Case for incremental speed increases

[nick]

Thought I would pass my latest observations with trying Chang's method of practice. After the second time in 2 years of trying the small sections of passages at 1.5 times the practise speed HS, I now understand why I given it up about a year ago: quality of each tone. Because one is trying to maximize speed, the notes are not 'full' tones, they sound thinner, and I noticed it is because the key is not depressed to key bottom. This second experimentation lasted about a month, and once I switched back to the gradual increasing speed on the metronome method, the sound was soooooo much better. Other posts had mentioned the mistake of this method, comparing it to someone trying to walk faster and faster until they are running, which because different muscles are used in the two activities, would never result in the goal speed. I learned how to type in this manner, by starting slowly once the keys were learned, and gradually increased my speed over time until I now can type very fast. Seems this analogy is more closely related to piano playing as well.Some may say 'just think how fast you could type with the Chang method'! The goal has been reached in that regard, so not necessary. So it would seem that whatever muscles are needed when playing fast will be developed as the speed is gradually increased. I have heard the 'speed wall' term, that one can move up on the metronome and then all of a sudden one can not go faster, as though to go faster from say metronome speed of 100 to 102 would require a different set of muscles. Holy Cow. Is that possible!! I think not. So if not there, then when? 112? So at 110 no problem but 2 more numbers, no way?

   Glad to put Changs method to rest and continue where I left off, at 92. Anyone have similar experience?

Nick

     

[Sketchee]

Eventually you will hit some speed wall.  You're definitely not going to be able to play faster and faster to infinite speed!.

[bernhard]

Eventually you will hit some speed wall.  You're definitely not going to be able to play faster and faster to infinite speed!.

Er…

Actually you can get to infinite speed (at least with a few notes): Play them as a chord (what could be faster than together?)

[bernhard]

Nick:

as though to go faster from say metronome speed of 100 to 102 would require a different set of muscles. Holy Cow. Is that possible!! I think not.

Not a different set of muscles, but a different set of movements. The difference between walking and running is not primarily in the muscles, but in the movement (the movement looks, feels and uses muscles differently). So yes. To go faster you will need a different set of motions.

In any case, you have no way of consciously willing a specific muscle/set of muscles to contract. All you can do is consciously will a certain motion – which will in turn use different muscles.

It seems to me that you are missing the point Chang makes (actually many other people have made the same point before, most famously Edwin Fisher).

He is not against gradual increments in speed or even slow practice. He is against starting from scratch with slow practice because you will not fathom what the correct movement at speed is, and you will be practising the wrong movements which will create a speed wall.

Therefore he says that you should start at top speed (by playing a few notes as a chord and slowing down) to figure out the appropriate movement at speed, and once you figure this out, then and only then you should do slow motion practice. In fact he actually says that slow practice is essential, as long as it is slow motion practice.

Just as you will never be able to run by speeding up walking motions (you will just walk faster and hit a speed wall). You need a different set of motions, and you can only figure out what these are by either running, or by watching someone run and trying to emulate such motions. Once you figure out the correct motion, by all means slow down and do slow (slow motion) practice.

By the way, certain motions cannot be done in slow motion any way. (Try juggling, skipping rope or water skiing).

Best wishes,
Bernhard.

[Sketchee]

Er…

Actually you can get to infinite speed (at least with a few notes): Play them as a chord (what could be faster than together?)

It might be pretty impressive to see someone play a whole Chopin etude in one chord with only their five fingers!  

[bernhard]

It might be pretty impressive to see someone play a whole Chopin etude in one chord with only their five fingers!  

No need to limit yourself unnecessarily.

Just lie down with your whole body on the keys.

[nick]

Nick:


Not a different set of muscles, but a different set of movements. The difference between walking and running is not primarily in the muscles, but in the movement (the movement looks, feels and uses muscles differently). So yes. To go faster you will need a different set of motions.

