Re: [tuning] Digest Number 3961

Gene,

If I may step in here. I think that this is, in part, unresolveable, but one shouldn't lose faith in face of an inevitable lack of resolution. Each listener, each musician, brings her or his own experience of music into resolving for themselves which elements are more or less important, and through which devices music appears or is construed to be "expressive". In my experience, those listeners and musicians with intensive experience in sound recording, and in particular, commercial sound recording, have a lot of nuts and bolts information to offer, but also seem to bring in the greatest prejudices in this regard; in the end they are often re-producing cliches that ought not be left unchallenged.

But I do believe that there are some strategies to deal with this.

The first is to assume that the soundfont at hand is good enough for some music, and that making it work for your music is more a matter of adding musical elements -- dynamics, phrasing, rubato, perhaps a bit of portamento -- than of tinkering with the font itself. As you are a mathematician, I could well imagine that you have some interesting ideas about how to implement such elements -- deforming metrical time over the course of a phrase and complimenting this with dynamic change, for example. In electronic forms, these effects need not be overstated, in fact, they should probably be on the subtle side. (My experience has been that, for some reason or another, when it comes to electronic sounds, we either want an impressive totally vivid, flat, flashy surface, or we want a surface with irregularities, maybe even vulnerabilities, with which we may become tender, intimate. While there are a few electronic pieces which balance these tendencies, it's probably safest to go one way or the other.)

The second, flashy, strategy is to find a soundfont that is honest about being electronic -- that is, it doesn't bother with faithfully imitating any conventional instruments, but rather enjoys simply being a set of synthesized sounds. It works for pipe organs, it worked for analog synthezisers. (Does such a sound font exist?). With such a font, you might be able to sustain a fairly flat dynamic environment, but I suspect that a bit of shaping would be useful -- a la Wendy Carlos, whose timing is very elegant.

The third approach is the one you've taken so far, which is to say upfront that the music is in the pitches and the rhythms, and that the realization is not the music. This is a perfectly legitimate, platonic even, tact to take, but the audience response is bound to be limited. Not all listeners are platonists, and many listeners are probably not ever going to be able to accept the legitimacy of such an approach. I think you have to either be resigned to this situation or be prepared to throw some bonbons and pralines their way.

My own strategy is simply to produce scores for performers to realize in real time. That's what I do for my work where intonation isn't an issue, and I cannot imagine segregating my music around tuning. Honestly, I don't enjoy producing recordings; I don't really enjoy listening to recordings. In short, I'm increasingly a luddite when it comes to the age of mechanical reproduction. The shortcomings of my approach are obvious -- not many people are going to go to the efforts required to realize these pieces, and this will restrict my audience, but I am unwilling to compromise on this point, and I would much rather have a few people hear the piece right, than many people hear it in a compromised form. This is not elitism, this is caring about how people listen.

DJW

> > Message: 19 > Date: Fri, 17 Mar 2006 00:58:08 -0000
> From: "Gene Ward Smith" <genewardsmith@coolgoose.com>
> Subject: Re: How do microtonal people hear?
> > If someone wants to be helpful, what would they suggest as an
> alternative, better sf2 string font than the rather harsh one I used?
> It shouldn't be mushy, should bring out the parts with clarity, should
> not have tuning problems and should be able to convery emotional
> intensity.

Daniel,

Well put. I would just like to add that there are certain facts about
human perception that do play into this well beyond prejudices about
recording. Constant noticeable vibrato, for example, will always be
fatigueing to human ears, even though many will choose to suffer
through it. Likewise, extreme compression is also fatigueing. I can
listen to the whole range from very computerized to humanly expressive
and variable, and I do definitely still appreciate the in-between.
Personally, I find extreme vibrato to be really distrurbing and distracting
in either case, including in a live performance. Sure, this is still subjective,
and I can't speak for everyone. But I'll certainly do what I can to ask
everyone, electronic MIDI composers, and live instrumentalists both
to please use anything beyond very subtle vibrato only as an ornamental
effect. A flute player I heard in college drove me to nearly having an
actual headache playing one of my favorite composers: Piazzolla.
I ask you to please believe me: I wasn't trying to focus on the vibrato
and have a hang-up about it, I was trying to enjoy the music.

An extreme example of this would be if I was visiting Gene and
he played a piece for me, but turned his speakers up to 120dB
which also distorted his speakers. It would be painful, and I'd
rather leave than listen. There's NOTHING intellectual here.
Gene's examples were not at all *that* bad, but my simple point
is that it isn't about expression or subtlety (though those elements
will certainly make anything more broadly accessible), it's about
recordings simply not being excessively fatigueing on the ears,
first and foremost. Though I'll still push for anyone bothering
to record trying to make it as nice as they realistically can.

