182,819 users have contributed to 42,261 threads and 254,932 posts.
In the past 24 hours, we have 3 new thread(s), 17 new post(s) and 44 new user(s).
I think that a lot of us forget that most sample use is technically physical modelling. For example, unless you play each articulation at exactly the dynamic that it was recorded, you are already modelling.
In my limited experience instruments like Samplemodeling's The Trumpet work very well for very dynamic music with lots of subtitles of articulation. However when you the put the legato up against the Vienna Trumpet legato, it is not in the same league.
DG
@DG said:
I think that a lot of us forget that most sample use is technically physical modelling.
Well, no. 😊 Physical Modeling is by definition not based on samples. It is based on a programmed model simulating the sonic events of an instrument by digitally recreating the physics. Actually, Wallander (sythesis) and Sample Modeling do both use no PM. For example Arturia Brass does. But I am not interested in how to achieve it, rather the result of it.
As I said earlier: VSL already broke some limitations of regular sampling and others followed that approach. Before that Gary Garritan was ahead of his time with his Orchestral Strings and the Maestro tool. It will always be like that! Others will come, invent something new and make older stuff look old. 😊 One must be naive to believe, that the Vienna Instrument in its present form will be up to it's time in 2-3 years from now. But the samples must not necessarily be outdated until then. I am pretty sure, that in the near future musicians are going to wish for more realtime playability with natural expression and that is something Wallander and Sample Modeling have achieved better. Has nothing to do with me not liking VSL anymore. On the contrary: I made substantial investment in it by owning almost every collection by now. I just hope they will see the signs of the time (that is a german term - I don't know if it translates literally) and apply some new technology to their products. Let's wait for the already announced Vienna Instrument v2 ...
And one comment on the Jazz versus Classic thing: up to now there was simply a lack of usable jazz samples, so developers tried to fill that gap. Maybe the Sample Modeling instruments are better for Jazz. That is certainly the main focus. But the same technology can be applied to a more classical style as well. Why shouldn't it?
Yes, physical modeling is actually TAMPERING with the basic acoustic nature of a sound, by altering the waveform. Layering dynamics is not the same thing. You are still hearing mainly a recording of a performance, not a newly created tonality.
VSL, unless I am mistaken, is fundamentally opposed to the alteration of waveforms that is seen with modeling. Because at least up till now, their philosophy has been similar to that of classical record companies - to represent as pristinely as possible the original sound created by extremely good orchestral musicians, and under no circumstances mess with its basic timbre. Of course legato or alternation instruments dissect the performances, but they are still heard purely though in a new time frame.
One thing that is good about this is it preserves the original musicianship of the players. What happens with some poor sap who plays an instrument that is then altered by ripping it to shreds and then reconstructing it with physical modeling? He is a non-entity, that is what happens, at lest in reference to his musical performance. But the VSL musicians are highly recognizable - for example the solo trumpet, flute, even the ensembles have very distinctive characters.
Not that I am opposed to physical modeling. It will probably become normal in the future but right now it has a very artificial sound compared to highly detailed sampling.
One other thing is how no one seems to understand that using samples in a serious MIDI performance is an art form unto itself. it is often actually MORE artistic than acoustic performances, even though that may seem heresy to the knee-jerk mentality which usually condemns MIDI.
It is not just a cheap substitute for orchestras in film scoring. If you listen to the Bacal Rite of Spring, or any number of Guy Bacos pieces, you will hear a true mastery that goes far beyond what is normally thought of as MIDI. It can be a powerful expressive medium for personal artistiic work. Jay Bacal stated that he derived a true musical pleasure from doing these performances, just (I might add) as a pianist or other soloist might, and Guy obviously has created original musical expressions of real beauty. I recently read an obnoxiously irritating book by Gerard Lanier, who thinks he is a big guru of Silicon Valley, and in this book he reduced MIDI to cell phone ring tones and "Up" and "Down" notes on the keyboard, stating that the original MIDI programming language is an example of old, outdated code that is simplistic and reduces the infinite possibilites of acoustic music to simple "On" and "Off" values. This assinine, incredibly stupid attitude is common among people who do not understand just how much can be done with this medium. It is a rare example of a programming language that is essentially infinite in its possibilities. VSL is a major part of that hugely expressive potential.
@William said:
[...] I recently read an obnoxiously irritating book by Gerard Lanier, who thinks he is a big guru of Silicon Valley, and in this book he reduced MIDI to cell phone ring tones and "Up" and "Down" notes on the keyboard, stating that the original MIDI programming language is an example of old, outdated code that is simplistic and reduces the infinite possibilites of acoustic music to simple "On" and "Off" value. [...]
