Excellent discussion. I'm toying with the idea of getting an old Lexicon MC 4 for experimenting with multichannel (don't tell me girlfriend). Does Logic 7 upmix stereo to a center channel as well, or only to the rear and side channels?
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Genelec on audio science
- Thread starter svart-hvitt
- Start date
- Joined
- Mar 12, 2018
- Messages
- 367
- Likes
- 3,905
I had a MC-12 for several years and I don't know what, if any, simplifications/changes were made to Logic 7 for the MC-4. The algorithm changed slightly over the years as David Griesinger made improvements. Overall, though, there was a deliberate attempt not to mess with the stereo soundstage, merely to generate a more enveloping/involving experience.
Sounds good! So I guess a couple of surround speakers together with an mc 4 could be a worthwhile investment, then? Might be worth a try.
Guermantes
Senior Member
Or it can be thought of as a creative space rather than a "gimmick" -- just as many classic rock albums use stereo creatively or installations use multiple speakers for the construction of sound architectures.
I have a few of those SACDs and there is one I definitely prefer in stereo (because the centre sounds distorted): https://www.discogs.com/Mahler-Maur...ymphony-The-Song-Of-The-Earth/release/7660895
But isn't the 3-channel replay of those recordings a happy accident (perhaps even a "gimmick")? The centre channel was originally a separate microphone positioned to make a concurrent mono-compatible recording and wasn't really intended to be heard together with the stereo microphone tracks. So for decades there was no thought of replaying them as 3-channels until domestic 5.1 surround introduced a centre speaker.
- Joined
- Mar 12, 2018
- Messages
- 367
- Likes
- 3,905
Some things are just too simple
. No delay means "omnisound", and no DSP to extract only the uncorrelated material from the discrete L & R material. I tried it at the time and gave it up quickly. However, if a marriage is at stake . . . the bar can be lowered, I guess
- Joined
- Feb 23, 2016
- Messages
- 20,696
- Likes
- 37,431
I tried this a few times. Once with Acoustat Two's up front and a pair of Maggie MG2's in the rear. I quickly gave it up too. I later retried it with an early digital delay unit that delayed only the rear channels for the purpose of Hafler setup. Didn't like it either. It was better than no delay, but sounded very artificial and 'tacked onto' the recording space.Some things are just too simple
No delay means "omnisound", and no DSP to extract only the uncorrelated material from the discrete L & R material. I tried it at the time and gave it up quickly. However, if a marriage is at stake . . . the bar can be lowered, I guess
OP
svart-hvitt
Major Contributor
- Joined
- Aug 31, 2017
- Messages
- 2,375
- Likes
- 1,253
- Thread Starter
- #268
GENELEC AND IMMERSIVE AUDIO
There are a couple of notable examples of implementation of Genelec speakers for immersive audio. Let me go through a few of them.
The first that comes to mind is University of Surrey’s 22.2 lab:
https://www.genelec.com/university-surrey-opens-222-listening-room
Another example, more of an experiment, is this 29 speaker (plus subs) setup:
https://www.genelec.com/genelec-drives-3d-audio-revolution-warsnare
A truly Finnish setup is the Finnish Musicians’ Union’s G Livelab 150 seat scene in Helsinki:
https://www.genelec.com/g-livelab
https://www.genelec.com/sites/default/files/media/References/Installations/case_study_g_livelab.pdf
In total 42 speakers are implemented:
«The system takes input from six microphones in the ceiling of the stage or from the stereo auxiliary output directly from the desk. Thus, the virtual acoustics can be used with a purely acoustic act, but also with an amplified act together with a normal PA system. The input signals are fed to a custom made 64-channel time-variant feedback delay network that generates natural reverberation, which is routed to the 42 virtual acoustics loudspeakers. The outputs are delayed according to the position of each loudspeaker to guarantee the proper localization of sound sources on the stage. The reverberation can be adjusted so that the space sounds anything from a dry bar up to a cathedral. In addition, the levels of reverberation on the stage and in the audience area can be separately adjusted to optimize both the support for the musicians and the spatial sound for the audience».
A more down-to-earth example, where the chain from performance to finished playback material is very short, is the 5 speakers plus subs surround setup in a control room of the Royal Opera House in Stockholm:
https://www.genelec.com/royal-opera-house-stockholm-takes-new-surround-studio-use
A couple of view points from Genelec on the issue of immersive sound, first a more marketing related comment from CEO Siamäk Naghian and then a more research oriented comment from Thomas Lund:
https://www.genelec.com/blog/immersive-audio-next-big-thing
https://www.genelec.com/blog/immersive-monitoring-perceptive-perspective
From my view point, it seems like the debate on immersive audio - as it is discussed at universities - considers stereo as a trivial problem. Which is not the same impression one gets when browsing audio forums. Is my impression correct? Why do we have this gap between universities and audio forums?
