tremolo with tuplets

[dobrek]

Hello,

How is tremolo used with triplets and other tuplets?

Swedish polskas often contain 1/4 triplets. The
attached file gives an example of this. As I play some of the polskas
rather slowly, I would like to sometimes use tremolo for the
triplets. How should I do this? Continuing with 1/32 or 1/64 tremolo
would be extremely difficult, and even if it worked, it would probably be
very ugly, because it would be a "fractional" tremolo, so to
speak. Should I adjust the velocity of the 1/32 or 1/64 notes?

Jaroslav

[Mark Robertson-Tessi]

Sometimes tremolo is played metered, but other times it is played free, so that is one option.

You could also subdivide the triplet 1/4 notes evenly, and then use that triplet speed for the whole piece. If you divided each triplet quarter into 4 strokes of tremolo, you would play the first measure:
4E, 4G#, 4B, 6A. Each measure would have 18 strokes, 6 per beat.

Cheers
MRT

[Jim Garber]

It is always a matter of taste, but frankly I can see no real reason to use tremolo at all, esp in a dance tune which is, after all, what a polska is. Tremolo is like a spice that you add to the food/melody. Too much and it overpowers everything. Just my very humble opinion.

BTW that notation is pretty strange with all those triplets. I am not sure that it is the right way to annotate that tune, but i also don't know that tune tho i have certainly heard other polskas. Do you have any examples of it being played?

[Jim Garber]

This may be a more sensible notation. At least I think this is the same tune annotated in 9/8 and played in this video. After hearing this beautiful and sort of mournful tune, I still would not use tremolo or use it sparsely for accents and flavoring.

[JeffD]

I would pick as written, no triplets, no tremolo. Teat that sound as ornamentation done with a quick little hammer-on. At least that would by my first try.

[dobrek]

Thanks to all who participated in this thread so far.

As concerns the tip to subdivide the triplet 1/4 notes evenly, I had also though of that. The problem with this method is that 1/16 notes would get 1.5 strokes and, which is even worse, dotted 1/8 notes would get 4.5 strokes.

Concerning the tip not to use any tremolo at all: First, this is a general question. I would like to know how tremolo is applied to tuplets, in general. Polskas aren't the only type of music that contains tuplets. Secondly, I do not use tremolo with all parts and repetitions of all polskas I play. (Actually I do not play the posted polska at all, it is just a polska I could find a picture of in the web.) I only want to use tremolo for certain parts of certain polskas. I agree that tremolo should not be overused.

[Jim Garber]

I am not quite understanding your question. Tuplets are just notes with differentiated rhythms so if you were to play them as tremolo, it would be like any other note. This is folk music and AFAIK there is no real ancient tradition of the mandolin in Swedish traditional music so I would use it where it feels right to you. If it were me, I would not tremolo on any tuplets, even the slower ones.

[dobrek]

This is folk music and AFAIK there is no real ancient tradition of the mandolin in Swedish traditional music so I would use it where it feels right to you..

My question was not about Swedish folk music. The polska was just an example.

I am not quite understanding your question. Tuplets are just notes with differentiated rhythms so if you were to play them as tremolo, it would be like any other note.

I believe in metered tremolo. Apart from the fact that I am not able to play "free" i.e. rhythmless tremolo, I don't believe that many people will like it, when they hear it, no matter if they know what a tremolo or a metered or a "free" tremolo is.

Now, if I want to use metered tremolo with tuplets, the notes will, in general, not get integer numbers of strokes. If, e.g., I want to use a 1/32 tremolo, then each quarter note will get 8 strokes. Hence, two quarter notes get 16 strokes. Hence, each note of a quarter triplet will get 16/3 or 5.333333333... strokes.

How do you play a note with 5 1/3 strokes?

(Sorry, I had to edit this, because it had wrong numbers first.)

[Jim Garber]

Sorry... I am no virtuoso player, but I think you are maybe getting too much into the math of this as opposed to the music. There is nothing wrong with studying and measuring your tremolo -- that is a good way to start out -- but at some point you need to listen to music and get the soul of it. Forget the metering after a while.

