Tuesday, August 09, 2016

Classical music's new audience does not want mono sound


A highlight of my recent listening was BBC Radio 3's broadcast of the superlative Haitink/LSO Mahler 3 Prom seen above. The BBC's Proms relays maintain a commendably high sound quality, but it is still worth auditioning them with a constructively critical ear. Listening to a Prom on a monitor quality audio system underlines the presence and impact of the sound. But that impact comes at a price: individual instruments and soloists are spotlit - e.g. the mezzo in Mahler's Sehr langsam Misterioso - and although the soundstage is wide the location of sections within the orchestra is ambiguous. These may be esoteric points, but they do lead down a path that raises important questions about how classical music can reach new audiences.

A post here in 2012 touched on the up close and personal signature sound of Proms broadcasts, while a post a year later reflected on how placing multiple microphones close to the musicians has, of necessity, become standard recording practice. Specific circumstances at the Proms including different platform layout each evening, the problem of intrusive audience and ambient noise, TV coverage requiring minimal visibility of microphones - can you spot the mics in the photo above? - and the famously poor acoustic of the Albert Hall, all dictate the use of multiple and close microphone placement. But multiple microphones picking up the same sound in different locations cause phase errors, because they are at different distances from the sound source [1]. This means when the two channels are combined in mono the sound of the instrument(s) is partially cancelled out by the phase discrepancy, in just the same way as reversing the phase of one speaker in a stereo system degrades the sound significantly [2]. Which is why many of the great stereo recordings of the past such as Antal Dorati's iconic 1959 Firebird Suite on the Mercury label used very few microphones.

When I worked on radio programmes at the BBC, which is many years ago indeed, we used to switch between stereo and mono monitoring because may people back then listened in mono - remember transistor radios? If your FM tuner/receiver has a mono/stereo switch and you live in the UK, there is a simple way to demonstrate the destructive power of phase errors. In an ideal mix the sound will remain virtually unchanged when folded down from stereo to mono, except of course for the change to a single sound source. But switch to mono on a BBC Proms broadcast and the sound degrades dramatically, particularly the upper strings.

Commercial reality dictates that most classical recording is now done on the fly at concerts. But studio recording is still the norm in the rock world where sound engineers are more wised up to the importance of listening in mono. For this reason a website devoted to rock studio best practice recommends:

"When soundwaves are (partially or totally) out of phase with each other, some frequencies can disappear. A word of caution here is that stereo sounds that have a lot of phase problems between the left and the right channel can sound really wide when played back in stereo, but lose some frequency content when combined to mono. A good practice is to occasionally listen to your mix in mono, using either a mono button on your mixer/audio interface or a plugin with a mono button on your master bus in your DAW [digital audio workstation]".
Rock audiences are conditioned to mono sound because DJs play their sets in mono, as clubs do not have stereo 'sweet spots'. Because of this most club sound systems are wired in mono; in response to this leading rapper Kendrick Lamar mixed most of his award winning album To Pimp A Butterfly in mono, and he estimates that 80% of his mixing is done in mono. Mono is also staging a return in personal listening as single Bluetooth speakers used in portable systems provide a mono source, and the soundbars that are popular in home cinema systems are point sound sources despite claims to the contrary. But in classical music Nada Brahma - sound is god - and the all-important new audience will not be won over by mono sound.

I can already hear the shouts of "This has nothing to do with classical music, because nobody listens in mono". Which is wrong: because at a Promenade concert, the only people who hear true two dimensional sound are those in the arena or the expensive front stalls seats. The rest of the audience is to all intents and purposes listening in mono because they are so far from the orchestra (The optimum ratio of distance from the sound source to width of stereo source is approximately 1.3 to 1. For a seat to the rear of the Albert Hall the ratio is 6.5 to 1). The Albert Hall provides a topical example, but the same applies in most large concert halls - a reader wrote in a comment of how at a Festival Hall concert "the Schumann symphony... could be heard playing in the distance like a phantom radio programme on the BBC World Service long wave" [3].

BBC Proms broadcasts and recent commercial recordings may have sonic impact and immediacy. But that sound is artificially contrived; which is not in itself as a bad thing as all recordings are artificial artifacts. However, the divergence between up close and personal broadcast/recorded sound and the experience of a neophyte in a cheap concert seat hearing an orchestra playing in the distance like a phantom radio programme may explain why the current generation of headphone listeners fails to engage with live classical music. An example of the trend towards up close and personal sound is the current Classic FM experimental binaural stream of the Philharmonia Orchestra playing the Die Fledermaus Overture - see graphic below. This was recorded using the standard binaural set-up of a dummy head with microphones in each ear cavity. (As only two microphones can be used in binaural recording and as a solid dummy head separates the microphones, phase errors are minimised). For the demonstration recording the dummy head was placed in the middle of the orchestra; this allows the listener to move from a cheap seat at the back of the hall with its mono sound, to sit in the middle of the orchestra and experience the power of the music coming from every direction [4]. The binaural Die Fledermaus can be sampled via this link - headphones are, of course, required.