Having said that, if audio is a problem to be «solved», and if one agrees that psychoacoustics has its merit, wouldn’t it be correct to demand that audio science develops both a model of the physics (speaker and room) and a model of the human mind - as well as modeling how the two models interact? Why is it then, that «audio experts» are mostly from the engineering school, and not from medical school or institue of psychology? Can audio be science without a model of the physics and the mind at the same time?
There are a couple of notable examples of implementation of Genelec speakers for immersive audio. Let me go through a few of them.
The first that comes to mind is University of Surrey’s 22.2 lab:
https://www.genelec.com/university-surrey-opens-222-listening-room
Another example, more of an experiment, is this 29 speaker (plus subs) setup:
https://www.genelec.com/genelec-drives-3d-audio-revolution-warsnare
A truly Finnish setup is the Finnish Musicians’ Union’s G Livelab 150 seat scene in Helsinki:
https://www.genelec.com/g-livelab
https://www.genelec.com/sites/default/files/media/References/Installations/case_study_g_livelab.pdf
In total 42 speakers are implemented:
«The system takes input from six microphones in the ceiling of the stage or from the stereo auxiliary output directly from the desk. Thus, the virtual acoustics can be used with a purely acoustic act, but also with an amplified act together with a normal PA system. The input signals are fed to a custom made 64-channel time-variant feedback delay network that generates natural reverberation, which is routed to the 42 virtual acoustics loudspeakers. The outputs are delayed according to the position of each loudspeaker to guarantee the proper localization of sound sources on the stage. The reverberation can be adjusted so that the space sounds anything from a dry bar up to a cathedral. In addition, the levels of reverberation on the stage and in the audience area can be separately adjusted to optimize both the support for the musicians and the spatial sound for the audience».
A more down-to-earth example, where the chain from performance to finished playback material is very short, is the 5 speakers plus subs surround setup in a control room of the Royal Opera House in Stockholm:
https://www.genelec.com/royal-opera-house-stockholm-takes-new-surround-studio-use
A couple of view points from Genelec on the issue of immersive sound, first a more marketing related comment from CEO Siamäk Naghian and then a more research oriented comment from Thomas Lund:
https://www.genelec.com/blog/immersive-audio-next-big-thing
https://www.genelec.com/blog/immersive-monitoring-perceptive-perspective
From my view point, it seems like the debate on immersive audio - as it is discussed at universities - considers stereo as a trivial problem. Which is not the same impression one gets when browsing audio forums. Is my impression correct? Why do we have this gap between universities and audio forums?
Having said that, if audio is a problem to be «solved», and if one agrees that psychoacoustics has its merit, wouldn’t it be correct to demand that audio science develops both a model of the physics (speaker and room) and a model of the human mind - as well as modeling how the two models interact? Why is it then, that «audio experts» are mostly from the engineering school, and not from medical school or institue of psychology? Can audio be science without a model of the physics and the mind at the same time?
Last edited:
At the same time it has to be noted that audio forums also think that short lengths of mains cable are not a trivial problem.
GENELEC AND IMMERSIVE AUDIO
There are a couple of notable examples of implementation of Genelec speakers for immersive audio. Let me go through a few of them.
The first that comes to mind is University of Surrey’s 22.2 lab:
https://www.genelec.com/university-surrey-opens-222-listening-room
Another example, more of an experiment, is this 29 speaker (plus subs) setup:
https://www.genelec.com/genelec-drives-3d-audio-revolution-warsnare
A truly Finnish setup is the Finnish Musicians’ Union’s G Livelab 150 seat scene in Helsinki:
https://www.genelec.com/g-livelab
https://www.genelec.com/sites/default/files/media/References/Installations/case_study_g_livelab.pdf
In total 42 speakers are implemented:
«The system takes input from six microphones in the ceiling of the stage or from the stereo auxiliary output directly from the desk. Thus, the virtual acoustics can be used with a purely acoustic act, but also with an amplified act together with a normal PA system. The input signals are fed to a custom made 64-channel time-variant feedback delay network that generates natural reverberation, which is routed to the 42 virtual acoustics loudspeakers. The outputs are delayed according to the position of each loudspeaker to guarantee the proper localization of sound sources on the stage. The reverberation can be adjusted so that the space sounds anything from a dry bar up to a cathedral. In addition, the levels of reverberation on the stage and in the audience area can be separately adjusted to optimize both the support for the musicians and the spatial sound for the audience».
