Have we eliminated the need for an orchestra?

[IOComposer]

I\'m hearing a lot of demos these days with ultra convincing orchestral emulations. Some that I honestly could not tell what was real and what was not...and I\'m a professional!

Have we finally arrived at the point where a sampled orchestra outshines the real thing?? Will Gary\'s strings actually make a sampled score PREFERABLE to the real thing? What keeps investors paying hundreds of thousands of dollars for live orchestras when a clever kid in his bedroom can whip up a comparable score for a couple thousand bucks worth of sample libraries?
-IO

[Damon]

I agree Jamey that orchestral mockups are getting extremely convincing. Have you heard James Newton Howards mockup for Dinosaur on the Keyboard mag site?(www.keyboardmag.com)
It\'s listed in the \'Features\' section under H. It is quite an impressive piece of work! They have the midi mockup and the real orchestral mockup and the only difference I really hear between the 2 are the brass (still a little synthy sounding in some places). Everything else is incredible. He must have sampled an orchestra himself because the pizzicatos are something I\'ve never heard before. Please take a listen if you haven\'t already. http://www.northernsounds.com/ubb/NonCGI/images/icons/wink.gif
Another person who has blowed my mind with mockups is Antonio Genovino from Italy. You can hear his music at mp3.com. His stuff sounds extremely believable.
I don\'t know if mockups will ever replace the real thing unless the composer can work in extremely fast time. I know it takes me a long time to do an orchestral piece, having to get expression to sound just right, tweaking out notes, etc. but I\'m sure there are guys out there who can crank stuff out in no time. I would love to know how long it took James to do his mockup for Dinosaur just to get an idea of how fast he can crank his stuff out.
There are some pretty incredible libraries out there as far as Xsamples concert harp, Gigastrings (which I\'m sure will replace all of the string libraries out there now), Miroslavs Brass and Woodwinds (I prefer his solo woodwinds to the enesembles), and AOs woodwinds. Brass still needs to be worked on as far as ensembles go IMO even though there are great sounds from within the available sample libraries, I\'ve yet to hear the holy grail of orchestral brass. Although there are great French horns on Miroslavs woodwinds and brass library and the trombone staccatos are fantastic. I\'m sure Nicks library has some great sounds as well from what I\'ve heard from his demo, I just haven\'t been able to afford it lately.
All I can say is if I hear a demo that sounds just like track 3, \"The Battle\" from Gladiator, I\'ll fall back in my seat. It would seem to me that making a mockup of that track would take forever, but I\'m sure there are some guys out there that can do it. I know I haven\'t reached that level of perfection yet. There\'s still so much to learn. http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif


[This message has been edited by Damon (edited 07-24-2001).]

[This message has been edited by Damon (edited 07-24-2001).]

[IOComposer]

I have heard the Dinosaur demos. That\'s not even close to some of the stuff I\'ve heard recently. I think I\'ve been hearing either custom stuff or very well EQ\'ed and mixed stuff from the standard libraries. I think that there are moments on Nick\'s new demo that definitely pass for ultra convincing.

However, that\'s not really the point of my post. I\'m really thinking about the ethical implications of what sample libraries and this development has on the traditions of the professional orchestra. What happens when our mock-ups start sounding so good that the producers and directors start asking, \"why don\'t we just cut in these tracks and save some money?\" It\'s bound to happen because of the foundation of the capitalist system. Buy low and sell high; spend a little, make a lot; Hire cheap labor and reap larger profits, etc.

Once this starts happening, the demand for orchestra will dwindle. It used to be \"ahh...that\'ll never happen...sampled orchestra will never be as good as the real thing\". But it\'s becoming as good. There are many examples now of it being as good and soon there will be floods of them. Soon, if you can\'t emulate an orchestra perfectly, then you won\'t work. It\'s already becoming that way in some of the circles that I find myself in. For a production that can\'t afford an orchestra, you\'re judged upon how well you can emulate one. The best emulation gets the gig. That\'s the way things have evolved and it\'s only going to be getting more and more like this from here on out.

Just curious as to what you people think of this evolution.

[Chadwick]

It\'s been a fact of life ever since someone used a moog bass instead of a real one. Ever since the LM-1 went into studios.

