@William said:

Wouldn't one way of achieving this be to record all samples - even solos - with all the instruments and players in position?

I know, Herb is rolling his eyes and shaking his head at the expense. But it would be an interesting, bankruptcy-causing experiment.

I thought of that straightaway - and then I thought about the cost and the size of the samples. And the size of the sound stage. Good idea though.

If you have read the announcement of the MIR-project, you may remember the last few sentences:

@Another User said:

[...]

The Future

The innovative ideas of the Vienna Symphonic Library are by no means exhausted here. Our ongoing research and development explores the cumulative effects of adding individual instrumental group resonances with each other – every instrument being, as it were, a space by itself. The goal is to blend all of the sonic nuances of individual orchestral instruments and groups with the acoustic characteristics of the room at large to create a complete acoustic picture with unprecedented realism. Combining the acoustic properties of the very small with the all-encompassing sonority of the very large is a revolutionary step toward achieving virtual acoustic perfection.

-> http://vsl.co.at/english/pages/profile/news/the_mir_project.htm

... so I think I can say that we are aware of the possibilities.

As far as I can tell, this is nothing that you will see in Vers.1, though ... [8-)]

All the best,

/Dietz - Vienna Symphonic Library

Good points Paul, Cm and William, although i can imagine it wouldn't be economically feasible, and Herb would be wondering if we've all gone mad!

Another thought i have is a 'digital chamber' track. And this is altogether more complicated.
Given the discussion so far is generally agreed on the 'resonance' theory, I wonder if its possible to create a 'digital chamber', in which is represented all the instruments in the arrangement. Now I hear some say this is Reverb, but in fact reverb is wholly assigned by track or group of tracks, and has no actual interactive resonance with the other tracks in the arrangement.

The digital chamber track would be one which takes each track recorded, and puts a 'resonance whisper' in the Digital chamber track.
Still not enough.

In the digital chamber, each whisper is encouraged to resonate according to a series of templates. e.g. Your template one is full Wagnerian Orchestra. The digital track contains references to all the instruments, and importantly their placement, and resonant 'ability'. You've already made the point about Percussion resonance, and i too have sat in a percussion section and felt this.

So in our 'Digital Chamber' the snare is rightly turned off, but the Gong, Tympani, basedrum, windchimes etc. all interact with instruments to produce a 'Orchestral Resonance' that may be much closer to live than before.
At this point I ask you forgive my terminology, as I am explaning this from the point of view of a composer and player, not a specialist sound engineer.

So our Digital chamber now contains a resonance instrument to be added to the overall arrangement.
And I wonder if it's also possible, given standard spatial templates, i.e. Room, Hall, Stadium, Neighbour's toilet etc., that the track automatically recalculates the resonance, and counter resonances, and applies them.

In addition to a spatial placement construct, such as i understand is the idea behind MIR, we then have an added element digitally 'interacting' the resonance between instruments.
As William said, there are no doubt some seriously gifted acoustical engineers among us, and i would welcome their views on this.

Regards to you all,

Alex.

I saw your post after mine, and i am suddenly a lot wiser about MiR and what it is you're trying to do.

Regards,


Alex.

@PaulR said:

I thought of that straightaway - and then I thought about the cost and the size of the samples. And the size of the sound stage. Good idea though.

It won't be that easy. E.g. C will resonate differently if you play the C below, the F below the D# below, the D# two octaves above on another instrument. Now taken we have playing instruments across all range, a piano could make a tuba resonate, and so on. So like the legatos, I think the only way to get it would to record all combinations. The cost of having every player on the rightsized stage will be minimal to these [;)]

Actually I don't know, but I thought some hardware reverb units, or more general, such that aren't convoluting impulses only, do some kind of resonating also at the reverb stage.

Some results are also coming from modelling instruments theirselves virtually, because all the resonances have to be considered then. As we all know the technolgy still has to develop a bit to get as real as recordings or samples for that matter. Also things like "body resonance" that Michiel Post is researching for his pianos are directed at this topic.

PolarBear

Do you know what would really add that touch of reality? Something so subtle, yet so intrisic to a realistic sounding recording? Well, what we need are samples of the audience coughing and shifting in their seats! What else compliments a quiet passage of a cello solo than some sickly old man clearing the phlegm from his throat! Come on guys, get on this now!

-BW

@Horse Opera said:

Do you know what would really add that touch of reality? Something so subtle, yet so intrisic to a realistic sounding recording? Well, what we need are samples of the audience coughing and shifting in their seats! What else compliments a quiet passage of a cello solo than some sickly old man clearing the phlegm from his throat! Come on guys, get on this now!

-BW

lol thats so funny, I was thinking about posting that over the weekend. I'd love to have a cough track to add to realism.

