Pauls Plan to Try It Himself. For Bernhard

[mound]

(this post began as a response in the How long to practice after.. (Bernhard, yours...) thread.

Here is what I came to realize (and this goes to the brutal discipline getting started with this method requires).

I had a section that I did learn after 7 tries. I kept practicing it.. At the 18 minute mark, I was thinking to myself "this isn't even close to speed".. It turns out I wasn't quite honest with myself.. I couldn't play the passage 7 times in a row without making a mistake.. Maybe once or twice, but there were flubs.   Yes I "learned it" -I knew the notes but I didn't pass the 7 repetition rule. I forced myself to cut the passage in half (even though I'd already worked on it almost 20 minutes.. I'm still getting used to the method here) and I was able to play it many more than 7 times at speed with no mistakes. I updated my log to show the modified session for reference tomorrow (more on this in a bit), and made a new session out of the other half.

I had another section that I had defined as a session, that was shorter. I was able to learn it after 7 tries. 6 minutes and 18 seconds of practicing this, I noticed I was at speed, playing w/o flubs.  Perhaps this session was a bit too short. I'm learning...

So a big part of this I'm finding, is with regard to the work we have to put individually, with our score, with our skills in mind and a pencil and paper in hand (or a computer as you'll see below). This is actually the only "hard" part of the method Bernhard suggests, and he can (and does) only point us in the right direction.  But the hard work upfront (that his students pay him to do for them) makes everything else SO INCREDIBLY EASY. I'm beginning to see why Bernhard has said "any lazy slob can learn 10 pieces in a month".   

It's taking the time and being brutally honest with ourselves in order that we can PROPERLY define the chunk size for each session and then PLAN how to proceed through them that takes all the upfront work.. Be organized and use a log. Even after 5 days now with a log and a timer and a lot of honesty with myself, I'm getting faster at saying "yeah, that'll be an appropriate size" chunk just by looking at the score.. the last 10 session sizes I've "guessed" by looking at the score were pretty much correct. Bernard says it over and over "try it and see". Of course I'll still be interested to hear his response on weather hitting speed during the session is important or not.

So anyway, why did I say "6 minutes and 18 seconds", what am I crazy or joking? No actually.  Check this out.

I practice on a digital piano in a home studio, with a laptop at my side for recording (I do other rock/electronica/fusion type stuff, but that's an entirely seperate world.)

Armed with all the information and thoughts I've taken from these threads and 5 days working at this method diligently, I went through my entire score and determined "right hand sessions", "left hand sessions" "hands together sessions". I then numbered them in the order that I should tackle them.  Generally it's a RH session, then a LH session, then HT session after the RH and LF sessions are mastered.  I won't know for probably another week the best way to order going from RH to LH to HT in so far as sessions in a day and sessions saved for the next day.

There are a total of 55 sessions I came up with that I have left to master before my piece is ready for the real work The piece is 140 measures long, and I've already put just over 3 weeks of what I'm now finding was inefficient work into it, so there were some phrases I already mastered. Nothing HT yet though. So I put them in an excel spread-sheet. 55 rows listing:
RH: 79-82
RH: 82-87
LH: 77-87

(etc.. and those above are just made up, that's the format)

I made sure there was overlap between sessions.  I then used my scanner to digitize the score, and used a screen-grabber to "cut out" chunks of the score in order to paste them into a corresponding column in the spread-sheet. (I haven't completley finished this, but it's based on a comment where Bernhard mentioned photocopying the score and using scissors to chop it up for the log, same idea only on the computer)

I put 31 days of columns out to the right (but that number could grow or shrink).. Each day column is two sub-columns "min. learned" and "min. practiced". There is also a "mastered column" that will receive a check when  I can play it perfectly on the first try of a next day, and thereby remove it from the daily rotation.

This spreadsheet sits open with a little MS Windows stop-watch utility next to it.   

I spent 2.5 hours two nights ago putting this spreadsheet together.. Well, only a few minutes putting the Excel format together, but 2.5 hours scrutinizing my score and coming up with the sessions that I thought would be appropriately sized, and then determining the order in which to do them based on my past 5 days of working through the method and getting an idea for myself of what the appropriate chunk sizes and order to work on them are. I also took into consideration the fact that my teacher requested a certain section of the piece be completed hands together next lesson, so I put the sessions that would allow for that to happen first.