If you do want to approach playing this way, I would get a lesson or two or more from a highly trained teacher who may have a technique of approaching this. I hope others here can contribute to this discussion.

[dobrek]

Sorry... I am no virtuoso player, but I think you are maybe getting too much into the math of this as opposed to the music. There is nothing wrong with studying and measuring your tremolo -- that is a good way to start out -- but at some point you need to listen to music and get the soul of it. Forget the metering after a while.

Music is all about soul. But before you can dare to think about the soul of a piece of music, you have to be able to play the right notes and to play them cleanly and rhythmically. No math, no soul.

By the way: I think you contributed a very good and useful piece of math to this discussion yourself: By reinterpreting the polska as 9/8. I think you are right that the first notation of the polska is mistaken. The key to understanding this polska is to view it as a 9/8 tune. I.e. you used some math to help us approach the soul of the tune.

Thank you for that and for posting the video.

[Jim Garber]

I am not against using math to play music at all -- I think it is very necessary in the understanding of how music works and of course there is a strong purpose for the proper technique as well. As I noted, I am a folk musician who plays some classical as well and I do often return to the basics or follow recommendations that may very well be different from what I have learned in the past or that would give me some new insight.

However, I am also not a perfectionist and if i waited until every note in a piece was an absolute jewel and perfect before actually playing in public, then I might as well quit playing -- it will not happen.

BTW here is a relevant thread on the subject: Tremolo Stumbling Block.

[Mark Robertson-Tessi]

I'm not an advocate of strictly metered tremolo, so I wouldn't use the following approach myself, but for what it's worth:

Option 1: Speed up the strokes slightly on the triplets if you want evenness. If you are playing 1/32 notes on the quarter-notes, you will have 8 strokes per quarter note. Two quarters will be 16 strokes. When you get a triplet quarter, just speed up a little so you can fit 18 strokes, which is 6 strokes per triplet-quarter. Three triplet eighth notes take up one quarter. Instead of 8 strokes, play 9, which is three strokes per triplet eighth. Does the slight speed up bother you? Then:

Option 2: The other alternative is to subdivide by a multiple of 6, which was my original suggestion. You pointed out that 16th notes will get 1.5 strokes. If you are playing the piece slow enough that you'd tremolo a 16th note, then you can double the overall tremolo speed. 12 strokes for the quarter notes. 8 for each triplet quarter. 6 for and 8th note, 3 for the 16th note, 9 for the dotted eighth. Everything that's a 16th or higher (both triplet and not) will fit into that scheme.

But then a quintuplet comes along and ... aaaaahhh! Well then, 420 strokes to the quarter should handle everything down to 16th notes including quint- and septuplets.

Again, IMHO this is too much overthinking of it and personally would not be an approach I'd use, but it's an answer to your question based on math. I doubt that many people who have superb tremolo have broken it down to this extent, but there it is!

CHeers
MRT

[Jim Garber]

Mark: You expressed it much better than I could. Thanks. I am struggling a bit with all this but could not quite put my finger on it. The people who can break all this math down are the excellent players who are also excellent teachers. Some excellent players just do it and may not be able to explain how.

[JeffD]

I believe in metered tremolo. Apart from the fact that I am not able to play "free" i.e. rhythmless tremolo, I don't believe that many people will like it, when they hear it, no matter if they know what a tremolo or a metered or a "free" tremolo is.

Until recently, I did not understand that there was such a thing as metered tremolo. I always viewed tremolo as the way mandolins can extend a note, its our long bow tecnhique. As such I always shoot for a tremolo so fast that its outside of time, you hear and reconize just the extended note. Like a bird's warble, you don't focus on the pulse. You are not counting the hemi-demi-semiquavers.

I have been trying to look closely at what exactly I do, but I am pretty sure it is not metered. I don't, for example, slow it down when playing the tune at a slower BPM.

I have made a mental note to go back and listen to my favorite tremolo players again, and determine if they are metered or not. On my list of "dream" tremolo, was the playing of Butch Baldassari. I have to listen again more critically.

In any event, a non metered tremolo, if done in a satisfying way, would solve all these problems. Would it not?