We\'re simply talking about a difference in scale.

[Thomas_J]

I think it\'s possible to emulate an orchestra so well, you could not tell it apart from a real orchestra playing. I can\'t say I\'m completely fooled by Nick Phoenix\'s QLB. Its strong side seems to be the loud staccato samples. That\'s not to say the library isn\'t good though!
Anyway, I would guess that such work (fooling a trained ear) requires first of all great samples, and second, great patience and third, great knowledge.
A hint of fresh creativity is also important. Listen to Hans Zimmer\'s Crimson Tide (Roll Tide) and if you listen closely you hear the french horns are actually the horns from miroslav vitous. (Could be a blend though) Loads of high freq. boost and rolloff on the mids and lows. And everybody is wondering how to get those horn samples.. http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif Use your creativity! If you don\'t have it, you\'ll probably never be a good composer http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif

I don\'t think anything can replace a real orchestra when it comes to feelings and devotion for music. You can probably create something that sounds real, but you\'ll never be able to get the real emotion that you get from live players.

Thomas

[Simon Ravn]

I don\'t think orchestras have to worry. I don\'t believe it\'ll be possible in any near future to emulate an orchestra using samples... Yes people like James Newton Howard and Hans Zimmer have some custom made libraries that can solve a lot of their jobs - samples that were recorded by these people, knowing how they compose and what kind of samples they would need. I am sure there are also a lot of sampled SEQUENCES and RUNS among them, because doing completely realistc fast playing strings I think will never be possible. There\'s just too much variation involved from note to note (and the fact that not every note is attacked also have something to say I believe).

[mahlon]

Well, while anything is possible, things don\'t always go forward technologically. There are always plateaus. Look how long we\'ve been working in music with the same technological tools for 20 years. In contrast, look how fast computer design and technology have developed into new territory. Think if music tech advanced that fast.

Yes, I think it will be possible at some point (assuming there are no major disasters in the world which set technology in general on that plateau), to right with some synthetic means and produce complex emotion, but it might be a while before we get to that. There\'s still a huge difference to me twixt live and the best midi, even though I\'d like there not to be.

Here\'s looking forward though,
Mahlon

[Endicott]

..

[This message has been edited by Endicott (edited 09-08-2001).]

[ursatz]

Agreeing with Simon and Endicott on this one - yes, the sampled sounds are getting better all the time, but no, there\'s a *long* way to go (if ever) before orchestras are superfluous. Fast runs, portamento, special effects of various kinds - there are *so* many things that a real orchestra can do as easily as falling off a log; it would take terabytes, maybe petabytes, of samples to emulate all of them. It\'s a big plus to have, e.g., harmonics, con sordino, and sul ponticello sounds in our strings. How about all three combined? Or any two? Add in all the other possible effects, and the explosion is unmanageable.

It\'s still true, and I suspect it will remain true for decades, that the really convincing realizations are produced by composing for the samples, not for a real orchestra. I\'ll eat my words when I\'m fooled by a realization of any movement of any Mahler symphony.

[Synth2k]

I agree that we are getting really close to emulating an orchestra with really great samples, but I think that we\'re still a few years off from being able to completely replace an orchestra with samples. I don\'t know how possible it is to capture the subtle nuances of an organic orchestral performance through samples. Every time you sit in front of an orchestra, conduct an orchestra or even conduct a small group of players the sound is different each time based on a number of factors and the performance and the way the instruments sound together each time can be completely different. Samples also can\'t represent the sound you get from the first violins when they\'re pissed off at you for giving the second violins a better part to play. :-)

I really think though, that we can use good orchestral samples to replace portions of or to enhance an orchestral recording. For instance on a limited budget that many composers are faced with when recording with real players, maybe funding will only allow for half of the string players that you really want, or half of the horn players that you require. Why not beef it up with samples to make up for the lack of the real thing? Maybe it\'s not better than the \"real thing\", but chances are it\'ll sound better than having gaping holes in your orchestration! And, I suppose if you can\'t get real players, using an entirely sampled digital orchestra with good samples is still going to blow your clients away. \"Wow, you did THAT on your keyboard?\"

Ryan.