This morning I got through 6 sessions (some took close to 20 minutes, some took 13-15 minutes.. one of them took 0, meaning I got it perfectly 7 times in a row on the first try) and two of them I realized should in fact be one, and one of them I realized had to be three. So I adjusted the spreadsheet. This is my learning process, the process we all have to go through that Bernhard can't define for us and days of reading and asking questions on this forum won't answer 

So.

Every time I sit down at the piano now (and for the foreseable future) I'll look at this spreadsheet. Whatever session is up next (that is, whatever is after the last one I worked on today because I ordered them as I saw fit) I'll start the stop watch and start the session. As soon as I've reached the "learned it again" point, I record the time in the "learned" column, reset the timer and then continue practicing it for no longer than 20 minutes total time. I stop before 20 minutes if I notice I'm playing it at speed more than 7 consecutive times w/o flubs and record whatever time that is.  In the instance above where I just wasn't getting it after 20 minutes, I re-evaluated my session size and created a new line in the spread-sheet cutting it in half.  I can use Excel to add "comments" (that popup only when rolled over) if I have any meaningful thoughts on what I did.

Tomorrow I will start all over at the beginning of the session list. If I get something right away, the "time to relearn" column will get zero minutes. (Bernhard, do you think if I get it in 0 minutes, I should practice it at all that day or is that a waste of time? I'm not sure yet) But if I do practice it more, I'll record that time. Henceforth, since that session is now mastered, it gets bumped off the list. (it's still visible on the sheet, but I change the color of the row which extends on out for 30 or whatever days) so I know I don't need to revisit it the next day or any other day.

So you could picture the grid, moving out from top left to bottom right, sessions getting bumped off the list (changing color), all the while showing me exactly where I should begin and what exactly I should work on anytime I get a chance to sit down at the piano.

So you can see, I have put in place a spreadsheet that pretty much enforces the process as I believe Bernhard has described it. It enforces the process, makes it beyond easy to know exactly what I need to work on every time I sit at the piano, it provides all of the log/journal capability AND, since I now know that I already have a clearly defined goal sitting there waiting for me ANYTIME, I'm finding I'm much more inclined to take advantage of that 15 or 20 minutes here or there that prior to putting this in place, I would have watched TV or whatever thinking "I can't practice in 15 minutes" (or not even thinking to go practice at all). This alone could add 3 or 4 sessions to a day easilly.

Why am I doing this? Because I'm insane.    Just kidding. It's because as I wrote up in another thread the other day, I tried it for myself for several days in a row, and results were astonishing. I've put a month into this piece already using my intuitive method, and I've mastered more measures in the past week than I did that entire month.  Once I've finished the piece (have it memorized, playable w/o the score) I'll upload a recording of myself as well as a copy of this spreadsheet for everybody to see and hopefully use as a model, and perhaps that will give back to Bernhard a bit, the man who knows everything and asks for nothing  (of course the performance will always grow as it matures don't expect too much right away performance wise!)

Of course over the next month or so this process will likely change a bit as I optimize it for myself, and the number of sessions (currently at 58) I'm guessing is going to shrink, because I think I've erred on the short side, but I'll learn if that's true or not.  I'm going to save a "snapshot" copy of the spreadsheet as it stands right now to compare how it changed by the time I've mastered the piece.  Also, to the thread the other day about "To Bernhard, Skipped Steps" - I think I know what to keep an eye out for, but if it becomes necessary, I can give each session's row in the spreadsheet a bunch of indented sub-rows containing a checklist of steps for me to repeat.

Lastely, Excel spreadsheets have "worksheets" - you get tabs along the bottom of a single file allowing you to have multiple spreadsheets in a single file. At the moment, I'm only working this single piece of music. Next week I'm going to add Bach's Sinfonia #9, and I will duplicate the format, add a new tab labeled "Sinfonia #9"  and so this Excel file will grow for every piece I'm working on.  Before starting work on the Sinfonia, of course I'll have to do all the preliminary work of deciding on sessions and what not.. Any time I have, I'll sit down at the piano, pick a piece (click a tab) and immediately know exactly what to work on.