I don't see the analogy to runnng as correct. In running vs. walking, the foot must cover a much longer step, more "ground " to get to such a fast speed, no amount of speeding up walking can possibly achieve. In slower piano playing vs. faster playing, each finger stroke(analagous to each foot hitting the ground) covers the same distance, key top to key bottom. This is why typing works with the gradual increasing method. Ask a person who types fast besides me. The fingers are just moving faster, but from key top to key bottom. As a fact, the movement is even smaller as the speed increases in piano playing as videos of pianists show. Any difference in movement from one speed to the next I would think would occur gradually as one moves up in speed. Again, I can't imagine achieving a certain  speed where 2 numbers higher  would be a problem. Even if a slightly different movement were needed to play 2 numbers higher, wouldn't one just figure how to do it?  

It seems to me that you are missing the point Chang makes (actually many other people have made the same point before, most famously Edwin Fisher).

He is not against gradual increments in speed or even slow practice. He is against starting from scratch with slow practice because you will not fathom what the correct movement at speed is, and you will be practising the wrong movements which will create a speed wall.

Therefore he says that you should start at top speed (by playing a few notes as a chord and slowing down) to figure out the appropriate movement at speed, and once you figure this out, then and only then you should do slow motion practice. In fact he actually says that slow practice is essential, as long as it is slow motion practice.

I must have missed the part about gradual increments in speed. From my reading of Chang, once the notes are learned well, there is little slow practice. The essential argument of his book is the fast practice of small segments.

Just as you will never be able to run by speeding up walking motions (you will just walk faster and hit a speed wall). You need a different set of motions, and you can only figure out what these are by either running, or by watching someone run and trying to emulate such motions. Once you figure out the correct motion, by all means slow down and do slow (slow motion) practice.

They are two different activities. It is possible to  actually run slower than someone walks.(slow running = jogging). Not analogous to piano playing in my opinion. Piano playing uses much smaller muscles, which need to be gradually developed to handle the higher speeds without wrong tension.  

By the way, certain motions cannot be done in slow motion any way. (Try juggling, skipping rope or water skiing).

I am not concerned with whether or not EVERY activity can be done  incrementally  speeding up, only with piano playing for speed. I have though, seen a method of juggling, where they use scarfs to throw up in the air since it slows the speed of movement. I also have seen people skip rope very slowly, and then speed up to very fast speeds. In water skiing the person just stands there waiting for the boat to go. The speed of the boat can either be the minimum required to pull the person up, which I have seen can be pretty slow depending on the weight of the skier, or very fast.

I will see in a pretty short time if I hit the "speed wall " before reaching my goal speed. As I mentioned, the main reason I left the other method was the sound of each tone. I really love the sound I am getting now so will continue.

Nick

[bernhard]

I don't see the analogy to runnng as correct. In running vs. walking, the foot must cover a much longer step, more "ground " to get to such a fast speed, no amount of speeding up walking can possibly achieve. In slower piano playing vs. faster playing, each finger stroke(analagous to each foot hitting the ground) covers the same distance, key top to key bottom. This is why typing works with the gradual increasing method. Ask a person who types fast besides me. The fingers are just moving faster, but from key top to key bottom. As a fact, the movement is even smaller as the speed increases in piano playing as videos of pianists show. Any difference in movement from one speed to the next I would think would occur gradually as one moves up in speed. Again, I can't imagine achieving a certain  speed where 2 numbers higher  would be a problem. Even if a slightly different movement were needed to play 2 numbers higher, wouldn't one just figure how to do it?  

Er…

A huge difference between piano playing and typing is that the piano keys are spread over 1.20m  and have 2 levels (black and white keys), while a typist keyboard will fall pretty much under the hands. The complexity of movement and co-ordination required to negotiate piano playing cannot be compared to typing for this reason amongst others. You talk about the movement needed to move a key from top to bottom. This is just one of the myriad of movements required in piano playing, and certainly not the one should be concentrating on, in the sense that the finger should be the very last link of a complex chain of movements that should start at the shoulder level and sometimes even further back.

A simple example will suffice. Get a beginner to play an arpeggio slowly. S/he will get away with several movements and co-ordinations that will prove impossible to do at fast speed. Increasing the speed of these inappropriate movements will increase the speed of the arpeggio only to a point, after which there will be speed wall that no amount of practising these movements will surpass. One such movement prevalent even amongst non-beginners  is passing the thumb under to negotiate the 1.20m of the piano keyboard. The correct movement for playing arpeggios at speed is a lateral shift of the hand combined with a pivot on the 3rd/4th finger. This movement can only be figured out if someone shows you, or if you investigate speed playing by using the chord trick and – for the moment – ignoring accuracy and subtleties of touch. Once you figure out this movement, then yes, go ahead and investigate touch subtleties and accuracy by practising slowly the movement you figured out by speed practising.