I often do an extended tremolo with one or two brief little hammerings a fret up - when I play a Bach piece or some baroque piece usually played on brass. I kind of imitate the double note tremolo thing that brass players do. It works well.

[Beanzy]

I think it really needs looking at the use of the triplet in the first place.
They are used to give a 'hurry up' feeling or a 'push' to a line and normally fit in a space devoted to even note sub-divisions as defined by the time signature. They inject energy by compacting more into the same space.
So whatever sub-division you choose you're just trying to achieve the same 'push' or 'flick' with the tremolo version. This can be done by leaving a bit of a breath before packing the triplet in the time-slot remaining.
Not a strict use of the triplet space but that breath can give it more emphasis so it doesn't get lost in the wash of tremolo notes.

[dobrek]

Thanks again to all who participated in this thread.

I would like to add a few comments.

First, I was surprised to notice that the general opionion seemed to be
that adding all this theory to the practice would make the practice more
difficult. My idea was not - or is not - to make things more difficult, but to
make things easier for me. My main concern is not my tremolo but my
triplets (or, more generally, my tuplets). To play triplets (tuplets) in a
rhythmically precise way is difficult on any instrument. Triplets (tuplets) of eighths
are comparatively easy: You just play three (n) notes to one (mental) tap of your foot.
Triplets of quarters are much more difficult, especially
when played slowly. You have two play three notes to two (mental) taps of your
foot. I am able to do this, but I have never been content with my precision in
doing it. A friend of mine, a very good Jazz pianist and much better
musician than I, also says he thinks playing tuplets is difficult and that he
would like to be able to play his tuplets more precisely. Using tremolo would
be a great way to practice difficult tuplets. If there was a general and clean
solution how tremolo is used with tuplets (which doesn't seem to exist), I
would *always* use tremolo for practicing tuplets, even when I didn't plan to
use it in the final version of a piece. If you are able to produce a
rhythmically clean tremolo and if you know how many strokes each note of a
tuplet is supposed to get, then, the only thing you have to do to play the
tuplet in a rhythmically clean way is count. You can do this very slowly and
then play it faster and faster. When you master the tuplet, you can leave the
tremolo away.

I said above that I didn't believe in free tremolo. I do not only not believe
in free tremolo with "believe" having the sense as in "I do not believe in the
Euro.", I also do not believe in free tremolo with "believe" having the sense
as in "I do not believe in God.". Well, almost in this sense. I think it is
possible to produce a free tremolo, but virtually noone does it.

I have two major reasons for believing this:

First reason: I listened two a lot of tremolo players in my life. I listened
to brilliant players, good players and bad players. I listened to mandolin
tremolo, mandola tremolo, octave mandolin tremolo, balalaika tremolo, banjo
tremolo, guitar tremolo, violin tremolo, nyckelharpa tremolo and probably a
few other tremolos also. I never heard a free tremolo. All I could hear
sometimes was imprecise metered tremolo. But that is not the same
thing as a free tremolo.

I think there is a simple reason why noone uses a free tremolo: It is
extremely difficult to produce it. (My second reason why I don't believe in
free tremolo.) If you don't believe this, try the following experiment:

First, tap half notes with your left hand and quarter notes with your right
hand. I bet you will be able to do this with a very high precision and to keep
it up a long time. It is easy, because your left and your right hand can act
in a coordinated way. Now try something more complicated: Tap quarter notes
with your left hand and triplets of quarter notes with your right hand. This
is already more difficult, because the degree of independency of your hands is
higher, so to speak. Obviously, there are even more difficult patterns: three
taps left, seven taps right. Five taps left, thirteen taps right, ... But all
this, though difficult, is still quite easy compared to the next task: Tap
quarters with your left hand and tap your right hand *in no specific
tempo*. Your right hand may be much faster or slightly slower or whatever than
your left hand. I don't care as long as they have a certain degree of evenness
(such as tremolo) and as long as there is no relation whatsoever between the
rhythm of your left hand and the rhythm of your right hand (as should be the
case for free tremolo). How long can you do this? I bet that after a very
short time, your left hand and your right hand will be coordinated. Your right
hand will get "metered". Note that this exercise is still quite easy compared
to real free tremolo on, say, a mandolin. When you play a real free tremolo,
your mind has to take care of even more things that just quarters in the left
hand and fast and independent taps or strokes in the right hand. I.e. in such a
situation you are even more unlikely to produce a "free" tremolo.

