 The Clarinet BBoard
|
Author: Ken Shaw ★2017
Date: 2005-07-27 20:51
The maker of a high-end electronic sampling collection has issued what's said to be pretty good cut down version for $250. It has variable "nuance" functions to make the output more lifelike. The target purchasers are young composers.
http://digitalmedia.oreilly.com/2005/07/20/garritan.html
It's no longer "if" but "when" it replaces live musicians, at least for composers who can't afford to pay humans.
Ken Shaw
|
|
Reply To Message
|
|
Author: GBK
Date: 2005-07-27 21:02
I just listened to the Flight of the Bumblebee sample.
Sadly, the clarinet sounds pretty authentic...GBK
|
|
Reply To Message
|
|
Author: diz
Date: 2005-07-27 22:29
GBK - I couldn't find any of those samples - but then again I didn't look particularly long and hard, help? Also - why wouldn't the clarinet sound authentic when it is actually a clarinet that has been sampled?
Without music, the world would be grey, very grey.
|
|
Reply To Message
|
|
Author: msloss
Date: 2005-07-28 02:21
Right between the eyes.
The Moog signaled the end of acoustic performance 40 years ago. The Synclavier sounded the death knell 20 years ago. The Kurzweil 10 years ago. The VO last week. I'm guessing musicians bitched about the cathedral organ once upon a time. The users of "synthetic" instruments have been doing it for decades. The only thing that has really changed is that the tools have gotten more sophisticated and the end product higher quality.
In the age of electronic enlightenment, Williams, Horner, Goldsmith (RIP), etc. are still/have been writing for full orchestra for movies. The limited budgets haven't pushed the work to electronics -- they've pushed the work to Eastern bloc orchestras. Outsourcing baby. That's the wave of the future.
|
|
Reply To Message
|
|
Author: Morrigan
Date: 2005-07-28 09:10
Sure composers might use this, but it still doesn't sound like real musicians to me - I can't hear any nuance! Sounds more like super-quality MIDI.
I work with composers at uni regularly - the best ones are the ones that work with live musicians often, and they are the ones whose music is better, easier to play and well-written, and, in my opinion, learn faster and become more successful than the rest. My statement is backed up too; a good friend of mine recently won a very prestigous award for a piece for clarinet voice and percussion, and he worked VERY close with us, not to mention he'd taken some clarinet and voice lessons.
You can't tell me a composer is going to write great music with a synthesizer. [On the other hand, most in the past have only had a piano right?]
(Perhaps if they never intend for their music to be played by real humans this would be beneficial)
|
|
Reply To Message
|
|
Author: John Stackpole
Date: 2005-07-28 12:27
Gee, I always liked the "spacey" sound of the [I forget for the moment] device where you waved your arms about within a metal hoop thus changing the electronic characteristics - capacity - of the sound generating circuits and the tones produced.
Isn't this about the ONLY "instrument" around which does not depend on small muscle training and coordination? Ideal for us klutzes!
JDS
|
|
Reply To Message
|
|
Author: GBK
Date: 2005-07-28 12:35
John Stackpole wrote:
> I forget for the
> moment] device where you waved your arms about within a metal
> hoop thus changing the electronic characteristics - capacity -
> of the sound generating circuits and the tones produced.
Theremin
...GBK
/bring back Clara Rockmore
|
|
Reply To Message
|
|
Author: David Peacham
Date: 2005-07-28 14:00
Diz asked "Also - why wouldn't the clarinet sound authentic when it is actually a clarinet that has been sampled?"
Because the whole is more than the sum of its parts.
The sampling is effectively an accurate recording of a clarinet player playing individual notes. The synthesiser still has to put those notes together. In a real performance, the joins between the notes matter almost as much as the notes.
You can think of this as being a bit like those computer fonts that try to represent a person's handwriting. The font can easily represent how you write a b c d as individual letters. However, in real handwriting, letters influence each other: if I write the word "that", the first and last "t" don't look the same. A sophisticated font can allow for some of the influences, but it is too complicated a task to allow for all of them.
-----------
If there are so many people on this board unwilling or unable to have a civil and balanced discussion about important issues, then I shan't bother to post here any more.
To the great relief of many of you, no doubt.
|
|
Reply To Message
|
|
Author: Bob Phillips
Date: 2005-07-28 14:47
I've spent hours with Finale trying to get musical sounds out of it. NO success.