I must have missed the part about gradual increments in speed. From my reading of Chang, once the notes are learned well, there is little slow practice. The essential argument of his book is the fast practice of small segments.

Yes, we tend to read stuff and select what we take from it. I strongly suggest you read Chang’s book again. I have read  and reread it hundreds of times and I always find something new, or that I now understand in quite a different way.

They are two different activities. It is possible to  actually run slower than someone walks.(slow running = jogging).

Thank you. My point exactly. Jogging is running in slow motion. If you want to learn how to run you start jogging, not walking as fast you possibly can. (There is actually a sport in which you walk as fast as you can).

Not analogous to piano playing in my opinion. Piano playing uses much smaller muscles, which need to be gradually developed to handle the higher speeds without wrong tension.  

I go back to my arpeggio example. Passing the thumb under uses different muscles than shifting the hand. So practising slowly some thing inappropriate will not develop the muscles appropriate for the correct movement. Once again the obvious distinction is not between slow or fast practice, but between slow practice and slow motion practice.

I am not concerned with whether or not EVERY activity can be done  incrementally  speeding up, only with piano playing for speed.

Of course you can and should practise the piano incrementally speeding up as long as you are doing it in slow motion. This means that the movements that you perform slowly are exactly the same movements that you would perform at speed. However, if you just practise slowly using all kinds of inappropriate movements – movements you are getting away with simply because you are playing slowly- then incrementally speeding up will land you straight into a speed wall. Isn’t this obvious?

Accomplished pianists usually do not need to bother with any investigative work at speed because they have enough experience to know what movements will work, and they can go into slow motion practice straight away. A beginner however will not have a clue and will simply do slow practice and ingrain all sorts of inappropriate movements. This leads me nicely into:

I have though, seen a method of juggling, where they use scarfs to throw up in the air since it slows the speed of movement.  I also have seen people skip rope very slowly, and then speed up to very fast speeds.

I will start with juggling. Juggling with scarves is very difficult. The scarves have unpredictable trajectories. People not familiar with juggling believe that they must “catch” the balls. But actually you want to avoid this at all costs, because it will create all sorts of irregularities in your pattern and you will end up dropping the balls. You don’t want to “catch “ the balls. You want to throw the balls so accurately that they always land in your hand. A completely different concept and a completely different aim, leading to a completely different approach both in regards to practice as in regards to teaching/learning.

If your aim is to catch the balls you will never juggle fluently and elegantly because the movements required to catch a ball are not conducive to it. So your aim from the beginning should be not to catch the balls, but to throw them accurately. If you succeed you will of course always catch the balls.

I can just imagine some juggling pedagogue-to-be thinking intellectually about it and coming up with: “Wouldn’t it be nice if we could slow down juggling? But this is impossible! Wait a minute! What if I use silk scarves?” And being already an accomplished juggler, our juggling pedagogue, tries it out and finds to his satisfaction that it works. However it only works because he already knows how to juggle.  When the beginner tries it, because of the slowness of the fall of the scarves and because of their erratic trajectory, throwing a scarf accurately becomes impossible, and he therefore is forced instead to concentrate on catching it. So you can see how an eminently logical idea that seems to make sense can actually be an educational disaster.

If you want to learn how to juggle properly, you start without any balls at all. You just move your arms in the way a professional juggler does. You must master this movement and co-ordination (I am not going to bother describing it here, just watch a video of Anthony Gatto) without any balls at all. This movement can be done in slow motion (and must) since you have no balls yet. Now add one single ball, and never ever stretch to catch it. Instead let the ball fall to the floor Go from hand to hand like that, always aiming to throw the ball in your hand, and if you fail, let the ball fall on the floor. There is no slow motion possible here, because the ball falls under a speed determined by the laws of gravity.