A more down-to-earth example, where the chain from performance to finished playback material is very short, is the 5 speakers plus subs surround setup in a control room of the Royal Opera House in Stockholm:
https://www.genelec.com/royal-opera-house-stockholm-takes-new-surround-studio-use
A couple of view points from Genelec on the issue of immersive sound, first a more marketing related comment from CEO Siamäk Naghian and then a more research oriented comment from Thomas Lund:
https://www.genelec.com/blog/immersive-audio-next-big-thing
https://www.genelec.com/blog/immersive-monitoring-perceptive-perspective
From my view point, it seems like the debate on immersive audio - as it is discussed at universities - considers stereo as a trivial problem. Which is not the same impression one gets when browsing audio forums. Is my impression correct? Why do we have this gap between universities and audio forums?
Having said that, if audio is a problem to be «solved», and if one agrees that psychoacoustics has its merit, wouldn’t it be correct to demand that audio science develops both a model of the physics (speaker and room) and a model of the human mind - as well as modeling how the two models interact? Why is it then, that «audio experts» are mostly from the engineering school, and not from medical school or institue of psychology? Can audio be science without a model of the physics and the mind at the same time?
Can you relate these set-ups to practical take-up for domestic use? It may find a home in commercial VR simulators but given the mediocre acceptance of Mch for other than better cheap home TV viewing systems it is problematic for domestic use. For gamers prepared to wear a helmet full of small transducers it may find a niche or maybe as fixed installations for tweaking concert spaces.
Recording studios would only need these arrays if there was a profitable market for them. Lots of mics. must surely be a movie directors nightmare.
I thought the links were more of an overview than a practical guide to usage. Interesting subject.
Last edited:
OP
svart-hvitt
Major Contributor
- Joined
- Aug 31, 2017
- Messages
- 2,375
- Likes
- 1,253
- Thread Starter
- #272
@Wombat , highly complex setups from a lab situation are - in my view - relevant for a domestic setting.
Research labs often look messy to the untrained eye; giving the impression that science is not relevant for you and me. However, that’s far from the truth in many cases. Research that gives insight into complex problems, where one can control for variables and use creativity beyond the limits of conventional thinking, can be highly rewarding for the ordinary man too.
I guess we may have different views on what is science; I would like to encourage science that is curious and not constrained by conventions and limits of habit.
Having said that, I have been to domestic setups that are as «messy» as these labs and complex setups. So in the exceptional case, «domestic» doesn’t mean much. Besides, the most usual «domestic» setup is the mono one-channel setup, I believe, considering the sales of home speakers:
Source: https://www.macrumors.com/2018/05/17/apple-first-quarter-2018-homepod-sales/
OP
svart-hvitt
Major Contributor
- Joined
- Aug 31, 2017
- Messages
- 2,375
- Likes
- 1,253
- Thread Starter
- #274
Yes, always keep an eye on the money funding the research.
One thing that comes to mind, of course, is the fact that the biggest application of multichannel audio is in modern cars. Samsung didn’t buy Harman due to their stereo efforts. It was the car market that caught the eye of analysts and the media:
https://www.forbes.com/sites/sarwan...essure-on-auto-tier-1-and-apple/#1e2ba475509c
On the distinction between «pure science» and «applied science»: That’s a very broad and big discussion. I believe even mathematicians would claim that their high-level thinking has application use just waiting to surface around the corner (that has indeed been the case in some instances, maybe PCM is a good example in audio?).
And adding to the wiring loom is easy plus speaker enclosure cost is minimal.
I can see the application would work well in a disorienting softening-up process for interrogation subjects - without the high level sounds currently applied. I assume the military would be well ahead of me here. Just being imaginative.
Just being imaginative. 
Allow me to review two previous topics before getting back to immersive reproduction.
Spectral balance and absolute level are both capable of inducing self reference on music production; two downwards spirals some say. It would be unreasonable to expect anything else from a playback system than a flat direct sound frequency response; and Loudness standards now define digital production leveling across formats (ITU-R BS.1770, EBU R128 and similar regional offsprings). Artistic reasons for squashing content are valid, but a competition on absolute level is futile; and taken out of the equation when delivery is Loudness normalised.
New physiological and psychological studies indicate that human perception is more reliant on learning, expectation and reach-out behaviour than thought just a decade ago. Actual stimuli from outside the body (exteroception) have less influence than the two former. Perceptual velocity (“nachrichtenfluss”) is also smaller than we tend to believe. Hallucinations or mental illness might even be better understood as merely a further downgrading of exteroception, with the mind free-wheeling and the outside world sampled less still.