The other night I had started doing all the mapping and session "defining" with a pencil and paper and even photocopied the score I was working on, but as I reached for the scissors and tape I thought to myself,  "Paul, what are you doing? You're a computer programmer, bust out MS Excel and make this even easier!"   

I only wish there were a way to do this (tracking all the numbers) against learning a piece "the old way" - but there's no way I'm going to waste my time practicing like that anymore, and realistically, I think the only way to track time practicing in an unplanned fashion would just to be keep a running timer every time you spent time at the piano on a given piece. So unfortunately, I don't think my spreadsheet will ultimately be useful for "proof" of the benefits in any statistical manner.. but that's really not the point anyway.


Bernhard, if you can see any glaring mistakes in what I've described, please do tell!

Oh, and lastly, I know this (and other recent threads) are all about "Oh Bernhard Bernhard Bernhard!" - but I want to point out that I have learned an immense amount from Chang's book (CC) about the actual techniques of practice. Of course I have learned many similar things from Bernhard, but Chang's book goes to excruciating detail in describing what you should actually be doing within the "sessions" that the methods Bernhard describes help you to plan. Bernhard and Chang share many similar thoughts on the physical practice techniques. If these two collaborated to add to the wonderful book Chang has published, oh man, look out!


-Paul

[mound]

oh one more thing.. The only confusion I still have with regards to this plan, I think I touched on it above, is how to lay out HT sessions.  Bernhard has said elsewhere how he tries to get HT as soon as possible. I had hoped to achieve this by doing something like:

SESSION 1: RH: 1-4
SESSION 2: LH: 1-4
SESSION 3: HT: 1-4

but if I had say 4 sessions available in a day, and I just started at the top, I'd do RH, then LH as two sessions. So far soo good. But what next? The HT session that joins the two new and unmastered HS sessions? It seems I'd have to skip over the HT session becuase it doesn't make sense that I should do a HT session before each seperate hand is mastered. Or should I? Preliminary work? With the understanding that it won't reach it's "zero minutes to mastery" until at least a day or so after each seperate hand is mastered..  The alternative would be for me to put the HT session much further down in the plan, so that chances would be good that both HS sessions were mastered by the time I got to it.. But that is delaying HT perhaps too long.  Or, perhaps I'm on the right track in so far as defining the sessions, but misguided in trying to apply any particular order to them? hopefully Bernhard can shed some insight on this.

-Paul

[bernhard]

oh one more thing.. The only confusion I still have with regards to this plan, I think I touched on it above, is how to lay out HT sessions.  Bernhard has said elsewhere how he tries to get HT as soon as possible. I had hoped to achieve this by doing something like:

SESSION 1: RH: 1-4
SESSION 2: LH: 1-4
SESSION 3: HT: 1-4

but if I had say 4 sessions available in a day, and I just started at the top, I'd do RH, then LH as two sessions. So far soo good. But what next? The HT session that joins the two new and unmastered HS sessions? It seems I'd have to skip over the HT session becuase it doesn't make sense that I should do a HT session before each seperate hand is mastered. Or should I? Preliminary work? With the understanding that it won't reach it's "zero minutes to mastery" until at least a day or so after each seperate hand is mastered..  The alternative would be for me to put the HT session much further down in the plan, so that chances would be good that both HS sessions were mastered by the time I got to it.. But that is delaying HT perhaps too long.  Or, perhaps I'm on the right track in so far as defining the sessions, but misguided in trying to apply any particular order to them? hopefully Bernhard can shed some insight on this.

-Paul

No you should not do this way. If you spend 20 minutes on one hand you are risking injury.

Instead, alternate hands. Initially you may need to repeat the RH 10 or 20 times, and then the LH 10 or 20 times (this should take no more than 2 or 3 minutes for each hand) but soon you will be able to do just a couple of repeats on each hand, and finally alternate one time RH one time LH. Try not to do more than 30 seconds on each hand.

Ideally you should choose passages that equally challenging for both hands, otherwise one hand is not going to be working enough while the other will be overworking.

