snow lizard said:
Well, if it can make people think about how to approach the craft, and not necessarily take it as gospel, there are some great suggestions in there.
And if it can lead to a thread where some actual audio discussion takes place and not just a bunch of myth regurgation and/or personal sniping, then it's a worthy start.
I, too, don't 100% agree with 100% of what Ed has in these abbreviated posts. I also have been gnawing at the bit to play editor to clean up and streamline some thematic rambling in it (probably mostly caused by the fact that these posts, as "long" as they are, are actually edited highlights from an even longer previous tome.) But I didn't feel that the differences were that large or that important to bother risking an unwanted and unintended dust-up with Ed and others. For the record Ed: it's a good couple of posts. Not perfect, but no such posts are; and certainly good enough not to warrant defamation. I'll thank you for posting them

.
But at this point this thread has opened up some room for non-sniping discussion...
snow lizard said:
The 85 dB monitoring thing is important because of what the Fletcher-Munson curves say about how we hear things. We're not going to hear the same bass response at lower SPL's. I don't think there's anything wrong with working at a lower level as long as your monitors are calibrated the same for what you're mixing and for your reference material. Other people have brought up the point that they like to mix at levels they're comfortable with for the most part, but that they do check at various different levels periodically to see what's going on with the bass response....
Fletcher-Munsion is one of those topics of which everybody seems to have a general understanding, but over time the some of those generalizations have almost morphed into mis-gneralization or misunderstanding. Please excuse a very short and simplified histroy review in order to explain a point:
First off, what many books and web sites pass off these days as "Fletcher-Munson curves" are NOT actually the Fletcher-Munson curves. Fletcher and Munson first developed the idea of charting average ear sensitivity by frequency back in 1933 or so (not the same Fletcher of Mercenary Audio fame, though whether they are related or not, I have no idea

). They performed their study using *headphones* and not loudspeakers, and there were also other weakensses in their methodology, including what frequencies they actually tested at (those smooth lines we see are largely interpolated educated guesswork) and what sample base of human test subjects they used.
Since then there have been many other studies and researches yielding their own "equal loudness contours" - which is what "Fletcher-Munson-style curves" should be called, and each generating debate as to their actual accuracy. The latest concensus standard for equal loudness contours is actually the international standard ISO 226, last revised about 4 years ago, and which differs signifigantly from the Fletcher-Munson data, particularly in the bass response end of the plots. The kicker is even ISO 226 charts get mis-printed and mis-represented as often as they are gotten "right", to this day.
One reason the term Fletcher-Munson has stuck, I think, even though it has been the subject of scientific doubt and debate since at least the 1950s, is that the still-popular A-weighting curve is based in large part upon Fletcher-Munson data.
Anyway, I explain all that just to bring up the point that the equal loudness coutours, whether Fletcher-Munson, Robinson-Dadson, ISO 226, or any of a dozen other such results curves come up with in the past eighty years, are nowhere near as accurate as their scientific appearance leads us to believe. Not only are very one of those curves *averages* based upon a sampling of human ears, and therefore curves that no one pair of ears are likely to match (including yours and mine), but there is still no concensus on just how accurate those averages even are. The ISO standard is one that's been agreed upon, but it's been agreed upon just to have a standard concensus reference to work from, not because the scientific community actually says they are all that accurate inthe real world. The equal loudness contours, whether they be Fletcher-Munson or Sly-Family Stone, are not gospel for any single set of ears.
Second, the idea that 85dBSPL is the ideal listening volume is in itself oversimplified and not necessarily true for many of the same reasons given above. The fact is, 85dBSPL is LOUD. Hell, that's as loud or sometimes louder than the SPL at nominal audience range for the live bands I work with on the club circuit, and for those, I'd prefer to wear ear plugs. If I wre to work in the studio at 85dBSPL for anything more than a few minutes at a time, my ears would fatigue so fast that I couldn't get a trustworthy mix in in one session.
On top of that, there are some things that - for my ears anyway - mix better at low volumes. Vocals, for example. I much prefer to fold my vocals into the mix at normal listening volume (which for those of us above the age of 17 is *not* 85dBSPL

). In fact, I'm like Snow Lizard; I do my mixing all over the volume range. My hand is on the CR volume control on my mixer almost as much as it is on my faders or my mouse. I have read Big Boy engineers say they like doing most of their mixing at control room "conversation level".
For me, this issue falls in the same category as monitor selection. One should choose mixing volume the way they choose their monitor; based upon their ability to translate what they hear to a good sounding mix. Anything more than that is just technical posturing.
G.