Oh boy, did the guy really write this? [:^)] That would show about as much insight as I have in horse riding.
Yes, physical modeling is actually TAMPERING with the basic acoustic nature of a sound, by altering the waveform. Layering dynamics is not the same thing. You are still hearing mainly a recording of a performance, not a newly created tonality.
You make that sound like it has to be a bad thing. Samples are inherently limited to a number of velocities, while more advanced technologies can use fewer and interpolate between them, giving a continuum of sound. I'd argue that the PM approach has the potential to recreate the original sound more accurately than crossfading dynamic layers.
@mike connelly said:
You don't need me to put anything together. Just try it for yourself. The "problem" with the sample modelled trmpet legato is that it pre-supposes that a sort of slide/portamento is what a trumpet legato sounds like, when actually with the real thing (and VSL) you can clearly hear the skipping across the harmonic series. I have a suspicion that the sample modelled trumpet could do this, with improved scripting though. It's just that right now it can't.
I also think that the trombone will be more successful, because trombone legato is a sliding legato.
DG
Maybe I just haven't heard the right demos, I have listened to them and the legatos in the modeled version sound as good if not better to me.
And unless it's a specific situation where the player is trying to do more of a rip or other effect, legato trumpet shouldn't have audible in between notes, if the player is doing that he's flubbing the part. I would hope that the VSL brass legatos don't have "skipping" during the interval transitions.
No, he is talking about the harmonic series being faintly audible in a legato transition in brass. It is. And it is NOT the same thing as an electronic portamento and if you don't know that try to use a pitch bend wheel to simulate a horn or trumpet legato. It doesn't work. A brass instrument does NOT slide perfectly smoothly between notes, except in the case of the trombone actually using a slide glissando. So physical modeling would have to emulate (fake, as opposed to sampling it) this lack of perfect smoothness.
@William said:
No, he is talking about the harmonic series being faintly audible in a legato transition in brass. It is. And it is NOT the same thing as an electronic portamento and if you don't know that try to use a pitch bend wheel to simulate a horn or trumpet legato. It doesn't work. A brass instrument does NOT slide perfectly smoothly between notes, except in the case of the trombone actually using a slide glissando. So physical modeling would have to emulate (fake, as opposed to sampling it) this lack of perfect smoothness.
Of course it's not the same as portamento, I don't believe anyone here has said that it is.
And "the harmonic series being faintly audible in a legato transition in brass"? I wouldn't agree with that, at least not when played properly. If you're hearing that on brass playing, I'd love to hear an example of it.
@mike connelly said:
Maybe I just haven't heard the right demos, I have listened to them and the legatos in the modeled version sound as good if not better to me.
And unless it's a specific situation where the player is trying to do more of a rip or other effect, legato trumpet shouldn't have audible in between notes, if the player is doing that he's flubbing the part. I would hope that the VSL brass legatos don't have "skipping" during the interval transitions.
Mike, I think that we're just going to have to agree to disagree.
DG
@mike connelly said:
And "the harmonic series being faintly audible in a legato transition in brass"? I wouldn't agree with that, at least not when played properly. If you're hearing that on brass playing, I'd love to hear an example of it.
It is difficult to isolate an example because it is a basic part of the timbre of brass. Just as a slur on a violin is characteristic, or a slur in the human voice is different. But if you played a brass instrument you would instantly know what I was talking about. For example, on horn if you play a slur from a middle C (concert F) to the C above slowly, you hear C, E, G, Bflat (slightly out of tune natural seventh) and then C. If you then repeat increasing the speed of the slurs, which is a normal practice to get a smooth "singing" legato, those natural harmonics gradually are skimmed by faster and faster. And when you play a normal performance, they are almost completely gone, but still affect the way the slur sounds. And all brass do exactly the same thing on notes that are connected without valves or slides which a brass player tends to favor for getting a really good legato.
If you then repeat increasing the speed of the slurs, which is a normal practice to get a smooth "singing" legato, those natural harmonics gradually are skimmed by faster and faster. And when you play a normal performance, they are almost completely gone, but still affect theway the slur sounds.
When you play normally, they should be completely gone, if it's "almost" then you need to keep practicing because you're just playing inaccurately. A good brass player should be able to do a clean lip slur without hearing any of the notes in between, at all. If valves are changing it's even easier to do cleanly. I would hope any brass sample library (at least an orchestral one) with legato intervals would have clean slurs on all intervals. If the player doesn't do it right, do another take (although it would be cool to have "sloppy" slurs as a separate articulation or keyswitch, but that would probably be more useful for jazz or other playing than for most classical).