Once you can consistently always throw the ball from hand to hand, you move to 2 balls. This is the really difficult step. But again the secret is to let the second ball fall to the floor, so that you get the basic movement right. Catching the ball is not important at all. Throwing it on an accurate trajectory is all that matters. If you succeed, the catching becomes assured. After that three balls is easy.

The parallels with piano playing/learning are so obvious that I will not expand on them.

But I will say that: get two groups of beginners. Teach one by the method outlined above. Teach the other using scarves. I have done this experiment many times with disbelieving but inexperienced juggling instructors. I always got my group juggling 3 balls perfectly after 5 – 15 minutes. The scarf group never got anywhere (we had one hour sessions) even after 2 or 3 weeks.

This leads us to skipping rope, which you may think has nothing to do with juggling. But the same principles are at work here.

Skipping rope slowlyish can be done? Yes, up to a point, but it is far more difficult and tiring than doing it fast. Like scarf-juggling it is the sort of thing a pro may do easily but that will wreck the chances of a beginner ever getting fluent. Like scarf-juggling, it is an advanced technique, not a teaching strategy. Of course this does not stop misguided (and usually well intentioned) instructors to insist on it. Heck don’t piano teachers insist on things that will never work as well?

Let us examine this in more detail. How slow can one skip rope? Well, I can do one skip every five minutes. I turn the rope over my head and skip. Then I wait five minutes, and repeat. Will that teach me how to skip proficiently? Of course not. Just like the point of juggling is not to catch the balls, but to throw them accurately, so the point of skipping rope is not to skip, but to co-ordinate my jumping up and down with the passing of the rope.

This again highlights the difference between doing something slowly and doing it in slow motion. If I do skip every 5 minutes, I am not really doing it in slow motion. I am doing it stop-start. The speed of the motion is the same. I cannot turn a rope over my head below the minimum speed determined by gravity. (I can increase it up to a point by adding my arm force to it). I cannot jump over the rope any slower than allowed by gravity. So what I am doing is perhaps jumping higher (to give more time for the rope to pass under my feet) and stopping after the rope has gone through. This may be slow practice but it is not slow motion practice.  

To understand what slow motion practice is you would have to watch a video of someone (who knows how to do it) skipping rope in slow motion. You would notice several things: the skipper actually jumps surprisingly low – sometimes less than an inch. His arms barely move: most of the movement comes from the wrists. His movements are smooth, cyclical and non-stop – one movement prepares and leads seamlessly into the other.

My slow practice on the other hand did everything wrong: The jumps are too high (because I am mistakenly assuming that the point is to skip over the rope), my arms and not my wrist are doing the work, and the whole movement is start-stop. Will I ever get fluent in skipping? Not if I insist on this procedure.

Obviously slow motion practice – as in the video in slow motion - is impossible. So what is the way forward? First get rid of the rope. Look at that video again, and do exactly the same movements: master the tiny jumps, and at the same time keep your arms in the same position ad turn your wrists. In other words: do exactly the same movements as the guy in the video, but without the rope. Once you become comfortable with it, get the rope, but hold it folded in one hand (grab both handlers in one hand). Now you are going to turn the rope around in one hand as you do the skipping movements (including the wrist movements). And the most important detail: you are going to listen very carefully to the noise the rope makes as it hits the floor. You must time your tiny jumps to the sound of the rope. Again, once you become proficient, try to do one single skip of the rope and stop.

This will now be completely different from the one skip I was doing before. Now I am doing the correct movement and I am being guided by the sound, my aim is not to skip the rope, but to co-ordinate my tiny jump with the sound of the rope. Of course, as a consequence I will skip. Then try three skips one stop and finally continuous skipping. Of course it may take time, but at least now your practice will give results, you will not just be wasting time in a conceptually misguided approach.

Unless you get the basic concepts correct (the aim of juggling is not to catch the ball, but to throw it accurately, the aim of skipping rope is not skip over the rope, but to co-ordinate the jumping with trope revolutions) you will not be able to devise a learning strategy to acquire proficiency, elegance and mastery in these activities. Your arguments about scarf-juggling and slow skipping although intellectually compelling show that you do not have much real experience with these activities (and if you have and if you are any good at them you are just repeating something you heard instead of paying attention to what you are actually doing).