Most classical and romantic music tends to work one hand far more then the other (usually the RH). So one thing that saves a lot of time is to actually work on two different pieces at the same session, but different hands. So – as I said  in another thread – if you are going to do Chopin op. 10 no. 2 which has a very challenging RH, but a very easy LH, you should really do the separate hands session coupling the RH of op. 10 no. 2 with the LH of Op 10 no. 12 (revolutionary) which has a challenging LH but a comparatively easier RH.

So now the scheme should look like this:

Session 1: Alternate RH and LH (they may come from different parts of the same piece or even from different pieces).

Session 2: HT

Or even better:

Session 1: Alternate RH and LH, then do HT (if doing the same piece).

Of course this assumes that you are tackling a completely new passage and learning it from scratch.

If on the other hand you have already mastered the whole RH and the whole LH of a Bach invention you should still alternate hands, but it may take you one or two minutes to complete each hand. That is not a problem because there is enough variety in the full piece to protect your hands from harm (very different form 5 or 6 notes repeated 200 000 times in 20 minutes.)

And yes, do not delay HT. Join hands as soon as you can (but not before! Exception: counterpoint music – there are other exceptions as well, but you will figure it out)

By the way, I am aware of your other posts, but they will take longer answers, so I am answering the easy ones first!

Best wishes,
Bernhard.

[Abstract Harmony]

I would like to see this spreadsheet...

[mosis]

I was wondering what you thought about the alternating hands method, Bernhard. Normally, I play one hand until it's mastered, then the other, and then either join or leave for the next day (after mastering HS again). I did not think that switching hands was actually efficient whilst trying to master the hands.

You answered this in another post, but I'm bringing it up here. How MUCH faster should I be playing a section? What if it's already a slow section that poses no difficulty, should it be played faster anyway?

I am sorry to be a bother, but please do try to get around to answering my question in this thread. It is very important.

http://pianoforum.net/smf/index.php/topic,4750.0.html

[mound]

No you should not do this way. If you spend 20 minutes on one hand you are risking injury.

Instead, alternate hands. Initially you may need to repeat the RH 10 or 20 times, and then the LH 10 or 20 times (this should take no more than 2 or 3 minutes for each hand) but soon you will be able to do just a couple of repeats on each hand, and finally alternate one time RH one time LH. Try not to do more than 30 seconds on each hand.

Arg. Now I'm confused again. I thought a given practice session was intended to work on one thing, using 20 minutes to get it to the point of 7 repetitions, and continuing practicing that one thing for the remainder of the 20 minutes.. One thing being a RH passage, a LH passage, or a HT passage.. If you try to do more than one of those things during that time, aren't you reducing the effectiveness or have I still missed the point?

Ideally you should choose passages that equally challenging for both hands, otherwise one hand is not going to be working enough while the other will be overworking.

Yeah, I think I let this idea slip my mind. Chang talks about it at length. Same with alternating hands.

The idea of working one hand from one piece and the other hand from another piece makes sense as well, I will keep that in mind once I'm working multiple pieces with these methods.
Right now I'm just working the Malaguena.

So now the scheme should look like this:

Session 1: Alternate RH and LH (they may come from different parts of the same piece or even from different pieces).

Session 2: HT

Or even better:

Session 1: Alternate RH and LH, then do HT (if doing the same piece).

Of course this assumes that you are tackling a completely new passage and learning it from scratch.

It's in this that my confusion must still lie. If I have a new passage I'm learning from scratch, and I'm going to do the "even better" session above where I alternate RH and LH and then do HT.. It would seem then, that I'm trying to master each hand seperately, as well as the hands together, simultaneously. This seems counterintuitive to me.. (though that might be the point) Would that not imply that I had to base my chunk size not on what one hand or the other could get in 7 repeats, rather, what HT could get in 7 repeats, thereby most likely making the chunk size even smaller, and in fact be the same as going directly to hands together?

If on the other hand you have already mastered the whole RH and the whole LH of a Bach invention you should still alternate hands, but it may take you one or two minutes to complete each hand. That is not a problem because there is enough variety in the full piece to protect your hands from harm (very different form 5 or 6 notes repeated 200 000 times in 20 minutes.)

Yeah, that makes sense.

And yes, do not delay HT. Join hands as soon as you can (but not before! Exception: counterpoint music – there are other exceptions as well, but you will figure it out)

See, "but not before"  - how can HT in the same session that you started HS possibly be the right time with that said?