Mike Connelly -
I am not talking about rips or glissandi. You are simply wrong. I have played horn professionally for many years, and the characteristic sound of legato is created by the partials in between the notes sounding almost subliminally quickly but definitely. If you actually do play a brass instrument, i am amazed you don't know this.
You can refuse to agree if you wish, but that is the only thing that distinguishes one instruments legato from another - the characteristics of the slide between notes. It is NOT a perfectly smooth slur. If you want proof, then make a recording and slow it down drastically.. You will hear differences in the legato connections, and they are caused by the natural partials in between the main notes.
One other thing on this and I will drop it and you can go ahead and contradict me to your heart's desire -- an example of what I am talking about can be heard in two instances - one, have a horn player play a smooth non-fingered slur of an octave in a higher range. Then have him play it more slowly. Then even more slowly, continuing to slow the transition down. You will start to hear the partials in between the notes. it is only because they are passed by and warped slightly that they sound smooth. But they are not perfectly smooth in reality.
The other situation is take a trombone player and have him play a gliss quickly on the slide between notes a fifth apart. Then have him play the same interval at the same speed but using the lips only. They sound totally different. One is an absolutely smooth transition, with all possible intervals between the two main notes almost exactly like an electronic portamento, and the other has only the intervening partials (which are very few in an interval of a fifth). But the tongued-only is NOT just two notes played one after the other NOR IS IT totally smooth. And the slide transition which IS totally smooth shows the difference.
Glad to hear that you are a brass player. You say you're amazed that I don't "know" this - sorry, but when I play legato intervals, I am able to do it without hitting the partials in between. I don't know why you're hearing something I'm not, maybe your horn playing needs some work, or maybe horn players just have a lower standard in this respect? I will readily admit that horn is by far the hardest of the brass instruments. Again, I understand your explanation about the "slow legato" but I simply don't agree that playing over an interval slowly enough to hit the partials in between would be considered "legato" and I don't think most brass players would either.
No question that trombone glisses are a special case. Does VSL (or any sample based library) have a way of covering that aspect of playing with any sort of flexibility? That certainly does seem like a situation where physical modeling could probably do a better job of getting the playability.
When I get a chance, I'll record some intervals and post them so you can hear that they can be done without hearing the partials in between.
Anyway, back on topic, if you want legatos to have some "slop" of the in-between partials, there's no reason that can't be included in the modeling as well.
Oh, and a trombone can only gliss a maximum interval of a tritone, the slide isn't long enough for a fifth - if you actually do play a brass instrument, i am amazed you don't know this. [;)]
@mike connelly said:
When I get a chance, I'll record some intervals and post them so you can hear that they can be done without hearing the partials in between.
i am not talking about whether you can do a smooth slur. I am talking about acoustics. What causes the characteristic slur sound of horns, or trumpets, or tubas, is the partials in between the main notes.
They are not absent, they effect the quality of the slur in a subtle but unmistakably "brass-legato" way. As opposed to "string-legato"
or "clarinet-legato" or whatever.
That is what i am talking about - the acoustics that affect the quality of sound.
never mind.
One other thing besides this boring trivia about slurs - no one seemed to address this:
Is it not true that sampling of the kind that VSL does is opposed to "physical modeling"? In other words, tampering with the basic waveform of the sound. It has always seemed to me that they were trying to represent the "pure" sound created by the original instruments.
Nobody responded to that suggestion i made earlier. So you do not see any difference? To you it is all the same? Maybe you would like a trumpet created with a pure synthesized tone if it were accurate sounding? I am not saying that is bad, just asking. Now it is bad because it sounds fake. But in the (probably near) future maybe some company will be able to totally accurately reproduce a trumpet without any sampling except as a guide to how to do the waveforms, changes, etc. So that is as good as a sampled trumpet? maybe it is. But it seems like a principle evolved here, that basically was to be faithful to the original real acoustic sound of the players. Or is that wrong?
could one of you guys post a wavelab frequency spectrum to prove his cause? [8-|] this is really fun! please continue your brass debate! [;)]
@William said:
s it not true that sampling of the kind that VSL does is opposed to "physical modeling"? In other words, tampering with the basic waveform of the sound. It has always seemed to me that they were trying to represent the "pure" sound created by the original instruments.
I think it is in terms of pure physical modeling. The best sounding results are coming from the sample modeling instruments which are based on basic samples combined with shaping, what is kind of a unique and new approach. Those samples where recorded in an unechoic chamber, so I'm not sure, if that can be applied to the rather dry (but not completely dry) VSL Sounds...