I find these activities, and the learning strategies to master them, to have a direct import in piano teaching/learning/practising. The main difference is that contrary to skipping and juggling you can practise the piano in slow motion.

Best wishes,
Bernhard.

[namui]

Hi Bernhard,

  You know I always follow your advices (with much success). Reading posts in this subject help me summarize my related question to you. Here it is.

  As Chang mentioned in "teaching" section of his book, piano teaching is full of contradiction. I see the issue for this subject to be priorities of "speed" and "tone". Nick pointed out that the right movement for "speed" (by neglecting "tone" first) hindered the tone quality development. Nick hasn't mentioned that he already tried slowing such a movement down (the slow motion practice) to improve tone. But I did and, in my case, found that in a few places, additional movement is needed for improving the tone quality. For example, from fast speed movement, some key is not pressed deep as Nick mentioned. In order to press it deep during slow motion practice, a little additional arm and finger movement is needed. Some of those tone improving movements contradict with the speed movement. In my case, I moved back and forth between the fast and slow practices to investigate possible agreement.
 
  I still haven't found a good agreement for a place (Mozart's 12 variations in Var VI, fourth finger plays distinguishably bad tone at note A of bar 5, note G# of bar 6, and note E of bar 7). My fourth finger strength is not that bad at other places in this piece, so I assume that it's the movement problem.

  So my question is - Could it be possible that the right fast speed movement obtained by neglecting the tone quality introduces a tone "wall" (in response to the "speed wall"), making it difficult to improve tone quality during slow motion practice ? If it could, do you have any suggestion on how the contradictation should be managed ?

Regards,

namui

[Derek]

I started out fast, but very sloppily fast. Over time, my technique converged on greater and greater accuracy and evenness. I did a handful of exercises though and this no doubt helped to fine tune my technique. I'm no virtuoso, though. not yet.

[nick]

I was surprised to hear Bernhard say incremental speed increase to reach ones goal is ok as long as the movements when slow are correct. I didn't have the impresson the 'Chang' proponents thought like this, and I went back to Changs book and couldn't find it, only the slow motion practice, but not to incrementally get faster. My practice movements are very good with economy of motion primary. So we are in agreement then.
   Progress is good, up to 100 on the metronome with the sound I love, solid and strong with large dynamic range. Only way to go in my experience.
On the juggling with flags, I saw a juggler on tv teach some people in a matter of minutes. It is true I have no juggling experience, but thought that was interesting. Balls may be quicker to learn, I have no idea and leave that to the juggling experts.
Nick

[bernhard]

I was surprised to hear Bernhard say incremental speed increase to reach ones goal is ok as long as the movements when slow are correct. I didn't have the impresson the 'Chang' proponents thought like this

Er.. I would not call myself a Chang proponent, although I think his book is extremely helpful as a well organised collection of several ideas that are generally known but not in that concise and logical format. He himself is very clear about this, that he did not originate the ideas he talks about.

In fact when I read his book I was under the dellusion that I have invented most of it! It was quite sobering to learn that it was all common knowledge well before the time I was born!

The reason he does not mention incrementally speeding slow motion is simply because itis not really necessary. Just switch between slow motion and final speed. If the slow motion is indeed correct, its main use is accuracy. Once you get that, just play the final tempo straight away.

Another point is that although it is called "slow motion", it is actually quite fast - just not as fast as the final tempo. It is slowed down just enough in order for you to be able to work on the details with accuracy. How slow you will need to go to do that will depend on you, your technique, the piece and your understanding of the piece.

Best wishes,
Bernhard.

[bernhard]

Hi Bernhard,

  You know I always follow your advices (with much success). Reading posts in this subject help me summarize my related question to you. Here it is.

Er.... don't follow my advice! Try it out first to see if it works!

I have not forgotten your question. It is an excellent question. But the answer is complex. I am working on it. Expect a mammoth answer

Soon, hopefully.

Best wishes,
Bernhard.

[nick]

Well Bernhard, I guess we will just have to agree to disagree on this subject as my experience proves otherwise. The tone when playing fast h.s. right away is not good, and I have found the gradual increase( It goes almost without saying that the motions are the most economical) allows the technique, strength, coordination, to develop. I cannot see how one can gain this without the gradual development with very small increments.
Nick