Those findings may help explain some of the different opinions on this thread, because time is a great influencer when listening. For instance, it takes a while to learn the sound of your loudspeakers placed as they are in the room, including to some extent separating direct sound from reflections using movement. Fine mastering engineers have invested that time and know how to tackle other potential confounders, e.g. “subjective frequency response”, by listening at a defined level. The less variation, the better the chance of a comprehensive judgement without spending months or years.
We have measured mastering rooms where objective direct sound frequency response wasn’t flat, but the acoustic setup typically excellent, e.g. minimising destructive interference and perceivable reflections at the listening position. Under such conditions, time is on your side, any change a deterioration and a flat frequency response not paramount.
However, considering audio reproduction in general, and moving between rooms, a flat frequency response is the reference all audio professionals should be familiar with. It’s a simple question of reducing unknowns and thereby time to (re-)learn. Other factors are important too, for instance smooth directivity, minimising crossover colouration in direct sound and reflections, listening level etc.; and a decent room with optimal monitor placement is naturally the foundation before any in-situ adjustments are added.
Because of the above, we use “slow listening” principles complementary to traditional subjective test schemes; but remain skeptical to absolute preference findings in listening tests relying on any recorded material, as the foundation itself is self-referenced.
Spectral balance and absolute level are both capable of inducing self reference on music production; two downwards spirals some say. It would be unreasonable to expect anything else from a playback system than a flat direct sound frequency response; and Loudness standards now define digital production leveling across formats (ITU-R BS.1770, EBU R128 and similar regional offsprings). Artistic reasons for squashing content are valid, but a competition on absolute level is futile; and taken out of the equation when delivery is Loudness normalised.
New physiological and psychological studies indicate that human perception is more reliant on learning, expectation and reach-out behaviour than thought just a decade ago. Actual stimuli from outside the body (exteroception) have less influence than the two former. Perceptual velocity (“nachrichtenfluss”) is also smaller than we tend to believe. Hallucinations or mental illness might even be better understood as merely a further downgrading of exteroception, with the mind free-wheeling and the outside world sampled less still.
Those findings may help explain some of the different opinions on this thread, because time is a great influencer when listening. For instance, it takes a while to learn the sound of your loudspeakers placed as they are in the room, including to some extent separating direct sound from reflections using movement. Fine mastering engineers have invested that time and know how to tackle other potential confounders, e.g. “subjective frequency response”, by listening at a defined level. The less variation, the better the chance of a comprehensive judgement without spending months or years.
We have measured mastering rooms where objective direct sound frequency response wasn’t flat, but the acoustic setup typically excellent, e.g. minimising destructive interference and perceivable reflections at the listening position. Under such conditions, time is on your side, any change a deterioration and a flat frequency response not paramount.
However, considering audio reproduction in general, and moving between rooms, a flat frequency response is the reference all audio professionals should be familiar with. It’s a simple question of reducing unknowns and thereby time to (re-)learn. Other factors are important too, for instance smooth directivity, minimising crossover colouration in direct sound and reflections, listening level etc.; and a decent room with optimal monitor placement is naturally the foundation before any in-situ adjustments are added.
Because of the above, we use “slow listening” principles complementary to traditional subjective test schemes; but remain skeptical to absolute preference findings in listening tests relying on any recorded material, as the foundation itself is self-referenced.
A long post. But how important is this to the average person and thus to those who pander to their requirements?
Lots of round-and-round on audio forums but the everyday citizen doesn't take much notice. Other things are more important.
That is not to say if something easy and affordable comes along they won't consider it. However, technology fatigue is a barrier. People are now more discerning re where to spend their harder earned income. Too many 'innovators' are asking for it.
Technology based products seem to be out-dated faster than the broader population is comfortable with. Silicon Valley is well aware of this. Features of products are often advanced well beyond consumer needs and they are noticing.
So, back to asking where this research is heading?
Lots of round-and-round on audio forums but the everyday citizen doesn't take much notice. Other things are more important.
That is not to say if something easy and affordable comes along they won't consider it. However, technology fatigue is a barrier. People are now more discerning re where to spend their harder earned income. Too many 'innovators' are asking for it.
Technology based products seem to be out-dated faster than the broader population is comfortable with. Silicon Valley is well aware of this. Features of products are often advanced well beyond consumer needs and they are noticing.
So, back to asking where this research is heading?
Last edited:
Multi-channel surround: an analogy..?
Perhaps what I was going on about yesterday could be described as follows:
Suppose you have a 3D TV. By whatever means are used, separate images are directed to the eyes, thus giving you an illusion of a 3D scene before you. However, it is limited to the scene only being in front of you. Some technologies require active glasses, some passive, and some use optical prism arrays.