I looked into the program and found that there are several kinds of fiddle available: Stradavarius, etc., but no Cremora. Also doubles, triples and groups of 9 or 10, so you can call up an entire fiddle section.
There were, however, no separate files for R13s and Concertos.
BUT, you can call up your clarinets in solos, trios and harmony. Yum, you can make up a choir, a clarinet quartet (the three Bbs plus the Bass).
And, at $209 street (list is #250) plus the computer and speakers you have alread is cheaper than the meanest eBay bass.
I've been tempted to move from Finale to Sebelius in order to get improved instrument voicing, but have stuck with Finale in order to swap music files with colleagues. After I pay off the new thumbrest, I might just re-think on this new thing. I'd like to preview my stuff before exposing it to my playing friends. I've wasted their time with arrangements that sound just awful. With Finale, they can sound awful at home, alone; and I can seek advice on how to fix discreetly.
Bob Phillips
|
|
Reply To Message
|
|
Author: Gregory Smith ★2017
Date: 2005-07-28 14:57
D. Peacham said:
"The sampling is effectively an accurate recording of a clarinet player playing individual notes. The synthesiser still has to put those notes together. In a real performance, the joins between the notes matter almost as much as the notes."
--------------------------------------------------------------------------------------
I would go further and say that in a real performance, playing between the notes IS what matters most. How technology would be able to recreate that most essential of qualities, the qualities that make human musical expression unique, is beyond my understanding. I don't belive that it's possible to do such a thing. And I hope that I'm not wrong in my belief.
Gregory Smith
|
|
Reply To Message
|
|
Author: Markael
Date: 2005-07-28 15:00
The genie is out of the bottle and we will never be able to put him back in.
|
|
Reply To Message
|
|
Author: Mark Charette
Date: 2005-07-28 15:01
Bob Phillips wrote:
> I've spent hours with Finale trying to get musical sounds out
> of it. NO success.
Finale has absolutely nothing to do with sound production; your MIDI subsystem takes care of that. Finale (and Sibelius) just sends commands to the MIDI subsystem.
Look into getting yourself a good MIDI system.
|
|
Reply To Message
|
|
Author: Mark Charette
Date: 2005-07-28 15:12
Gregory Smith wrote:
> I would go further and say that in a real performance, playing
> between the notes IS what matters most. How technology would be
> able to recreate that most essential of qualities, the
> qualities that make human musical expression unique, is beyond
> my understanding. I don't belive that it's possible to do such
> a thing. And I hope that I'm not wrong in my belief.
It is possible already to connect the notes smoothly or articulated or tongued legato or anycombination thereof, with tone color changes to boot. I've played with doing that, and it's a very lengthy process for me and often results in real junk. I'm using my system as a toy rather than a real instrument.
However ...
At the moment, I'll wager the cost of an artist to put together such a track with multiple instruments outweighs the cost of just hiring people to play the instruments. However, with residuals involved, it might end up being cheaper sooner than later ...
|
|
Reply To Message
|
|
Author: Don Poulsen
Date: 2005-07-28 15:14
Here's the amateur and noncomposer chiming in.
Listening to a few of the tracks, it sounds to me as if the instrumental sounds are fairly faithfully reproduced but, as others have said, they lack nuance and sound somewhat mechanical to me. It's the difference between being able to play the music perfectly as notated and playing musically. For example, the solos in "Shenendoah" seem to lack the slight variations in volume and tone that a quality English horn player or flutist would provide in their interpretations.
If I started composing or orchestrating music, I think this would be a great tool to have, as it would either help me validate what was in my head or would allow me to experiment with different instrumental combinations sufficiently to get the job done. But, it wouldn't replace a live performance. If anyone should be concerned, it should be piano manufacturers over the loss of sales to composers.
|
|
Reply To Message
|
|
Author: Kalakos
Date: 2005-07-28 21:09
I was going to respond earlier, but didn't. Now I will, after thinking about it for a while longer. This is my personal feeling on the synthesizers etc..
