Re: [abcusers] problems with the R: field
Jack Campin wrote: I think there are already examples where extra information may need to be added in order to make abc unambiguous. A simple example is making | Ac Bd | sound a little more like |Ac Bd | simply by adding R:hornpipe to the header. except that hornpipes aren't always played dotted. You would need yet *another* level of extra information to say the style you're using is one where this dotted interpretation is appropriate. There is nothing to stop you defining "hornpipe 2" and entering an appropriate stress program for that. Henrik Norbeck has multiple definitions for Polska, identified by an extra letter following the word. I haven't included them in BarFly because I don't know enough about Scandinavian traditional music to get them right, but anyone who understands that music could do it for themselves, and if they send their stress programs to me I'll distribute them with the next version. The R: field is long due for deprecation. There is no standard list of what rhythms it covers and what to do with them, and nobody seems interested in making it extensible in any way that would allow different users to agree on what their extensions mean. Why not just let it die so that the name can be reused for something more important and more definable? I think a lot of people find it very useful, although I agree that it would be nice to have an interchange format for stress programs. And some of the rhythmic types found in folk music are unimplementable by any playback software. A slow strathspey is intrinsically a form where the player is *expected* to do their own thing with the rhythm. They are only ever played solo. What is a MIDI program supposed to do with this? Rubato driven by a random number generator? Anybody who expects that degree of interpretative ability from an abc program is deluding himself. He should be using MIDI to record a human musician instead. Phil Taylor To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
RE: [abcusers] problems with the R: field
I use the R: field in all my music. Since all my songs are old hymns, I put the meter in the R:, such as R: C.M. R: 8,7,8,7 R: 6-8 etc. John Henckel alt. mailto:[EMAIL PROTECTED] Zumbro Falls, Minnesota, USA (507) 753-2216 http://geocities.com/jdhenckel/ To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] problems with the R: field
De : Pax [EMAIL PROTECTED] A : [EMAIL PROTECTED] Objet : RE: [abcusers] problems with the R: field Date: dimanche 4 fvrier 2001 19:43 Hi All on the List With a truckload of respect I ask you to leave the R field alone. I use it all the time as I am only a new player of Irish Music and that field tells me whats what and when one has thousands of tunes one needs the R field. Most people would not know if a tune is an air or reel etc etc. Yes, R is a useful field. Please let it in the abc standard. Christophe To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] problems with the R: field
Phil Taylor writes:
I think a lot of people find it very useful, although I agree that it
would be nice to have an interchange format for stress programs.
And John Chambers writes
Something that I've thought could be useful in a player: People using
them to learn tunes could benefit from a basic sort of "style" list
that would modify tunes to fit a style. The point here would be to do
the standard, stereotypical things of that style. It's not a tool for
producing masterful music without human intervention; it's a tool for
helping novices learn the basics of a style. Most musical styles have
a lot of things that are conventionally done, often without the
musicians being very aware of what they're doing. Incorporating such
things into a player could lead to a good teaching tool.
I think these are very good ideas. If there were a standard
format for writing stress programs, and if someone cobbled together a GUI
to make them easy to write, then people would be able to produce--for
instance--tutorials for various styles for distribution on the
internet. ("This is how a roll on G should sound ~G Here it is in a
reel |DG~G2 BG~G2|cAFA dBcA|... . Press the slow-the-decorations button
to hear how the notes should be formed..." And even--there's no reason
not to be ambitious--- "This is the way the two-three slur in Sligo-style
bowing sounds...FAuAB AFvED|...") This wouldn't be as good as having an
instructor, or even an instructional video, but it would be better than
the written word alone, and would be compact, easily available, and
useful for a learner isolated from good instruction, as is (too) often the
case. Now that I think of it, if I ever get my hands on a Bulgarian
gaida, I'll be looking for something like this...
Cheers,
John Walsh
To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] problems with the R: field
Jack sez: | except that hornpipes aren't always played dotted. You would need | yet *another* level of extra information to say the style you're | using is one where this dotted interpretation is appropriate. ... | And some of the rhythmic types found in folk music are unimplementable | by any playback software. A slow strathspey is intrinsically a form | where the player is *expected* to do their own thing with the rhythm. | They are only ever played solo. What is a MIDI program supposed to do | with this? Rubato driven by a random number generator? This is true only if you're expecting an abc player to produce beautiful music. But this misses an important use of such players: A number of musicians who prefer to learn tunes by ear like to feed abc to a player, preferably in a "loop" mode, to learn the tune. Such people don't want any cleverness or randomness from the music. A "mechanical" version that is the same every time is just fine for their purposes. They are only trying to get the notes into their heads, after all, and then they will do their own interpreting. For such people, recognizing "R:hornpipe" or "R:shottish" (in all its variant spellings ;-) and playing the notes in a 3:2 or 2:1 ration is fine, and helps them get the basic feel into their fingers. The same would apply to a strathspey, which could be played exactly, or the dotted notes could be "overdotted" to match the style. But doing this mechanically and the same every time isn't a problem; it's useful for the person learning by ear. Something that I've thought could be useful in a player: People using them to learn tunes could benefit from a basic sort of "style" list that would modify tunes to fit a style. The point here would be to do the standard, stereotypical things of that style. It's not a tool for producing masterful music without human intervention; it's a tool for helping novices learn the basics of a style. Most musical styles have a lot of things that are conventionally done, often without the musicians being very aware of what they're doing. Incorporating such things into a player could lead to a good teaching tool. To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
Re: [abcusers] problems with the R: field
Jack Campin writes: The R: field is long due for deprecation. There is no standard list of what rhythms it covers and what to do with them, and nobody seems interested in making it extensible in any way that would allow different users to agree on what their extensions mean. Why not just let it die so that the name can be reused for something more important and more definable? Actually, it's used very effectively by Abcmus and (I think) a couple of other player programs to give tunes the proper accent and swing. It makes them a lot easier to listen to. It uses the tune type and meter to figure out the primary and secondary beats for emphasis, and does things with the relative length of notes to get closer to what a human might play. If you replace R:hornpipe with R:reel in the header of a hornpipe, it'll come out sounding like a reel (well, somewhat) and vice-versa. The list of tune-types is user-extensible, and the style for each tune-type can be defined in detail: you can adjust it to play the phrase |ABcd| anywhere between straight and |ABcd| (or, for military-style bagpiping, even further). I'm sure you could set it up to play middle-Eastern types too. This feature is designed for control of the rhythm, and I don't think it's possible (yet) to use it to define quarter-tone scales for specific tune-types, but I know there is support for them somewhere in the program: I had a couple of tunes come out sounding *very* strange: I was using a Q for some special purpose, and it turned out Henrik had a default which made the notation QA play the A a quarter-tone flat! So...hey! ask Henrik! John Chambers writes, Something that I've thought could be useful in a player: People using them to learn tunes could benefit from a basic sort of "style" list that would modify tunes to fit a style. The point here would be to do the standard, stereotypical things of that style. It's not a tool for producing masterful music without human intervention; it's a tool for helping novices learn the basics of a style. Check out abcmus... Cheers, John To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