You might object: But this or that great mandolinist does not know the notion
of free tremolo. Or: This or that great musician knows what a metered and what
a free tremolo is and he says he uses free tremolo.

It is one thing to master something and a completely different thing to know
how it works. Almost all people master at least one language. But I think it
is safe to assume that noone understands how a single natural language
works. If we understood one or two natural languages completely, i.e. if we
knew all the rules according to which they are used, we could implement this
knowledge in programs. We could write programs that mastered natural languages
in the same way in which humans master them. There would be perfect machine
translation and there would be robots we could talk to. Often people who
master something to a very high degree do not only not know how the thing they
do works, but even hold completely false beliefs about what
they do. An example: Almost all motorcyclists, and that includes brilliant motorcyclists,
believe that to drive a
right curve, they have to steer to the right. This is true for walking pace,
where no sloping position is used. At higher speeds, you have to steer
slightly to the left in order to get into the sloping position needed to drive
a right curve. If you try to drive a right curve by steering to the right at
such a speed, you will have an accident.

You might still object: O.k., let us assume for a moment that there is no such thing as a free tremolo.
Let us suppose that every tremolo is a metered tremolo. But obviously there are very many
great musicians who never worried about metered vs. free tremolo or who don't know the difference.
These musicians have a brilliant tremolo without thinking about it. So why should we bother? Why
should we practice metered tremolo?

That some persons who are extremely talented and who have been practicing
their instruments since the age of 4 many hours a day manage to do something without ever thinking
about it does not mean that it was a bad idea for me to think about it. And, who knows how good
these musicians would be, if they had done things in a more systematic way? They might be even better.

[Beanzy]

For me I've always liked Bickfords approach to tremolo, where he realises that we are trying to give the impression of a single sustained note with the tremolo. Anything else he contends is not tremolo merely rapidly played notes.

[Jim Garber]

Maybe I am more than a little confused by your approach. is your goal to play precise tuplets or to have a precise tremolo? Or both? You start talking about tuplets and then go off on a long set of arguments for metered tremolo. I would think it better to separate the two goals.

It sounds like the only reason you want to apply tremolo to tuplets is a method of subdividing the notes, which I certainly understand. However, musically, it makes little sense to me (with possibly the exception of quarter triplets or pieces played very slowly) to apply tremolo to these notes. If tremolo is merely a way to subdivide the notes in order to measure your tuplets precisely then it may work for you as a practice technique but would probably have little use in actual performance.

In terms of practicing a metered tremolo, I heartily endorse your right to do so, and would certainly think it would help in your playing cleanly and precisely. There is nothing wrong with learning the mechanics os playing on the road to mastery. in fact, i think that is what a good teacher can do for you -- open your eyes and ears to how things work or can work to make your playing improve. The best of musicians even those who may be considered masters of their instruments always talk about learning even after decades of playing.

Dobrek: I don't think you mentioned whether you work with a teacher or not. It sounds like you are deep into this, passionately involved and some one-to-one guidance would be highly useful to answer your questions. People here are interested and knowledgeable and this discussion may go on for pages, but there is really no substitution for qualified instruction.

[Mark Robertson-Tessi]

dobrek,

Upon further thought, I think the issue I'd have with trying to learn tuplets using consistently metered tremolo is that a tuplet is usually (but not always) polyrhythmic only for the duration of the tuplet. In other words, for a triplet you don't necessarily want to carry that polyrhythm of 3 through the rest of the non-triplet notes that precede or follow it. Taking the example of a quarter note triplets, to evenly meter a tremolo over this, your quarter-note beat would have to have minimum 3 strokes (or any multiple of 3). However, the rest of the piece often will not necessarily have that rhythmic undercurrent. Imposing a 3, 6 or 12 stroke flow under the basic quarter-note beat is (often) changing one of the rhythmic layers. In other words, you will be fighting the actual pulse of the fast subdivisions which are straight eighths, 16th, or 32nds.