It's kind of like going to a dance and the difference between having a disc jockey playing recorded music, or having a real band or orchestra. For me there is no comparison. I'd rather hear (and/or dance to) real, live music! All the rest is just "muzak" or as my kids called it "elevator music." For me, recorded real musicians are better than electronic synthesizer recorded, but for dances, shows, plays, etc., there is no substitute for live musicians. I guess my main reason is because with "live" musicians there is always a give and take and spontaneous creativity happening BETWEEN the players (and also the audience). No synthesizer music has this because even if one "live" person is playing the synthesizer, the tones are all pre recorded and the nuances (yes, and even the mistakes) are not there. The "magic" of music is removed.
Just my opinion, but what do I know. I'm a Greek folk musician who doesn't even read music!
Best regards,
John
Kalakos
Kalakos Music
http://www.TAdelphia.com
|
|
Reply To Message
|
|
Author: DavidBlumberg
Date: 2005-07-28 21:21
Not only is it possible to have the connections between the notes, but also the shading of tempi, etc.
The days of "frankenstein midi" are over. I can take a measure of music and have over 50 tempo changes just in that one measure - and I'm speaking about the difference between 120 and 122 MM - so it can be extremely subtle.
With that said, maybe it could replace live musicians for much studio work, but live concerts will remain forever.
|
|
Reply To Message
|
|
Author: DavidBlumberg
Date: 2005-07-28 21:24
Just listened to flight - doesn't sound much like a Clarinet to me at all!! Still sounds completely synthetic.
Strangely enough the old SoundBlaster AWE 64 has a quite realistic Clarinet sound. Not the upgraded SoundBlasters which have tons more memory - but don't sound very real like the 64 does.
Even the articulation of the 64 sounded like a Clarinet.
|
|
Reply To Message
|
|
Author: DavidBlumberg
Date: 2005-07-28 21:30
Mark - finale has everything to do with it as it is what loads those sound fonts (or another program which is compatible with those soundfonts).
Just any midi program won't do it.
Finale 2006 now is including Garritan Personal Orchestra as an option for $189 for current finale users.
|
|
Reply To Message
|
|
Author: Mark Charette
Date: 2005-07-28 22:13
DavidBlumberg wrote:
> Mark - finale has everything to do with it as it is what loads
> those sound fonts (or another program which is compatible with
> those soundfonts).
> Just any midi program won't do it.
I'm sorry, that's not correct. The synth (which is the software that loads/plays the notes) is controlled by the MIDI signals that Finale generates. Finale acts as a patch map and sequencer.
MIDI is a protocol, not software (I know, I wasn't clear, and people often confuse MIDI with the pieces of software that generate/interpret the MIDI signals)
Any MIDI patch/sequencer software can control the synth when you map it correctly.
|
|
Reply To Message
|
|
Author: DavidBlumberg
Date: 2005-07-28 22:22
I thought that some soundfonts were proprietary and designed for specific programs. So you are saying that it can be mapped with any midi program, even the $30 or free ones?
|
|
Reply To Message
|
|
Author: Mark Charette
Date: 2005-07-28 22:33
DavidBlumberg wrote:
> I thought that some soundfonts were proprietary and designed
> for specific programs.
The soundfonts or samples are (you can buy them separately). The synth that can access/play the soundfonts and samples is a different and sometimes expensive thing. The sequencer that causes the synth to play notes, change volume, do pitch bends, etc. is a different piece of software (logically, anyway, and is often separate in pro gear), and the piece of software that says which instrument or instruments to play on each channel is another piece of software (again, at least logically).
That doesn't mean that the $30 piece of software can load the soundfonts or samples (though a free piece of software I received with a soundcard can load/control/reshape all the Vienna soundfonts); the software has to be compatible with the fonts used. There are a number of synths that can load the same soundfonts. The whole purpose of the separation of the bits used to play music is to give the end user (composer, etc.) near infinite flexibility in the choice of tools.
oftentimes a program like Finale or Sibelius will bundle a relatively tame synth/sequencer/patch system with their software, and add a plug-in to bypass the MIDI subsystem (which today is somewhat limited). But that also means that another notation program or sequencer/patch system can use the synth. It also means you can replace the synth relatively easy if you need higher quality.
It also adds complexity for the average casual user ...
|
|
Reply To Message
|
|
Author: Gregory Smith ★2017
Date: 2005-07-28 23:24
Mark Charette said:
"It is possible already to connect the notes smoothly or articulated or tongued legato or any combination thereof, with tone color changes to boot."
