18 replies to this topic

[Ligneo Fistula]

I quickly rigged up some rough 'n' ready VBA in Excel (!) to generate random rhythms, from about 60 sub-units, at 3 levels of 'complexity'. However, I realise now that I don't strictly always count the pulse when I play. It seems simple patterns (like four semiquavers or dotted quaver-semiquaver) are parsed much like reading text rather than through counting each beat or sub-beat.
 
The issue now is that I'm struggling to cope with examples like this:

 
How would you go about tackling this.  Do you use pattern recognition while feeling the beat and therefore don't need to count, or count every beat or half-beat (tuplets excepted)?
 
Frustratingly I also seem to struggle counting when notes are tied:

 

What systems and strategies do you use?

 

Thanks in advance.

[Gordon Shumway]

Err, maybe depends on the composer and the speed (Bach, Bartok, Debussy?). If your examples were slow enough I'd count the quavers initially and do it all mathematically (same as playing 2 against 3, or 3 against 4 on the piano). Later, when I know the piece and have a feel for it, I do it by pattern recognition (or do I mean "feel"?), but randoms by definition have no pattern, so that's out. If you are grade 3-ish, I'd advise you not to let this kind of thing (overformulation of art) prevent you from looking at real music - it's a bit like trying to teach a kid table tennis with handcuffs on, maybe?

[elemimele]

Complicated question. I play mostly early music, where things are never random, but where notation is often an approximation of what should be played. For a complex slow movement with loads going on, I count initially in the smallest unit necessary for me to decode what's going on, probably quavers. As I get more familiar with the flow of the melody, I count longer (it's analogous to walking: when the landscape is unknown and complex, you plod along a step at a time, checking for trip hazards and pitfalls - as you get more confident and the way seems smoother, you raise your eyes and get the bigger picture, looking to the horizon and striding out with a sense of where you're going).

 

I'm full of admiration for both your programming skill and your ability to play random rhythms! On the whole, though, I prefer music that has patterns, and finding what that pattern is, so that I don't have to play each note metrically, is part of discovering a piece. Playing loads of music, so that these patterns are ingrained and immediately recognised, is part of learning to sight-read for me; sight-reading isn't about converting written notation to the corresponding sound, it's about recognising what a piece of music should be, from the markings on the paper, and producing that music with musicality. For that, I need at least some ability to see the patterns. The pulse is a row of regular fence-posts on which to hang the patterns.

[agricola]

If you don't count (feel) the pulse when you play I would start with that.  Complex rhythms must be understood according to the note values and not their visual spacing on the page although many people try (unsuccessfully) to do it that way.  I second both of the previous posts.

[LoneM]

Depending on the speed, I might find this difficult to play at sight. When I meet this stuff in orchestral music I sometimes mark it up with a pencilled vertical dash above the main beats. As others have said, you must start with the pulse.

[HelenVJ]

Those examples seem unnecessarily tortuous - reminding me of the Hindemith 'Elementary Training' book that we had to endure as students. Any resemblance to real music is co-incidental. Try some Bach (French Suites?) or Bartok ( Mikrokosmos, maybe the 3rd vol). Plenty of rhythmic training to be had there.

[Arundodonuts]

Two comments.

Yes those are complex rhythms.

It looks like they should be countable in quavers as others have said but I think I would start by counting in semi-quavers very slowly at first with a metronome

 

I agree with HelenVJ too, there are some ferocious rhythmic patterns in Bach requiring some diligent counting and sub-division.

[Ligneo Fistula]

Thanks for everyone's input.

 

For context, I'm trying to improve my sight reading, which has let me down from when I first started reading notation, so I'm attempting to play as much music as I can from different sources.  And with it poor rhythm recognition/timing is a recurring problem when trying to sight read, and I particularly struggle with tied notes and syncopation. Hence, the training app I cobbled together.  I accept now that the 'level 3' examples above are perhaps too ambitious for me, however I wanted to know how experienced musicians tackle rhythms like these (e.g. counting vs familiarity/feel) and, if the former, how to count.