@Horse Opera said:

Do you know what would really add that touch of reality? Something so subtle, yet so intrisic to a realistic sounding recording? Well, what we need are samples of the audience coughing and shifting in their seats! What else compliments a quiet passage of a cello solo than some sickly old man clearing the phlegm from his throat! Come on guys, get on this now!

-BW

lol thats so funny, I was thinking about posting that over the weekend. I'd love to have a cough track to add to realism.

Funnily enough there are some "scoring-noise" samples posted on Northern Sounds at the moment. It all seems rather strange as I've spent years removing the chair squeaks from recordings [:D]

DG

If they are the same I got from Bela (I believe that is the source) they overdo it more than anything. People walking around isn't usually happening at performances (though I remember a piece where exactly this was a part of the composition - somebody help me out with the name). There was also one sample from VSL also but it somehow didn't make it to the users it seems. Maarten used it on at least one of his demos for VSL. But yes, that airy feeling is missing a bit in the world of clean samples. I also think that more or less good mixing skills contribute to it, too (e.g. see what Thomas_J did with libraries some of us wouldn't even like to touch once).

PolarBear

@PolarBear said:

I People walking around isn't usually happening at performances (though I remember a piece where exactly this was a part of the composition - somebody help me out with the name).

Are you thinking of HPSCHD by John cage and Lejaren Hiller? When it's performed live you're supposed to walk through the performance area, in between the harpsichord players and the speakers.

Very funny...!

And then you could add the dulcet tones of a mobile phone going off in the middle of a desperately tremulous solo, and the selfish bastard looking at someone else trying to indicate 'it wasn't me.'

Regards,


Alex!

that happened at the alys stephens center when the orpheus chamber orchestra came to town. The worst thing was, the cell phone was playing some old pop song from the 70s >.<

We'll also need some samples of obnoxious teenagers saying things like, "this music is boring!" and "mom, how long do we have to be here?" since I seem to always sit in front of those kinds of people whenever I go to the Bowl.

-BW

hehe that reminds me of the story of one of the first performances of 4'33"

apparently at one of the colleges in the US (indiana maybe) the practice rooms are right above the concert hall. Now, they're well insulated and durring a normal performance you can't hear anything but THAT day there happened to be a piano student preparing for his jury. When they got around to doing 4'33" the audience was treated to 8 bars of piano mu sic followed shortly by loud screams and explitives over and over. [:D]

A few years ago i played in small orchestra for a charity christmas event in aid of Legacy in Australia.
During a particularly sweet violin solo arranged for the event, a young, very attractive lady wearing nothing but a very small christmassy thong, jumped onstage and 'waved' her assets in front of the violin player. Like a true professional he kept going, never missing a note and with no sign of 'anxiety'. The young lady tried her hardest to put him off, but he refused to stray from the musical path, and finished the solo to a rousing cheer of congratulations from the audience, as much for his 'determination' as the quality of his playing.

When asked afterwards how he kept his composure under such a 'challenging situation', he smiled and replied," She could have been stark naked and covered in jam, it would have made no difference. She was a woman, and I bat for the other team!'

Regards,

Alex!

this is all very amusing!

But doesn't this kind of underscore the point that of all the things that make sample librarys seem unrealistic, is sympathetic resonance of nearby instruments really that high on the list? I can imagine such resonance would make things sound different, but hardly more realistic. Does a woodwind quintet sound less realistic because it's missing the cello section through which to resonate? You could ask the same of any chamber group. Of course not!!

I would submit that until you fix the more glaring shortcomings of orchestral libraries - tedious articulation selection process, bad mixing by user, bad orchestration by user, bad composition by user - futzing about group resonance is a wasted effort.

Give me predictive articulation selection, automated mixing based on what's going on in the score, and a good ear, and that'd make a real difference! [:D]

Metro, I read your post carefully and offer the following....

Is sympathetic resonance really that high on the list?
In my opinion, yes. It's part of the mechanics of sound that create that 'live' sound.
Does a W/W quintet lose something because there's no strings to resonante through?
Yes. And i've played in many. W/W players the world over will tell you how difficult it is to creat a full and complete sound without the presence of other instruments. In a chamber of some sort be it large or small, the W/W RELY on the resonance from surrounding objects to enhance the sound. It's even worse playing outside, with little or nothing to aid the sound picture, and the lack of resonance makes playing and trying to give the audience a complete sound, hard work.
If you've played in a chamber group, you would understand string players in particular have to do more work trying to 'fill' the sound so it sounds complete. This is difficult to describe, but it's as if there is a instinctive acceptance that resonance is an important part of the complete sound, and players do their best to fill that resonant space dependent on dynamics and good balance of sound.
As you have described, bad mixing, bad orchestration, bad composition, all by USER.