-------------------------------------------------------------------------------
It may be possible to manipulate all of these inflections technologically, but are you saying that the end result is virtually indistinguishable from the real human voice or the real clarinet?
Are you saying that I could be fooled into believing that the technological end product - for instance, a clarinetist playing the slow movement of the Mozart Concerto - would sound as though an actual human being were playing? That if I were to hear it for the first time, I would actually say "Who is that?" not "What is that?" (not that I haven't actually asked "WHAT is that?" after hearing a couple of clarinetists over the years!) .
I still have a difficult time believing that it is possible. Do you or anyone else have sound clips of this so that I could hear it for myself?
Gregory Smith
Post Edited (2005-07-28 23:25)
|
|
Reply To Message
|
|
Author: Terry Stibal
Date: 2005-07-28 23:39
The agony that we hear now over the advent of mechanized music is nothing compared to that of the great sound revolution in the 1920's.
Remember, if you wanted "public music" in 1922, that meant a musical group. Every "quality hotel" had their own group or groups, every quality movie theater, every entertainment venue...all were provided with three to ten man (not many women back then) unit that got paid real money to make music.
Then, boom. Along came "modern recordings" (flat disks versus cylinder recording), amplification, radio, movies with sound. My grandfather told me that the AFM local dropped off over 50% between his arrival in the United States (1919) and 1930, and I believe what he told me.
Something new is always going to come along. Normal horn playing replaced multiple natural horns. Radio entertainment has covered so many musical bases that you can't even begin to list them. I remember when the Six Flags park up near Saint Louis had a ten piece pit; last time there (back in 1990 or so) it was down to a percussionist and a synth.
Facts of life in the musical world, and like it or not it's going to come. One day, those computers that just ten years ago were so limited will be able (with the right prompting by a musician or one skilled in the use of music software) to handle all of the nuance that currently isn't there.
Forty years ago, synths were whiney electronic machines. Sure, you got a pitch that was precise enough, but few would have compared it with "real music". They were a novelty, and no one thought them a threat to anything but good taste.
Now, they're getting almost good enough for most purposes. Not for our purposes, perhaps (I can almost always pick out when a string synth is being used, and certainly can tell when the clarinet and trombone parts are being "faked". But, for Joe and June Lunchpail who want to see a Broadway show but don't want to spend $95.00 a seat, it's more than good enough, despite our feelings to the contrary.
Ten years from now, who knows? I know that, with a wind controller and a good synthesizer, you can do some pretty amazing things. It's only going to get more amazing as our data crunching ability grows.
And, keep this well in mind: you can make a damn'd good living at any number of trades, and still play all of the clarinet that you want.
leader of Houston's Sounds Of The South Dance Orchestra
info@sotsdo.com
|
|
Reply To Message
|
|
Author: Mark Charette
Date: 2005-07-29 00:01
Gregory Smith wrote:
> Mark Charette said:
>
> "It is possible already to connect the notes smoothly or
> articulated or tongued legato or any combination thereof, with
> tone color changes to boot."
> -------------------------------------------------------------------------------
>
> It may be possible to manipulate all of these inflections
> technologically, but are you saying that the end result is
> virtually indistinguishable from the real human voice or the
> real clarinet?
I think it's technologically possible for an instrument or group of instruments, but currently improbable due to the tremendous effort that would be involved. The voice is more complex and further off since we also expect a different channel of "intelligence" (words as a second channel, not sounds).
Improbable today ... but not all that far off. Once the tools are sufficiently shaped for an artist, the artist will construct a new "thing". Right now the interface to the tools isn't sufficient (most were created by a programmer rather than an artist).
|
|
Reply To Message
|
|
Author: DavidBlumberg
Date: 2005-07-29 00:02
It absolutely is not possible currently.
The Clarinet tone is not the same.
It can however be done on Piano and a Pianist would most likely not be able to distinguish among real or not. There aren't many who can do it that realistically, but it can and is done.
|
|
Reply To Message
|
|
Author: Mark Charette
Date: 2005-07-29 00:22
DavidBlumberg wrote:
> It absolutely is not possible currently.
>
> The Clarinet tone is not the same.