------------------
Sonic Control
www..com (\"http://www..com\")
Giga Users Network at
www..com/gigasampler/ (\"http://www..com/gigasampler/\")

[This message has been edited by Synth2k (edited 07-24-2001).]

[SCARBEE]

If we look at the computer graphics used in Films and games the gab between reality and magic seem smaller for each year. If you look 10-15 years from now we will probably regard samples from year 2001 just as simple as we look at Fairlight samples today. We can not predict the technology of tomorrow, but when you see computer animations that can simulate waves (Titanic) it makes me sure that we will see VERY realistic sounding libraries. Someday we will have AI controlling the instruments, where we can buy different AI\'s - a player, conductor, etc. His way of conducting or playing will be analyzed and put into a morphing sampler.
So if you choose guitar and Santana you will have his way of playing and moving around the positions on his guitar making his famous riffs...

cheers

Thomas

[This message has been edited by SCARBEE (edited 07-24-2001).]

[Jamieh]

<BLOCKQUOTE><font size=\"1\" face=\"Verdana, Arial\">quote:</font><HR>I really think though, that we can use good orchestral samples to replace portions of or to enhance an orchestral recording. For instance on a limited budget that many composers are faced with when recording with real players, maybe funding will only allow for half of the string players that you really want, or half of the horn players that you require. Why not beef it up with samples to make up for the lack of the real thing? <HR></BLOCKQUOTE>

I agree with that statement. I have done that to beef up (or fix) mediocre live groups before. Blending samples in with a real recording sounds much better than an all sample recording IMO. I think blending is something that is being done quite a bit in Hollywood these days.

[PatS]

<BLOCKQUOTE><font size=\"1\" face=\"Verdana, Arial\">quote:</font><HR>Originally posted by Synth2k:
I agree that we are getting really close to emulating an orchestra with really great samples, but I think that we\'re still a few years off from being able to completely replace an orchestra with samples.<HR></BLOCKQUOTE>

Maybe I\'m over-reading your comment here, but do you really want samples to replace an orchestra? (I don\'t think that\'s what you\'re saying, but I\'ll play devil\'s advocate for the moment.) I definitely want samples that sound as good as the real thing (read: I like illusions as much as the next person), and I want performances based on these samples to sound as human as possible, if not specific to certain groups of humans, such as the LA Philharmonic. But I would never want them to replace a real orchestra, be it in the studio or in the concert hall. If anything, I want more musicians in these places, not less. Having said all that, I do recognize the need to use samples for cosmetic repair or to enlarge the sound of an orchestral recording. But completely replace an orchestra? Not in my lifetime, I hope! After all, consider not only the aesthetic value but also the social value of a live orchestral performance--the dynamic interaction between ensemble, audience and environment. How do you capture that?

[This message has been edited by PatS (edited 07-24-2001).]

[rkmusic]

Hi,

When you are in front of the flames of a fire you could be there a long time watching and watching... What happens really for all of us, humans, is that the forms and colors and sound of the flames are never, never the same, by themselves, and you notice that. What you see is impossible to be equal moment after moment. The mockup of an orchestra is done by a succession (or sequence, what an exact word to explain that!) of \"short movies\" (notes) so to say. And the samples are like photographs or little movies also, but since theses are recorded, what happens is like to watch the past always with repetitives scenes. The twoo situation are quite different. When you are listening a real orchestra you only watch (if you are listening) forward to the future because you have never, never listened this material before, not the notes, the composition, but the sound itself. I mean the perception is so fine...
Based on samples (past) a library will never replace an Orchestra, but it´s very usefull for mockups or real productions also, why not?
When I hear a good, very good orchestral demo, I am really amazed, because of my love for music, of course, but always \"something\" tells me that... You know what I mean?
This \"somthing\" is our human perception that is very large.
At the beginning of sampling the notes had to be looped, right? But since we have more memory capacity (RAM or HD) we like much the non looped samples, and then we like the different up & down bowings of strings, and then we have the consideration of alternative samples for repeted note like the \"c\" in a C-D-C-B phrase to avoid the repeted sample, because you hear them, and then..., and then...
The matter is the love for the real thing accordingly to our perception.
Anyway I am a defender of growing spectacular Orchrestral Libraries (not all of us, musician, can have the possibility of having the Chicago Philarmonic Orchestra under his fingers) for many reasons but the most important for me is the capacity of Creation.
Just an opinion (excuse my poor english)