So why do sampled libraries get the blame for these?

Tedious articulation process. You probably won't like this, but the fact that we have a library CAPABLE of seperate articulations is for me at least a wonderful step forward. When computer music was in its infancy, there were no articulations and the sounds were, compared to today, hopelessly synthetic. (I speak as one who writes orchestral music)

Much of the sampled technology we have today is young, maybe the last five years. There were attempts before then to provide a 'reality of sound', but in today's marketplace the standard has gone through the roof. And who knows what will happen in the future?

Predictive articulation selection.
I have a problem with this, at least in part. If yiu want to write in a particular style on a regular basis, then your suggestion has merit. If you wish to follow the same articulative pattern each time then ok.
But what if you don't? And if you wish to produce a unique sound of your own, you'll still have to build another 'automated articulation template' which may be beneficial in the long run, but as current music trands change and grow, may become redundant very quickly.

In one sense you are right though. The articulation input process in any library is a challenge to musicians and developers alike. And again, i refer to the days when we didn't have such a wonderful selection, and we had to work very hard, for long hours to reproduce the same articulated sounds you take for granted now. Imagine having to build your own each time, because there was little or no choice 10 and twenty years ago.

Automated Mixing process dependant on the style of the score?

Hmm...
Interesting concept. Only problem is, you would be bound to the same 'mixing style' as everyone else, and if an audience were to face a choice between your 'cloned' mix of a composition, and one that's maybe a little bright, or not neccesarily 'right', they may well choose imperfection over repetition!
I like the fact that things never turn out the same way twice, and i also appreciate the skill it take to mix well. I've met many a fine sound engineer in my earlier days, and discussions between them about the merit of brighter brass, or more sonorous strings was as passionate and loud as our discussions about articulation, or compositional worth!

Each one had his or her own 'style', unique, sometimes flawed, but individual just the same.

As for a good ear, sadly, no one can help you with this. Study and a lot of listening may help, but a good ear, the same as an instinct for composition, and a gift for orchestration are abilities not only reliant on study and learning, but your own natural ability or lack of it.

My regards to you,

Alex.

My goodness, Alex, what a thorough reply! Thanks for reading my post so carefully. Some of the stuff I wrote was meant in jest, at least partly. I've only played in a few orchestral or chamber settings, and we never worried about the resonance of our surroundings. (the "liveness' or reverb traits of the room, of course are a very different story) I'm not a professional performer, however, so i'll take your word for it. Maybe the preferred recording location for small ensembles should be inside furniture stores!! [:)]

The point of referencing all the things that USERs don't do well is this: resonance modelling won't fix any of those user problems. When you listen to user demos, the range of quality is stunningly huge. Some are amazingly unrealistic, others are very realistic. The demos that are exceedingly realistic, to me anyway, don't scream out "except for the lack of resonance modelling." They sound great period. So I'm not blaming orchestral libraries at all, rather pointing out that user error is the biggest factor in fake sounding orchestrations.

And Alex, I LOVE the fact that we have so many articulations! Back in 1988 I spent so much energy trying to make my new Korg M1 sound like an orchestra. The tools we have now are simply jaw-dropping in comparison. But with 129GB of VSL sample data comes new problems. I believe the future is going to be software-assisted articulation selection. It won't make your music any better, but it will make the lines you write sound natural with less effort. Of course, underneath it all, you'll always have the ability to go in and edit individual articulations. Having software help won't make it sound any more canned than a violinist who's spent years studying "proper technique" sounds canned. But we'll see - the proof will be in the hearing!

My automated mixing suggestion was mostly a joke. That probably wouldn't provide much more of a solution than a Finalizer, as mixing/mastering really is an art that can't be automated. MIR promises to be some sort of front end for a mixing environment, so who knows what might become a lot easier in the future.

And yes, my comment about my ear was also a joke!! I'm not really asking for VSL to fix my judgement in composition! The underlying point is, however, that a good ear can make almost any set of tools sound great. Just listening to Jay Bacal's example of the chamber strings was amazing. (the thread where the new user is complaining about the chamber strings) It sounded so good, and of course, it was done without any of my fantasy software help. VSL is simply stunning.

Anyway, Alex, thanks for an entertaining discussion...

Yes, it's true. An entertaining discussion.
And i admit to a large sense of humour when i wrote the reply!

My regards,

Alex.

part of the "user" side of things can be fixed on the sample/technological side. Right now there is quite a learning curve to go from classicaly trained composer to digital composer/engineer. Making the tools we use easier to use, more intuitive will greatly help in the creaton of realistic orchestral performances.