Tuesday, December 22, 2015
Friday, December 18, 2015
I’m not even close to being the first guy to gag on the bullshit marketing terms applied to non-existent audio equipment qualities.
Immersive: As in “I felt the music was all around me. . .” (Other bullshit terms used for this generally non-existent quality are “open,” and “air.”) Bob Carver made a nasty audiophile device called the “Time Lens,” back in the 80’s. If you like “immersive,” you’d love this thing. This awful device screwed with mid and upper-mids L/R phase relationships so severely it was hard to locate any sound sources in some rooms. Carver had some really fancy explanations for why sound seemed to be originating everywhere but from the speakers, but simple test equipment (an oscilloscope) demonstrated the fact that frequency-dependent phase relationships were being messed with and the result was semi-entertaining until it gave you a headache. A surround sound mix can be “immersive,” if the speakers and listener are properly placed. Otherwise, stereo means “solid” which will always put the listener in a position facing the sound sources and, at best, a good mix and proper playback system will provide left, right, close and distant perspective. And that’s it. If you hear signals behind you, your system or room acoustics are screwed up.
Warm: (aka “creamy,” silkiness,” and “natural”) Ideally, you might hope this means some mild addition of even harmonics—mostly 2nd—distortion contribution that might seem pleasant. Usually, what most people call “warm” is the result of slew or crossover distortion and is in no way pleasant if you are used to listening to low distortion reasonably accurate sound reproduction. Distortion-wise, we tend to “like” what we’re used to hearing and if we spend enough time with low fidelity equipment and recordings we’ll learn to love that. That is not, however, a good thing.
Presence: Three decades ago, I had the pleasure of getting to mess with a Mark Levinson stereo power amplifier when I was a design and test engineer at QSC Audio Products. The Levinson claim to fame from advertising and reviewer hype was “presence.” hen I put a Mark Levinson amp on the bench, what I found was a slowly rising upper-midrange shelf that peaked at about 3dB at 3kHz (along with the associated phase distortion created by this built-in EQ). Otherwise, there was nothing special about this overpriced equipment in any regard. It wasn’t quick, it wasn’t particularly stable when super-sonic signals or reactive loads were introduced, and it wasn’t particularly low distortion (IMD or THD wise). A few moments with some simple electronic test tools would have dispelled the magical claims from a power amplifier with a silly built-in EQ, but audiophile magazines shy away from test equipment and objective listening tests.
Wide Dynamic Range: Dynamic range is a fairly straight-forward specification in audio equipment: “the ratio between the largest and smallest values of a changeable quantity.” At the most impractical theoretical extreme, the human ear is capable of detecting 140dB of aural signals. That assumes 0dBSPL as the threshold of hearing, which quickly deteriorates from infanthood to adolescence with age and noise exposure to more practical values of 15 to 30dBSPL. This also assumes the maximum sound pressure a human can tolerate and discern is 140dBSPL; a dubious claim at best. 140dBSPL is, for example, typically described as a full-throttle jet engine at 50 meters. A typical gun muzzle blast generates an impulse sound pressure level of 140dbSPL or more and that exposure is commonly known to cause permanent hearing damage. Relative to 1V (0dbV), the theoretical perfect (but practical) unity gain amplifier will produce 123dB of dynamic range. This can be bumped up to about 129dB if the relative maximum is 10VRMS. So, regardless of the unrealizable human hearing maximums, 123-128dB of dynamic range is pretty much as good as it gets in the real world. This relatively straight-forward specification and its associated practical limits are regularly ignored in the bullshit marketing world where numbers as insanely silly as “200dB” are quoted regularly and stupidly. Keeping in mind the analog thermal noise threshold, it’s useful to remember that every combination of amplifiers that produces an exponent of 2 (2, 4, 8, 16, 32, etc.) will reduce the combined dynamic range by 6dB. “Floating-point audio” or 64-bit or any other bullshit aside, physics and biology put an end to the realistic gains we’re going to get from playing with numbers.
This list of bullshit words is not by any means complete. I’ve left out color and taste sensory analogs, for example. A good rule to apply to modern ad-driven reviews and marketing claims is “any asshole can write marketing bullshit.” There is no one checking these claims with test equipment or even half-decent ears. So, if it sounds too good to be true, it is.
Saturday, December 5, 2015
Friday, December 4, 2015
In a possibly hopeless attempt to convince a friend that her vocal group could benefit from a little microphone application knowledge, I went looking for some artwork to help with my descriptions. Luckily, Sterling Audio had exactly what I needed. The illustration at right is one of the points I was trying to make. My friend’s group is a fairly large women’s vocal group and their bandleader semi-wisely selected a collection of six low cost large-element condenser microphones with which he lines up in even spacing across the stage, randomly distributing the microphones across 11 performers. The original intent was something like the above illustration. Obviously, 11 into 6 produces a significantly different dynamic, but the flawed concept is the same.
Performance-wise, the problem is even more dramatic. During a recent show, I solo’d up each vocal mic and discovered that the vocalists could be counted on one of two possibilities: 1) some singers naturally gravitated to grouping around a mic and singing “together” and 2) some naturally filtered to an abandoned mic and sang “solo.” If you’ve never sang in a group this might surprise you, but if you have this will be a no-brainer: the solo vocalists were consistently off-time and key and the together vocalists consistently delivered stronger, tighter, more in-key performances. So much so that I found myself naturally fading-out the solo mics and blending the group performers by ear, as I pulled down disagreeable faders and pulled up the performances that blended well.
Most recordists think the best way to record any signal sources more complicated than one voice is to distribute as many mics across the various sources as possible. Hell, hip-hoppers put two mics on one voice to optimize phase distortion and tonal weirdness. Here’s where I disagree with “common knowledge,” regarding the best way to record/reproduce/reinforce a group vocal. Sterling Audio’s article suggests, “There may not always be enough resources for separate microphones or recording tracks” as a reason for simplifying the number of microphones involved. I think the recordist better have a damn good reason for using more than a stereo pair in this kind of situation. “Depth of field” comes from the various time delays caused by the distance of the sound source from the mic. Vocal groups, horn players, and ensembles are often perfectly capable of figuring out where they should be, physically and sonically, in the mix and allowing them to perform for the microphone as a listener often produces incredible results.
The fascination modern recordists have for vintage recordings, oddly, doesn’t seem to extend to vintage recording techniques. Due to the lack of console and record media channels, the old-school guys were forced to learn something about microphone polar patterns, acoustics, and music. Today, “more is better” and too many recordings have the spacial characteristics of a pair of headphones.
POSTSCRIPT: A friend and ex-student, Steven Sullivan, took all of this to heart and produced a record limiting himself to no more than 8-tracks and very few effects.