Richard Krull

[tomhartman]

<BLOCKQUOTE><font size=\"1\" face=\"Verdana, Arial\">quote:</font><HR>Originally posted by Damon:
I agree Jamey that orchestral mockups are getting extremely convincing. Have you heard James Newton Howards mockup for Dinosaur on the Keyboard mag site?(www.keyboardmag.com)
It\'s listed in the \'Features\' section under H. It is quite an impressive piece of work!
[This message has been edited by Damon (edited 07-24-2001).]

[This message has been edited by Damon (edited 07-24-2001).]<HR></BLOCKQUOTE>


The Dinosaur demo is a mindblower. Even most of the brass sounds good to me...brass has been a real problem. Would love to know what horns he\'s using and what he did for the voices.

Strings were terrific. Once in a while you could hear the high violins as a sample, right at the end of the note, but geez, I\'ve used real players that didn\'t sound that good.

When you listen to the orchestral version the biggest difference is just a matter of air and presence. The demo is so good that it\'s fair to say the two are different, but I\"m not sure if they are worth what it cost to do the session with real guys...especially under dialog.

Most of the problems with doing convincing orchestrations, outside of the samples, has been dynamics to me....when I listen to a score or an orchestral work, the expressiveness is consistent...something hard to do with samples unless you can load every nuance imagineable. Howard did a great job with dynamics in this demo.

I do believe anyone who thinks they can consistently spot a sample nowadays is kidding themselves...unless it\'s a bad sample or the composer used it in stupid manner. There\'s plenty of junk out there, but things like this demo are pushing the edges of anyone\'s ability to discern the difference...very impressive.

[Jamieh]

I also agree that there is a long way to go. I have never heard a fully sampled piece that fooled me for more than a few seconds, especially if it tries to use brass. Zimmer has some stuff that is a blend of real and sampled instruments that can fool you for a bit, but if you have a good ear you can definitely tell it is sampled (like Crimson Tide)

The question in my mind is, knowing that the average film goer is a musical idiot, when will producers decide that they don\'t care if the music is fake because the audience doesn\'t care? Heck, audiences have already proven that they don\'t need a plot or charater development anymore. When will directors decide that they don\'t need a real orchestra for the soundtrack either?

Personally I find that my music is MUCH better when a real orchestra records it. While I can play in all the notes the exact way I hear it in my head, giving the music to a bunch of people that really know what they are doing with the individual instruments improves the quality greatly. Good players can make a good composition great.

[Endicott]

..

[This message has been edited by Endicott (edited 09-08-2001).]

[Chadwick]

I think Thomas is right, but that we are a very long way away from that day.

Just as Gigastudio got rid of the loop problem as well as offering us the clever addition of dimensional switching to aid performance, in the future (not soon), it may be possible to marshall all the elements of an orchestra just as a conductor does now.

In order to be able to have the organic variation pointed out by Endicott, we\'ll need to break the orchestra down to its individual components. ie record each member of the orchestra seperately.

That way, we might be able to assign some smart algorithm which \'plays\' all the players when we push C3.

If we CAN do this, then we run into the next biggest problem, which was pointed out in the Giga brass thread.

Even if the artists are recorded in situ on the same day with the same gear, you can\'t pull up four individual trumpets, combine them, and assume that you\'ll hear something which sounds like the four guys actually played the sample TOGETHER.

When an orchestra plays, all the indiviuals bend to the will of the ensemble. If three of the trumpets are a tad sharp, the fourth player will \'brighten\' his playing, if the other guys are playing behind the beat, he\'ll try to lay back a little.

So we have a conundrum.

We need ensemble recordings because when the \'whole\' of the orchestra plays together it creates something which sounds better than the \'sum\' of 100 individual recordings. On the other hand, in order to emulate the controlled anarchy of a living orchestra we need control over the individual elements.

These two seem mutually exclusive.

Even if they weren\'t, has anyone asked Gary how much time, head scratching and intense concentration went into HIS emulation of the string section?