You might be interested in looking for classes in Dalcroze Eurythmics. I took a 4 year sequence of it, and it delves into rhythm in ways that are both very strict and very liberating. Certainly there's a place for the mathematics of rhythm, but I think that's only one side of it!

Cheers
Mark RT

[manwithnoname]

I came to mandolin by way of percussion, and have given quite a bit of thought to tuplets, artificial rhythmic groupings, etc.

Dobrek, you're on the right track for playing accurate time. The key to all working with all rhythmic groupings is understanding fractions- more specifically, working with the lowest common denominator (LCD). For example, if you want to play quarter note triplets accurately, take the LCD of the quarter (2) and triplet (3), or 6. This, in effect, is 2 beats of normal eighth note triplets. To get the timing of the quarter note triplet, accent every other one. If the beat looks like this X--X--, you're going to play X-X-X-. This works for all subdivisions of any beat- sextuplets over 7, 2 vs. 3. vs. 4. vs. 5, etc.

Subdividing is key.

That said, you can easily come off sounding like a robot if you're doing math problems in your head while you're playing. You still have to feel it.

[JeffD]

This is a very interesting discussion to me, free tremolo versus metered tremolo.

The intervals of tremolo may occationally exactly match an integer subdivision of a quarter note, but I assure you that would be by pure accident. And in your comments, which I really enjoyed by the way, you did not mention the key indicator - which is I tremolo at the same rate regardless of the beats per minute. If I play a waltz, lets say, with some extended tremolo - I tremolo at the same rate if I play it up to dance tempo as I do if I play it as a slow air.

I cannot believe I am the only one.

[Mark Robertson-Tessi]

There are plenty of examples of free tremolo out there. I just took a few cuts on the Bluegrass Mando Extravaganza CD and slowed them down and the tremolo is what I'd call free. "Divisions" of 10, 11, 12 and 13 per bar in a waltz were evident. Slight speed fluctuations happened as well. You could start arguing that anything that's not even is just timing issues, but it sure doesn't sound like there's an attempt to make a subdivision. Rather, it sounds like free tremolo. Also, if you take the tapping exercise dobrek suggested and replace it with fast 2-finger drumming in one hand, you'll find that it's much easier to keep the tapping rates independent.

Cheers
MRT

[dobrek]

There are plenty of examples of free tremolo out there. I just took a few cuts on the Bluegrass Mando Extravaganza CD and slowed them down and the tremolo is what I'd call free.

Can you post a link to such an example? This sounds very interesting and I am always ready to change my mind in the light of new evidence.

[dobrek]

That said, you can easily come off sounding like a robot if you're doing math problems in your head while you're playing. You still have to feel it.

I think the first step in practicing something should always be to learn to play it with the highest precision possible. (I am often too lazy to follow this principle, though.) I wouldn't mind sounding like a robot at this stage. Only great musicians are able to sound like robots. When you are able to play something like a robot, all the movements have become automatic. Then you can forget all the theory and start feeling the music. Then there is no need to think while playing any more. I don't think that such a way of practicing destroys anything. If you like music, your feeling for it is strong enough. It won't get killed by a little precision and mathematics. Rather, the precision is the basis of your ability to "feel" the music and to express the way you feel it. A musician who is practicing to feel music without a solid technical basis is like a boxer training his will power without being able to use an uppercut.

Besides, to divide a few numbers and to count a few strokes is not *that* much mathematics.

[August Watters]

I don't think practicing playing with accurate time, when you're playing tuplets, is that much different than practicing playing, say, 8th notes in time. If you want to have fine control over how you play in tempo, the first step is to play the subdivisions of the beat accurately, whether it's quarters, 8th notes, or tuplets.

Awhile back we discussed a Buck White tune from the Bluegrass Extravaganza album. Buck was clearly playing six to a beat, except when he pushed the tempo a little bit and got more notes in. The background mandolin (probably Dawg) was playing a harmony part, also metered but behind the beat so the note changed an instant behind Buck. Both parts were examples of a metered tremolo, played expressively. The exceptions prove the rule, I think!