David,
Somehow I think you haven't heard really good samples; since a particular sample is just a digitized recording (just like a CD), there will be times when it's spot on perfect (or at least undifferentiated from a CD recording of the exact same thing). The problem comes in the cheaper samples where the samples are recorded in only one manner and then re-sampled to create different pitches/timbres/articulations/etc. The more samples you have of those things, the less mathematical work to interpolate needs to be done, and the more "real" the sounds become - including, in the high end (and very expensive) samples, the correct key noises at a barely audible volume to add realism.
One of my workmates (and advertisers here) has a pretty complete professional recording studio in his home replete with tens of thousands of dollars in gear, and what I've heard him create is absolutely amazing.
|
|
Reply To Message
|
|
Author: DavidBlumberg
Date: 2005-07-29 00:58
Ok, I'll give you that - probably haven't heard everything out there. I have heard the Garritan and one other really high end sampler set but don't recall the name of it.
What's the name of the sample set that your friend has?
|
|
Reply To Message
|
|
Author: Mark Charette
Date: 2005-07-29 01:38
He uses the Kontakt software system with (I'd be guessing, but) probably $60K worth of professional samples. He provides "thickening" for a number of Motown and Electronic music groups based out of Detroit. In his spare time so far - he still needs his day job to make ends meet.
Just for reference, a starting point for sampled wind instruments in Kontact format would set you back near $1000 from East-West, and if you want something better you can always rent a studio, a producer, and a player ...
|
|
Reply To Message
|
|
Author: DavidBlumberg
Date: 2005-07-29 02:25
Ah! I just looked it up in my e-mails with that name and yes, that was the other system that I had heards of.
But haven't heard anything from them before.
|
|
Reply To Message
|
|
Author: Markael
Date: 2005-07-29 02:25
I would like to build onto Terry Stibel’s thoughts on technological change.
Jazz Bassist Milt Hinton played violin for silent movies as a kid. The advent of talkies led him to switch to bass.
The fallout from technological change is always somewhat unpredictable and paradoxical. Videos didn’t not put the movie houses out of business, as predicted.
You might say that the pipe organ was the first synthesizer.
Just about every digital piano has an “electric piano” sound. The Fender Rhodes type sound was once probably the closest approximation to the piano sound; now it is almost like another instrument that supplements but does not supplant the true sound of a piano.
Imitation is the sincerest form of flattery. None of us wants to be replaced by a synthesizer, but at least it is good to know that the real thing is the standard that is being copied. The real thing always will be the standard, even though there will also be more and more alternative musical sounds .
There will continue to be clarinet players blowing in high school and college marching and concert bands, and many will join the ranks of unemployed clarinetists, a troupe so often bemoaned on this board.
Is the glass half full or half empty?
You can look at the parade of clarinetists as increasing competition for a shrinking market. True, to a point, but also, they help stoke the thirst for good music. The fierce competition will ensure that the very best will put out good stuff, and the rest of us will want to hear it.
The downsizing of the orchestra pit is, from one perspective, just another example of mechanization that has cost some jobs, but in the process, has made goods and services more affordable for everyone, including poor musicians. We are, after all, consumers as well as producers of music.
How many of [italics] us could afford season tickets to the symphony, or to go dancing every Saturday night to a live orchestra on the scale of Shaw or Goodman?
You can attend small budget drama productions that include a scaled down orchestra, with perhaps a handful of wind instruments and a keyboard to fill the gaps.
Back in the classical days, wasn’t it mostly the wealthy who got the chance to hear Mozart and Hayden, et al? And would ordinary schmucks like most of us even get a chance to blow a clarinet?
|
|
Reply To Message
|
|
Author: msloss
Date: 2005-07-29 12:28
I almost posted yesterday again, but I wanted to see how the thread developed.
Yes, it is possible to do a great deal of what Greg asks -- essentially getting that "connective tissue" between notes, as well as shaping the interior of the individual note as well. The tone production algorithms are sophisticated enough to have parameters that equate to timbre, vibrato, intensity, etc. that can be controlled by gestures, wind velocity, lip pressure, finger strikes, etc. If the synthesis patch is well written, or the sample library is deep enough, you can do a great deal.
BUT... just because it can doesn't mean it does. I view these tools as raw material the same as a mpingo tree and a vat of molten brass. It takes a craftsman to make a tool out of it. Even then, a beautifully constructed clarinet is still just mpingo and metal until it is played well by an accomplished musician. These electronic instruments have to be PLAYED. With a wind controller, or even with a mouse if I'm that patient (I'm not) I can ellicit control over the instrument to get the nuances that differentiate true music from sequences of notes.