Seems to me we might be talking about a lifetime commitment if samplers ended up capable of playing every available nuance.

Anyone want the job?

[Damon]

I totally agree with you Chadwick. To get an orchestral mockup to sound incredibly real takes alot of hard work and time. Sometimes I think it would be easier and faster to just have your sequenced parts translated into a program like Finale and hand it over to the orchestrator and conductor and not have to spend hours in front of the computer tweaking out individual velocity notes, expression, etc. I actually find that part of composing the most tedious and time consuming. I certainly would be sweating bullets if a director like Ridley Scott told me he needed the entire Gladiator soundtrack done entirely with orchestral samples in the small amount of time that most composers have to score a film. Talk about stress! http://www.northernsounds.com/ubb/NonCGI/images/icons/shocked.gif


[This message has been edited by Damon (edited 07-25-2001).]

[This message has been edited by Damon (edited 07-25-2001).]

[SCARBEE]

Let\'s not forget, that great music with lots of soul has been made using synths. Just thing of Vangelis score to Blade Runner. His trings sounds worked didn\'t they?. Also again the water in Titanic or the computer animated people that fell down, when the ship went down - didn\'t it have the right feeling? Who wasn\'t sacred of the Waves in \"The perfect storm\"? A real ocean would have millions of mere variations, but you don\'t need that.

I have went to many classical concerts (I am an Opera fan...) and my favorite composers - other than Opera are Brahms, Poulenc and Prokefief. I have heard concerts with different orchestras and also compared different CD\'s: The difference is HUGE. So what you want is the best sounding orchestra, not a second rate one.

It will be done. It is just a matter of time and money. Imagine recording each member of the Chicago Symphonic orchestra?

Still - a good song can have much feeling, even using bad quality samples.

Thomas

[This message has been edited by SCARBEE (edited 07-25-2001).]

[dnortana]

Some very interesting thoughts here on this thread.

While a totally convincing and realistic orchestra simulation may still be a ways off, I will not be very surprised when it becomes possible. Just look at what is being done with graphics in film. Imagine if the same talent and energy being applied to animation were turned to musical simulation - I think we\'ll be amazed....

But look at today\'s status of orchestral simulation. With a good piece of work, most of the public will not know the difference, or even be able to choose between the two in an A/B comparison,as one of you points out.

I see the current status of electronic music technology having two significant impacts.

One, the possibility for lesser known musicians to produce music for film/video for production companies that could not otherwise afford to have an original score. Canned, licensed music gets replaced, to some extent, by new, original work. But a guy seated on a stage with a keyboard and bank of Giga dcomputers is not going to replace the live orchestra.

This same thought applies to original music and audio CD\'s - just look at all the Indie labels. And some of it is very, very good.

Two, a brave, new, musical world has been opened to anyone with an interest and the means to make a \'relatively modest\' investment in equipment and samples. Joe Blow can now compose an orchestral piece - and hear it performed! There will be some musical genius appearing as a result of this technology being developed.

As an aside, I often wonder what a guy like Beethoven, or Mozart,or Haydn would do if somehow he got \'time-warped\' into my studio, and magically appeared seated in front of my Kurzweil, and all the synths and computers connected to it. Would that be a trip, or what!?

Regards,
Trond.

[SCARBEE]

[QUOTE]Originally posted by dnortana:
[B]I often wonder what a guy like Beethoven, or Mozart,or Haydn would do if somehow he got \'time-warped\' into my studio...

Beethoven would rage against pirates and Microsoft. Mozart would jump into the tecno-scene and require better loop facilities in Gigastudio and Haydn would make 100 more Symponies for Rare instruments.

Wagner would work for NemeSys/Tascam and be a library developer too. Listz would demo libraries at the NAMM shows.

Thomas

[gigaDiga]

Bach would have coded GigaStudio in the first place, invented a new midi format, and then STILL made the best music EVER made...

doesn\'t it make you sick http://www.northernsounds.com/ubb/NonCGI/images/icons/wink.gif

p.s. oh yeah and I did always think GigaStudio should have better looping facilities... http://www.northernsounds.com/ubb/NonCGI/images/icons/wink.gif

[This message has been edited by gigaDiga (edited 07-25-2001).]