-Aaron

--- In tuning@yahoogroups.com, Daniel Wolf <djwolf@...> wrote:
>
>
> Gene,
>
> If I may step in here. I think that this is, in part, unresolveable, but
> one shouldn't lose faith in face of an inevitable lack of resolution.
> Each listener, each musician, brings her or his own experience of music
> into resolving for themselves which elements are more or less important,
> and through which devices music appears or is construed to be
> "expressive". In my experience, those listeners and musicians with
> intensive experience in sound recording, and in particular, commercial
> sound recording, have a lot of nuts and bolts information to offer, but
> also seem to bring in the greatest prejudices in this regard; in the end
> they are often re-producing cliches that ought not be left unchallenged.
>
> But I do believe that there are some strategies to deal with this.
>
> The first is to assume that the soundfont at hand is good enough for
> some music, and that making it work for your music is more a matter of
> adding musical elements -- dynamics, phrasing, rubato, perhaps a bit of
> portamento -- than of tinkering with the font itself. As you are a
> mathematician, I could well imagine that you have some interesting ideas
> about how to implement such elements -- deforming metrical time over the
> course of a phrase and complimenting this with dynamic change, for
> example. In electronic forms, these effects need not be overstated, in
> fact, they should probably be on the subtle side. (My experience has
> been that, for some reason or another, when it comes to electronic
> sounds, we either want an impressive totally vivid, flat, flashy
> surface, or we want a surface with irregularities, maybe even
> vulnerabilities, with which we may become tender, intimate. While there
> are a few electronic pieces which balance these tendencies, it's
> probably safest to go one way or the other.)
>
> The second, flashy, strategy is to find a soundfont that is honest about
> being electronic -- that is, it doesn't bother with faithfully imitating
> any conventional instruments, but rather enjoys simply being a set of
> synthesized sounds. It works for pipe organs, it worked for analog
> synthezisers. (Does such a sound font exist?). With such a font, you
> might be able to sustain a fairly flat dynamic environment, but I
> suspect that a bit of shaping would be useful -- a la Wendy Carlos,
> whose timing is very elegant.
>
> The third approach is the one you've taken so far, which is to say
> upfront that the music is in the pitches and the rhythms, and that the
> realization is not the music. This is a perfectly legitimate, platonic
> even, tact to take, but the audience response is bound to be limited.
> Not all listeners are platonists, and many listeners are probably not
> ever going to be able to accept the legitimacy of such an approach. I
> think you have to either be resigned to this situation or be prepared to
> throw some bonbons and pralines their way.
>
> My own strategy is simply to produce scores for performers to realize in
> real time. That's what I do for my work where intonation isn't an issue,
> and I cannot imagine segregating my music around tuning. Honestly, I
> don't enjoy producing recordings; I don't really enjoy listening to
> recordings. In short, I'm increasingly a luddite when it comes to the
> age of mechanical reproduction. The shortcomings of my approach are
> obvious -- not many people are going to go to the efforts required to
> realize these pieces, and this will restrict my audience, but I am
> unwilling to compromise on this point, and I would much rather have a
> few people hear the piece right, than many people hear it in a
> compromised form. This is not elitism, this is caring about how people
> listen.
>
> DJW
>
>
> >
> > Message: 19
> > Date: Fri, 17 Mar 2006 00:58:08 -0000
> > From: "Gene Ward Smith" <genewardsmith@...>
> > Subject: Re: How do microtonal people hear?
> >
>
> > If someone wants to be helpful, what would they suggest as an
> > alternative, better sf2 string font than the rather harsh one I used?
> > It shouldn't be mushy, should bring out the parts with clarity, should
> > not have tuning problems and should be able to convery emotional
> > intensity.
>

--- In tuning@yahoogroups.com, "Aaron Wolf" <backfromthesilo@...> wrote:

> An extreme example of this would be if I was visiting Gene and
> he played a piece for me, but turned his speakers up to 120dB
> which also distorted his speakers. It would be painful, and I'd
> rather leave than listen. There's NOTHING intellectual here.
> Gene's examples were not at all *that* bad, but my simple point
> is that it isn't about expression or subtlety (though those elements
> will certainly make anything more broadly accessible), it's about
> recordings simply not being excessively fatigueing on the ears,
> first and foremost. Though I'll still push for anyone bothering
> to record trying to make it as nice as they realistically can.

One fact of life about using soundfonts is that they are based on
samples, and the samples are likely to have vibrato. I don't know why
strings are so difficult to get right, but I've claimed before (though
some people tell me I'm all wet) that the strings are the limiting
factor in current soundfonts.