Accuracy vs. musicality

Musicality or accuracy?  Another word that has come up is “quality”

What is musicality?  What is accuracy?  What is Audio Quality?  Here are some references.  And of course, always refer to Linkwitz. 

https://www.stereophile.com/content/whos-right-acccuracy-or-musicality

https://www.audiocheck.net/audiotests_matt.php

https://www.scribd.com/document/227415732/AQT-A-New-Objective-Measurement-Of-The-Acoustical-Quality-Of-Sound-Reproduction-In-Small-Compartments

When recording engineers are being taught how to listen to find the right mic for a particular instrument they begin by listening to and comparing the sustain of the real instrument to the mic’d version of the instrument.  Here they hear the tonal quality of the real instrument and how the mic changes the tonal quality.  This kind of is in the direction of accuracy, accurate tonal reproduction.   Makes sense, linear recording and playback sounds musical.

But no, that’s not all there is to it.  The recording  engineers are next taught to carefully listen to the attack transient that precedes the sustain.  One might think of the attack transient as the sonic presentation of the sustain.  It’s where the sustain comes from.  The attack is very short lived, a fleeting event compared to the long winded sustain.  It’s hard to concentrate on and capture such a short lived event in order to evaluate mics but it can be done.  Curiously, out of 6 seemingly identical mics, only one might come close to actually capturing the true sound in the attack part of the tambourine.  While all of them fairly equally represent the sustain of its little cymbals.  Musicality is rooted in the clarity of the attack transients. 

And so, the audiophile might want to look toward the faithfulness in the attack transient instead of the faithfulness of the sustain as a clue to musicality.  There is no problem hearing the attack transient if silence precedes it.  The problem is to hear the attack transient within a musical piece, which is filled with many other attack transients.  There have been studies relating to the speed of separate sonic events in music and is speech.  The overall average rate of man-made sonic events is 8 separate events per second.  Each sonic event begins with some kind of attack transient which evolves into a sustain.  The attack transients occur so frequently that the much longer lasting sustains become overlaid one on another, creating a set of running complex and ever changing chords that tag along with the music, fitting and blending everything together.   

Ideally we will have 10 dB of quiet immediately preceding each attack transient.   In a large hall the reflections of one attack transient are delayed and weakened by diffusion so as to not interfere with the next attack transient and the sound sounds musical.  But in small rooms, our hi-fi listening rooms, the very early multiple reflections of the attack transient are quick and strong and they fill up the desired succeeding quiet and overlay themselves onto the low level details of the next attack transient, obscuring the listener’s perception of the next attack transient, aka sound masking.   

A test that measures the tonal clarity has been developed, actually back in the golden days of hifi, back in 1986.  It is called MATT, Musical Articulation Test Tones and is a gated sine sweep, 8 tone bursts a second, 1/16th second burst followed by 1/16 second silence.  The clear Ta-ta-ta of ascending and then descending tones is audible over headphones.  But in most rooms, yodeling turns into gargling. 

This means that the sounds are where they should be in time but the articulation within each sound, the dynamic aspect of each sound, is lost to the excessive sustain produced by poor, undeveloped room acoustics.  To get a fast (musical sounding) room you have to seriously quiet down the front 1/3rd of the listening room, the area where the sound is being created so you mostly only hear the direct signal from the speaker and the instantaneous ambience or running reverb is at least 10 dB below the direct at all times.     

So, what we see is people choosing speakers that can somehow penetrate the running reverb noise floor.  Speakers with overly bright directional top-end might help or those with large midrange horns are very effective in delivering a strong direct to running reverb ratio signal.  And yes, dipoles are excellent at this because they do not project sound vertically nor laterally, but only forward and backward.  It’s the component of sound projected vertically which is then caught and stored between the parallel floor and ceiling surfaces, early vertical reverb buildup, which lingers too long and floods into the next 1/16th second of desired quiet.  Same with side to side and the subsequent lateral early reverb buildup.   Notice woofers and sub woofers are running totally out of high speed dynamic control. 

Musicality is in the record and it’s in the signal being played by the speakers.  The problem is not the equipment it is the room obscuring, literally masking, our ability to hear successive attack transients which means that all we get to listen to is the running aggregate set of sustains left over from the preceding rapid set of inaudible (masked, drowned out) attack transients. 

Down load a MATT test for free and play it over headphones and over your sound system.  Audition the musical clarity response curve of you audio system, remembering that the last link in our sound systems is the room acoustic link.         https://www.acousticsciences.com/products/matt-acoustic-test-cd