Even further, I am an accomplished wind player, and can get terrific results from "wind" patches, but I'm only a mediocre pianist and it shows with percussive patches, the same as if I put down my clarinet and sat down at a piano. Just because anybody can walk into Guitar Center and buy this stuff doesn't make it easy to replace "real" musicians. You still have to be a real musician to get quality results from the technology.
Philosophically, I'm opposed to the sample libraries as proxies for live instruments simply because even the best isn't good enough for art music. However, they are adequate for jingle houses and other non-artistic applications where the standard is lower. We can carp all day about the loss of jobs in that venue because of it, but turn back the clock and look at what multi-tracking did to the business. One violinist could come in and play an entire string section. One reed player could come in and play oboe, flute, clarinet, sax and bassoon. Where did those jobs go? It is what it is.
So, in the context of art music, and this is my grounding in experimental composition, the sample libraries can form the foundation for creating virtual instruments that could never exist in the real world -- a wind-powered violin, a bowed clarinet, etc. I can also play in pitch systems that all the microtonal technique in the world can't replicate on "real" instruments. I can create sounds that no physical object could generate, and place it in an acoustic space that could never exist.
As Greg asks, with the technology, is a sampled instrument or voice performed well indistinguishable from the real thing? Nope. The human voice is the white whale of sound synthesis. We are still 5 - 10 years away from a credible artificial speech technology that gets all of the formants, articulation and nuances of speech correct. A true singing voice is even further off. Sampled clarinet? No. The variations we are capable of are practically infinite with the real thing. Even with a very detailed controller, the sample library would have to be incredibly deep (translate to gigabytes, maybe terabytes of samples) to cover every possibility. From there all the gradations between samples have to be interpolated somehow. The potential to do this lies in physical modeling with a computer, and the processing horsepower doesn't exist yet to do this in real time. All the variables of the instrument, the chaotic flow of air, the shape of a player's oral cavity, embouchure, etc. make the model too complex right now. And let's say 5 years from now it gets done, somebody still has to play the model to make it behave like a clarinet. Think of it like trying to build a clarinet out of papier mache. With enough time and effort, it could be done, but the materials aren't suited to the task, and the result will be unsatisfactory. This technology isn't well suited to trying to recreate what we already have, it is for creating new things that are a logical progression from the raw materials (bits and bytes, capacitors and transistors) being used. Me thinks it simpler to buy an R13 and spend the technology on things I can't buy in a music store.
My studio is littered with MIDI toys using samples, FM synthesis, analog synthesis, hybrid, etc. And for all that, when I want to record an acoustic instrument I can't play (or play well), I call up a friend rather than calling up a patch.
|
|
Reply To Message
|
|
Author: David Spiegelthal ★2017
Date: 2005-07-29 13:53
Things a synthesizer will never be able to reproduce include the random, awful squawks/honks/squeaks and semi-chaotic flubs I (and others of my ilk) produce while playing (these of course add the 'human element" to our performances, yes?).
|
|
Reply To Message
|
|
Author: Mark Charette
Date: 2005-07-29 13:55
msloss wrote:
> The variations we are capable of are
> practically infinite with the real thing. Even with a very
> detailed controller, the sample library would have to be
> incredibly deep (translate to gigabytes, maybe terabytes of
> samples) to cover every possibility.
But, Mark, to record one piece of music doesn't require the depth of samples you'd need to cover all possibilities (and of course, the mathematics of the interpolation become simpler). If, as Greg asks, could just the Adagio be recorded via sampling/synthesis today without someone knowing that the "player" wasn't playing in a live recording, I think the answer is yes, it's within the technology of today (non-real-time, but that isn't a requirement, just as it's not a requirement for those computer-generated movies).
The whole idea of samples is to get the "right" ones so interpolation is really possible; I'm not sure that the sample libraries available for clarinet today are of sufficient quality since the demand isn't there; contrast the samples available for just trumpet with those available for the entire woodwind family - trumpet sample libraries are worlds ahead since the demand is significantly higher (I can buy a set of trumpet samples with 64 velocity levels; I'm lucky to get 6 on clarinet ... )
And of course you still need an artist to put the track together. I don't think that will change.
|
|
Reply To Message
|
|
Author: Hank Lehrer
Date: 2005-07-29 15:00
Hi,
Let me take this thread in a slightly different but closely-related direction.
