruments like guitars, etc. Of course, this can be dealt with in one of two ways: choose a suitably colored mic, or move the mic further back from the source. But the latter is undesirable or impractical in a lot of cases, leading to the need for mics whose time behavior is deliberately imperfect.
2. Volume translation: In a large number of recording situations, the playback volume will be very different from the volume of the original source. A drum kit miked in close proximity exhibits peaks of 120 dB SPL or more when played hard, but virtually nobody will actually listen to the final mix so that the drums are really that loud. An interesting challenge with these kinds of sources is to get them to sound clean and clear and punchy and powerful at low volumes. In the end, it turns out that a flat frequency response is unsuited to accomplishing that feat in many cases.
One thing I've noticed over the years is the phenomenon of 'it sounded great in the studio on the day we recorded it, and then I got the final mix and it sounded lousy in my car.' One of the main reasons for this is often a tendency to turn it up until it sounds good in the studio. When monitored in a control room at 80 dB, something like a drum kit sounds bigger and better when you leave all of the warmth and dynamic range in it, but that doesn't translate well to low volumes on inexpensive playback equipment with background noise. When you turn it way down, suddenly it sounds dull and muddy.
Personally, I think of it like this: The challenge is to make the mix's 'size' or 'scale' match the lower playback volume. It has to be scaled down, which often involves boosting presence, dipping low mids, compression, etc. It can be difficult for the artist, who walks from the studio into the control room and suddenly hears it coming back sounding much smaller in comparison to what they heard in the other room. Audio perception is very easily skewed by comparisons and relationships.
But one reason a lot of producers like mixing on little, boxy monitors like NS10s is it forces them to think on a smaller scale with the mix, which yields results that translate much better, and also still sound good when you turn them up.
Those are just a couple of reasons why perfectly accurate signal chains are not always suitable. In many cases we're not really trying to accurately capture an acoustical event (though, in some cases, we are).
Comments?
Regards, --Dan  
Posted by dan.dan
03/16/2004