[elemimele]

I'll admit I'm fundamentally lazy. I don't like to learn to do things that I won't have to do in real life. Unless in real life you want to play music that has complex patterns that largely disregard any underlying pulse, I don't think you'll often be encountering timings as complicated as those you showed. Nearly all traditional music can be subdivided into pulse units, which makes it much easier to know how to count. It's necessary to look at a bar or two of the music and spot what's happening: if it's in crotchet beats (from the time signature), for example, is each crotchet often being divided into a quaver and a lot of wiggly-stuff with loads of flags? If so, it might be good to count it in quavers on the first attempt. That way, if we have a quaver, a load of wiggles, and then another quaver, it might be obvious that the load of wiggles are spread evenly across one quaver-"beat", and you can treat them almost like a tuplet. Agricola is right that the spacing on the page is irrelevant, but the grouping on the page is not. Usually music composers and editors will have divided their notes in a way that exposes the pulse. Often this also makes it easier to count in a unit one (or even two) times shorter than the pulse, where necessary.

[LoneM]

Those examples seem unnecessarily tortuous - reminding me of the Hindemith 'Elementary Training' book that we had to endure as students. Any resemblance to real music is co-incidental.

 

I agree that they may be too complex for the OP's current stage, but they are not uncommon in 20th-century music. Think of the Rite of Spring!  Our amateur orchestra hasn't played that (yet) but I remember tricky rhythmic passages in Bernstein's West Side Story, and there have been other pieces where, as I wrote above, we found it helpful to mark the main beats in pencil.

[HelenVJ]

Sure! I'm not saying complex rhythms don't exist. But I can't think of much contemporary piano music that would be notated like this - and there's a reason for that .

[SingingPython]

Elemimele and LoneM have made some excellent points which resonate with my own experience.  I've played a lot of orchestral stuff as an amateur violinist and am also fundamentally lazy - I always aim to do something with the least effort necessary to reach a successful outcome!

When I looked at your examples, I did go straight to thinking in a crotchet pulse, but quite slowly.  I noticed that I aimed to feel my main pulse with a large motor movement - arm or foot, and also gain a sense of each rhythmic pattern continuing through / filling the beat.

When I was still in youth orchestra (had the luxury of being in a very good one up to age 24), if I saw something rhythmically complex my "go to" tactic was to listen to the brass - chances were, they had it to and I'd do whatever they were doing ...

And I agree absolutely that if I can't easily see where the pulse is in a passage, I will mark it in.  The first goal of sightreading is to be in the right place at the right time - even if you play nothing / no right notes.

[elemimele]

... related to that, it occurred to me yesterday that as a single-line instrument player, I still prefer to have sight of a version with the accompaniment. Usually the pulse will be far more obvious there. There are pitfalls; in old editions (but not often in new) the bass and melody might not be aligned properly, so there's a risk that a particularly twiddly quaver's worth of stuff in the melody might fill the space occupied by a whole crotchet in the bass, a misleading situation. But if the music is laid out well, it does mean you can see at a glance that a whole bundle of stuff in the melody must add up to a simple time-interval that can be read from the bass.

[Gran'piano]

... that a whole bundle of stuff in the melody...

-er,  I thought I was the only one who used such phrasing for a beautiful sequence of notes.

[EllieD]

Thanks to my teacher, I played Peter Dickinson's "Blue Rose" which is a take on the McDowell "To a Wild Rose", but in a blues style. I would never have tried it on my own, the rhythms looked scary! But once you've broken it down, played it slowly, got a feel for it - it's actually not that bad. The examples above aren't that scary if you clearly mark where each beat begins, I would start off counting in quavers. Where rhythm gets more complicated is having two different things going on at the same time. Imagine playing the first of LF's examples with the right hand, and the second with the left. Or being the violin solo doing the first while the orchestra plays the second. Now we're having fun!! 

 

BTW, I've looked at Grade 8 sight reading and the rhythms are nowhere near that complicated, I think only a seasoned pro would probably dive into those as sight reading, and I wouldn't hold it against them if they did make a mistake!