Purists may shudder at hearing classical music so up close and personal, but, whether we like it or not, this is what new audiences conditioned to headphone listening want. Everything in the concert hall from dress conventions to silence between movements has been questioned. But there has been very little constructive debate about the sound itself. Because any debate usually descends into a destructive argument about introducing sound shaping and other new technologies into the concert hall. Questions and answers are being confused here. First we must debate whether concert hall sound is the reason why live classical music is failing to engage new and young audiences. Only when we have debated that question and arrived at the conclusion that the sound is a barrier to audience engagement should we start debating possible solutions.



[1] Transparent acoustic screens are often deployed between orchestra sections at the Proms to reduce sound leakage.
[2] Another potential source of phase errors is the trend towards multi-channel formats. In view of the BBC's claim of financial penury it is surprising that covert experiments with broadcasting the Proms in 4.0 surround sound continue. An experiment involving folding the 4.0 stream down to mono would be illuminating.
[3] Not all concert halls suffer from the problem of poor sound in the cheaper seats; examples of where the problem does not occur include Symphony Hall Birmingham and Snape Maltings. The sound quality in the more distant seats is determined by complex acoustical factors including ratio of depth to width of the hall and overall size. To illustrate the role of size, the Albert hall has a seating capacity of 5000, whereas the acoustically famed Musikverein, Vienna and Concertgebouw, Amsterdam seat 1744 and 1974 respectively.
[4] In the past the BBC have experimented with binaural broadcasts of the Proms and samples of the 2016 concerts are available online as binaural streams. In 1978 a play without dialogue by Andrew Sachs titled The Revenge was also broadcast in binaural sound by the BBC. This was an important experiment that is worth auditioning; an audio file is available which provides an introduction to binaural sound - listen here.

Photo of Prom via @bbcproms. Any copyrighted material is included as "fair use" for critical analysis only, and will be removed at the request of copyright owner(s). Also on Facebook and Twitter.

5 comments:

Pliable said...

Reader Aaron Cassidy points out that I have missed mentioning that samples of 2016 BBC Proms are available in binaural sound - http://www.bbc.co.uk/programmes/articles/2913JxRtQl3ZTvw0wz5C4D1/proms-in-binaural-sound

My article has now been amended to mention this. But I must mention that despite researching this article for some time and undertaking many online searches, these streams of the 2016 BBC Proms did not come to my attention. I try desperately hard not to sound too negative about the BBC. But if less time was spent by them plugging the BBC brand and more time was spent talking to the audience intelligently, I for one would view the BBC in a more favourable light.

Bodie said...

I can imagine the conductor becoming a mic stand of sorts, wearing a hat or helmet with with mics near his ears. Audience members could elect to wear headphones during the concert in order to hear what the conductor hears, rather than what they would normally hear in their nosebleed seats. I love hearing the different locations of the instruments in the binaural example you linked.

kirkmc said...

While this is interesting, I'm a bit confused. There is no such thing as mono or stereo sound at a concert. The sound goes in all directions, crossing over and folding, extending and spreading as it resonates in a hall. Calling it mono sound is simply wrong. With that logic, any piano recital I hear would be in mono, even if I'm in the front row.

Also, this:

"First we must debate whether concert hall sound is the reason why live classical music is failing to engage new and young audiences."

Seriously? Young audiences - who, as you explain, listen to DJs and club music in mono - decide not to go to a concert in, say, Royal Albert Hall because the sound is in "mono?" Concert hall sound is immersive, certainly better than pretty much anyone short of a billionaire can hear at home. The thought that a young person would actually reflect on this, and say, "You know what, I'm going to skip that symphony concert because the concert hall has mono sound" is, frankly, ludicrous.

I do, however, agree that the sound of concerts when filmed tends to be annoying. You mention the spotlighting of instruments; that should never happen. It's just wrong. You notice this often on DVD/Blu-Rays of classical concerts, when the director switches to, say, the clarinets, and you hear the clarinets more clearly than before.

Pliable said...

kirkmc, can we first start by unpacking your statement that “The thought that a young person would actually reflect on this, and say, "You know what, I'm going to skip that symphony concert because the concert hall has mono sound" is, frankly, ludicrous”.