Adding a third 'channel' would make no sense because you only have two eyes. It would not be a statement of the obvious to say that three or more channels were better than two - as has been said about multi-channel audio.
You could don a VR headset, which would allow you to synthesise an all-round 3D scene - although not necessarily a 'recorded' one. There would be a difference between being able to turn your head and look up and down, and actually being able to get up and walk about. Also, VR headsets are uncomfortable.
But here, I think, is where the temptation towards 'gimmick' comes in... If you were to place loads of ordinary 2D TV sets around the room, 90% of people would think they were in a genuine 'immersive' video environment. This is the same as those VR headset videos that are not 3D, but allow you to turn your head and look around as though surrounded by a large spherical TV set. Most people don't even care that it isn't 3D.
Not only that, but such a scheme would enable you to say "Hey! Look we can even get up and walk about and it still makes sense. This truly is a holodeck". But it wouldn't be. Most people probably wouldn't care, though.
This is what I am thinking is going on with much of this multi-channel audio (even just the idea of adding a centre channel). It is certainly 'surround' and 'multi-channel' and kind of 'immersive', but it isn't '3D'. Whereas plain common or garden stereo is 3D. But maybe the distinction between the two isn't important and no one cares.
There may be ways to 'project' a 3D scene at the listener, using '2D' transducers and head tracking - I suppose BACCH et al. are on this path. It would be great if it worked, and didn't have any side effects and felt real when you turned your head rather than just 'headphones without headphones'.
But in the meantime, I think that plain stereo with speakers performs quite a trick: it creates a genuine 3D image, but it also blends with the real room acoustics so that the listeners' own speech, shuffling of the feet, etc. feel quite consistent with the music. It creates an illusion of 3D, yet 'makes sense' even when turning your head. It seems natural to 'face the music' as in 99% of musical performance situations.
Given limited resources, should a person concentrate on getting good stereo or comprehensive multi-channel? I would say the former. If resources are not limited, then sure, go for good stereo and multi-channel - but don't use a centre channel when not listening to surround sound: it constitutes the meaningless 'third channel' that you don't need if your stereo setup works well.
Perhaps what I was going on about yesterday could be described as follows:
Suppose you have a 3D TV. By whatever means are used, separate images are directed to the eyes, thus giving you an illusion of a 3D scene before you. However, it is limited to the scene only being in front of you. Some technologies require active glasses, some passive, and some use optical prism arrays.
Adding a third 'channel' would make no sense because you only have two eyes. It would not be a statement of the obvious to say that three or more channels were better than two - as has been said about multi-channel audio.
You could don a VR headset, which would allow you to synthesise an all-round 3D scene - although not necessarily a 'recorded' one. There would be a difference between being able to turn your head and look up and down, and actually being able to get up and walk about. Also, VR headsets are uncomfortable.
But here, I think, is where the temptation towards 'gimmick' comes in... If you were to place loads of ordinary 2D TV sets around the room, 90% of people would think they were in a genuine 'immersive' video environment. This is the same as those VR headset videos that are not 3D, but allow you to turn your head and look around as though surrounded by a large spherical TV set. Most people don't even care that it isn't 3D.
Not only that, but such a scheme would enable you to say "Hey! Look we can even get up and walk about and it still makes sense. This truly is a holodeck". But it wouldn't be. Most people probably wouldn't care, though.
This is what I am thinking is going on with much of this multi-channel audio (even just the idea of adding a centre channel). It is certainly 'surround' and 'multi-channel' and kind of 'immersive', but it isn't '3D'. Whereas plain common or garden stereo is 3D. But maybe the distinction between the two isn't important and no one cares.
There may be ways to 'project' a 3D scene at the listener, using '2D' transducers and head tracking - I suppose BACCH et al. are on this path. It would be great if it worked, and didn't have any side effects and felt real when you turned your head rather than just 'headphones without headphones'.
But in the meantime, I think that plain stereo with speakers performs quite a trick: it creates a genuine 3D image, but it also blends with the real room acoustics so that the listeners' own speech, shuffling of the feet, etc. feel quite consistent with the music. It creates an illusion of 3D, yet 'makes sense' even when turning your head. It seems natural to 'face the music' as in 99% of musical performance situations.
Given limited resources, should a person concentrate on getting good stereo or comprehensive multi-channel? I would say the former. If resources are not limited, then sure, go for good stereo and multi-channel - but don't use a centre channel when not listening to surround sound: it constitutes the meaningless 'third channel' that you don't need if your stereo setup works well.
Similar threads