When I was a young pup and made a living teaching people to fly airplanes, I completed a corporate jet type-rating in a Cessna Citation. I worked for FlightSafety as an instructor and one of the benefits was to be able to use the simulators for the CE-500 and CE 501 (nice benefit, right). After about 16 hours in the first officer seat and 16 hours as captain, I got into a “real airplane” for the first time. Three perfect instrument approaches to minimums was the result and I thanked the Cessna people for making the airplane just like the simulator (my little joke). But then, the simluator or the airplane did not just "fly" itself; a pilot is still needed to make many critical decisions too numerous to mention. However, an awful lot of routine things can be done by using technology.
My point is that this new and amazing technology could be used a training aid rather than a replacement. As I have indicated above, there can be a lot of realism found in training devices. Why not use this technology for many of the learning or trial experiences for musicians as well as composers? Am I making sense here?
HRL
PS I listened to a recording of a community college band I play with. On Morning, Noon, and Night in Vienna, the clarinet solo played by our principal almost makes me weep it is so beautiful in person. The recording did not in any way even begin to capture the warmth and presence of her live performance. I believe that I am safe in saying I do not expect to see technology replace musicians anytime soon.
Post Edited (2005-07-29 16:53)
|
|
Reply To Message
|
|
Author: Ken Shaw ★2017
Date: 2005-07-29 15:12
Michele Gingras has been playing concerts on various electronic wind instruments for at least 20 years, with the most delicate and subtle expression. I think she could plug in a good electronic clarinet voice and fool most of us.
At least as I read the Garritan site, his program permits the same kind of nuances as the Electronic Wind Instrument.
Of course it takes a musician to make music. And I think the discipline of learning to play an "old fashioned" instrument, plus the long exposure to music-making that this involves, is necessary to create the sort of musician who can make music with the Garritan orchestra.
Still, I wonder. My nephew is a high school senior and makes amazing music on the electric bass guitar -- a notoriously inflexible beast, as I'm sure Mark will agree.
A great musician can make great music on any instrument. A decent musician can make decent music on any instrument. For those whose musicianship outstrips our technical accomplishment, this could be a godsend, even though you have to learn to play it, just as you do any other instrument.
Ken Shaw
|
|
Reply To Message
|
|
Author: msloss
Date: 2005-07-29 17:00
Well, I dunno. The Mozart Adagio covers a lot of ground on the instrument, and is a stunningly transparent look at the clarinet. I honestly think all the different shades and colors required to make this movement sing would require a pretty broad sample set to pull off. One of the most difficult things about clarinet synthesis is articulation. In terms of pure tonal character, I can get pretty close using basic FM synthesis and not even use samples. The articulation is so much more than just an envelope. There is noise and contour that is highly complex and infinitely variable on the front and back end of every note. Very difficult to replicate with the kind of variety a real player employs.
Nothing is easy with clarinet. Now french horn, that's another story. That characteristic "frak" in french horn articulation can be mocked up in a way that it is pretty convincing. So much so that good synthetic articulation even masks a not so great sample or patch. Trumpet is a similar situation. Not easier to synthesize than clarinet, but easier to fool the ear with certain aural cues. And with all those great sample libraries, still ain't gonna get Bud Herseth or Phil Smith in a box.
If those thousand monkeys took a break from writing Hamlet to do this, eventually they'd get it. Possible? Yes. Practical? Not with today's technology. It is an interesting academic exercise, and totally not worth the investment. CGI hobbits and Jedi masters will rake in hundreds of millions. CG clarinet? I'm thinking slightly less...
And just to bring it back to composers, I have yet to meet a young composer that actually can afford to pay musicians to play his or her work. Once out of the ivory tower of school, fuggedabouddit. Sample library is better than nothin' t'all as a composition tool. Problem is the midi manuscript programs allow them to write and play back stuff that is absolutely unplayable in meatworld because the computer can do it, and then they can't understand why you can't play ppp staccato 64th notes at qtr=184 two octaves above the staff.
|
|
Reply To Message
|
|
The Clarinet Pages
|
 |
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
.jpg){kind=small}