You offer no independent support for your statement that my proposition is “frankly, ludicrous”. This means that my proposition is in your view frankly, ludicrous, which is simply a somewhat unhelpful way of saying that you disagree with me. As I say in the post any debate usually descends into a destructive argument , and pejoratives like “frankly, ludicrous” are typical of that destructive tendency, particularly when they come from what is effectively an anonymous source. On An Overgrown Path often expresses dissenting views, so it comes as no surprise that your view differs from mine. Both of us are unable to supply any objective data proving that the sound in the cheaper seats of a large concert hall is, or is not, a barrier to the engagement of new audiences. Which underlines my point that this possibility needs at least to be given more serious consideration.

I suspect that you are not a regular reader because there has been extensive previous discussion, much of it hyperlinked in this post, of how the reference sound for music consumers has changed dramatically in recent years due to the hegemony of portable audio players and headphone listening. The term 'mono' that I use to describe the sound heard in different parts of a concert hall is a convenient shorthand for non-immersive, unidimensional, distant, unengaging, and a number of other less punchy adjectives that can be used to describe the kind of sound that a reader eloquently described as a “symphony... playing in the distance like a phantom radio programme on the BBC World Service long wave”. The sound that the reader was describing is that heard in the London's Royal Festival Hall, and I stand by my proposition that during a piano recital in that hall the sound heard at the back of the hall will be significantly less engaging, particularly for a newcomer to classical music, than that heard in a front seat.

You express the view that “concert hall sound is immersive, certainly better than pretty much anyone short of a billionaire can hear at home”. But that statement does not make any attempt to define 'better'. What is better to a seasoned concert goer may not be better for a newcomer. Would someone new to concert going rate the sound heard at the rear of the Royal Festival Hall as 'better' than that heard in the Philharmonia binaural demo?

I make no claim to infallibility. But as Alex Ross has said on Twitter, there is much to absorb and ponder in this post. So let's try to absorb and ponder without being dismissive. As has been discussed here before, concert hall sound is an artifice determined by 19th century acoustic conventions. Times, expectations and technology are changing rapidly, and it is not unreasonable to question whether these legacy conventions may now have passed their sell-buy date.

To avoid unproductive circular arguments I would be grateful if further debate on this specific comment is confined to adding new and relevant information.

iarful said...

I was grateful to see this reminder of the artificiality of the current Proms sound (especially as it is informed by a former professional's understanding of the reasons behind it, both technical and artistic) and mainly wanted to add a reflection from the perspective of musicians, to complement the article's concern with the audience. But with reference to kirkmc's comment, I remember (vividly) my first exposure, well over 30 years ago, to an ambisonic recording of a concert (reproduced without height). My first reaction was that it sounded almost like mono - because hall sound does from most seats after the artificial conditioning of stereo reproduction. My ears soon adjusted to the much greater realism of course and were astonished at the end when I heard applause from all around me followed by people talking in the row behind, their seats tipping up, etc. Stereo sounded absolutely pathetic for a long time afterwards! As far as I am concerned the 'near-mono' of ambisonic rules OK - except that, sadly, it still doesn't, decades later.

A week ago I was involved in a discussion on reproduced sound which included two professional pianists. The first was horrified to learn how the Proms sound picture is constructed in the control room from multiple close microphones with artificial reverberation, etc. - just deeply shocked. The other teaches in a music faculty which also runs a sound engineer course, the students on which hone their skills by recording concerts or rehearsals at the musicians' request. She related how there is also the option of recording without the students' help, where a pair of microphones descend from the ceiling and the digital recorder starts. She said that very often she prefers the results produced by this set-up. When I said how I often felt with modern reproduced sound it was impossible for the ear to focus its attention as it can in real life, to concentrate on this or that strand in the music or a particular performer (no doubt caused by the ultimate lack of coherence you describe, and in the early days of digital by its poor performance at low level) so that there was a sort of barrier beyond which one could not hear, she agreed enthusiastically: "Of course!" For all the limitations it may have, the 2 microphone set-up will give a truthful account of balance between the different musicians and just the sort of coherence which is necessary for the ear to do its work, for one to listen actively. So for very different reasons, those who produce the music are also less than satisfied with current practice.

As I write I have just been listening to the afternoon repeat of the Brahms 2nd piano concerto from the Proms. A review I read commented that Peter Serkin was at times all but inaudible in the hall. There was of course not a hint of this in the Radio 3 balance. This is an 'improvement' for the listener at home, but that old Quad slogan of "the closest approach to the original sound" has been well and truly abandoned - for good and for ill. To my mind there is a lot more of the latter than most people think, especially if we want audience members to be able to use their ears (and the brains that are connected to them).