Originally Posted by marvin100
So Mark, am I right, then, that you agree with the conventional wisdom that binaural recordings are really best with headphones and don't offer a significant improvement over loudspeakers? You really seem to know what you're talking about and have a lot of experience, so I'm curious what you think about this.
No, I'm not saying that (see post #38 or #39 - I can't recall which number is correct, but it is one of mine).
What I am saying is this: From an aesthetic point of view, whatever tools one chooses to use to create a specific sonic image are valid. Thus, if you want to use binaural as part or all of a speaker-based mix and the Jeklin, XY, ORTF or whatever as part of a headphone mix, then do it. To say that a purely binaural recording should never be played over speakers is like saying that a painter should not use an orange peel as a brush because it isn't a brush; if that particular "brush" helps the artist achieve his or her artistic vision, then it's the right brush
for that particular job - even if when someone says 'paint brush', an orange peel does not come to mind.
However, from a perception
standpoint, or rather, accuracy of perception then yes, I agree that a purely binaural recording is best heard over headphones due to elimination of crosstalk etc - but this also depends upon the acoustics of the space to some degree.
Let me explain. In the product development sector (those people who test the way a car door slam sounds, or a powered sunroof, or an accessory switch sounds...or for that matter, the sound of that hyper-expensive dishwasher, hair-dryer...or whatever) binaural recording and playback (over headphones) has been the de-facto approach since the early 1990's and in some locales and industries, even sooner. This is because binaural is best at capturing the acoustics of a space (assuming that headphones will be used for playback), and since the acoustics of a space play a major role in what we perceive, it is critical that the jurors (who listen to such sounds) have the most accurate playback so that it mimics what the item would sound like were they seated in the car, or standing in front of the dishwasher.
Put another way, no other typical recording technique (other than binaural) has been adopted by the test, measurement, and instrumentation community for human perception - that is, I have seen precious few test specifications that equate binaural to a stereo technique. For example, you will never see a test specification that reads "place the binaural mannequin head (or equivalently, the XY or ORTF pair) at + 100 mm (x), +240 mm (y), and - 30 mm (z)".
There is a very good reason for this, and it has to do with the spatial and temporal accuracy that binaural affords the recording, and yes, when these product sounds are evaluated, the jurors are all wearing headphones because it is understood that in assessing human response to stimuli, how we respond to stimuli depends upon how the stimuli are presented
to us - this tenet is of paramount importance. I'm dead serious when I say that organizations like car companies, appliance manufacturers, or even hand-tool companies have spend enormous sums of money on labs designed to support and execute this sort of acoustical testing and analysis, as well as human perception to sound - and all of them that wish to be taken seriously rely upon binaural recording and playback as part of the design process.
So again, one must consider the aesthetic, preference, the physics, and the math and in the end, use what gets the sonic vision across to the intended audience.