Ok, lets take an example. The following is a 6 bar phrase taken from the Malaguena. Lets pretend for the sake argument that this is the piece in its entirety.



Lets assume that my RH can handle the RH part of bar 1 into the first beat of bar 2 based on the 7 rule.  My LH based on the 7 rule can handle bars 1 and 2 since they repeat.  How would you break this into sessions?

By the way, I am aware of your other posts, but they will take longer answers, so I am answering the easy ones first!

No problem my friend! I figured as much.

I would like to see this spreadsheet...

I'll share it once I can be sure it won't mislead anybody.


Thanks!
-Paul

[mound]

just an update.. I believe Bernhard is working out some responses to my recent enquiries,  but in the meantime, I've continued to experiment based on some of his comments. I'm finding so far:

1. It may be counterproductive (for me at least) to try to plan/spreadsheet the whole piece from the start. At least w/o lots of experience doing such a thing.. Logging (in the same format I described above) what I did "today" and then planning "tomorrow" based on today seems to be working better for me at least. I still know what to do next, each time I come to the piano, because I noted the next plan before leaving the piano "last time" but I'm not so concerned about trying to work sessions that may not have been defined appropriately.

2. Although I'm still a *bit* unsure about how to know when it's too early to bring HT, I'm now trying to do HS and HT work within the same session, it seems to be working OK even though I don't quite understand how according to these methods I can be putting HT before each HS session has had at least an overnight learning period. But none-the-less, it does seem to work, especially when I am conscious of alternating hands.. more on this to come I hope.

3. Repeated Note Groups.. HOLY COW.. I tried them on this passage of fast cascading downard pentuplets (that's what you call a 5 note "triplet" right?)



It took me 57 minutes to get through the whole set of groups. The next day, MAN! the runs were so easy! It's almost hypnotic running through the breakdown of an entire set, but you really nail every single transition!  I did find myself questioning "how many repeats do I do for each group?" I basically started each group very slowly to really feel the note, then if the chord attack worked, I'd start there (infinitely fast) and then slow it down (this goes to Chang's "parallel sets" and "phase degree") and I'd just kinda repeat slowing down, speeding up until it felt mastered.. No set time or number of repeats. Some groups were easy, especially where the chord attack worked, some were hard when it involved shifting my hand to the left..)

I also did an experiement with another repeated note group set on this:


That's image shows the first of 4 repetitions (it's cut off, but that dotted line at the top is an 8va) - it repeats dropping quickly through 4 octaves. I actually broke it into groups of 12 23 34 etc... all the way up to the 1234567. I asked myself "since it repeats, do I have to do each group all the way down the 4 octaves or is it enough to go through one octave?" since my arm position moves and therefore the hand position ultimately shifts a bit, I chose to do the whole 4 octaves - anyway, since LH and RH are the same, I did each group HS (slow, infinite fast and in between) and then HT at speed..  It took me just over 3 hours to get through the entire set of groups..  My hands were not sore at this point (probably because of the way I alternated them), but I don't believe a set of groups that takes that long is appropriate (somebody correct me if I'm wrong). I tell ya, the next day though, I was able to play the entire cascading run with HT close to speed with very few flubs, something I couldn't come close to doing before working the repeated note groups. The movements and maintaining that octave shape of my hand were just so easy, suddenly I didn't really have to "think" about each of the jumps..  Also, since each of these was an octave, there was no possibility of "chord attack" or "parallel sets" between each eigth note, at least not how I understand them.  Question: do repeated note group exercises ever need to be repeated or are they a one time benefit thing?



back to practice!

-Paul

[bernhard]

I was wondering what you thought about the alternating hands method, Bernhard. Normally, I play one hand until it's mastered, then the other, and then either join or leave for the next day (after mastering HS again). I did not think that switching hands was actually efficient whilst trying to master the hands.

Alternating hands is essential. But you have to use a bit of common sense here. You cannot start alternating hands straight away, you must first “learn” each hand.

So let us say you have a passage to learn. Start by deciding on the size you are going to tackle by using the 7 repeat criteria. By the time you pared the size to won you can tackle (HS) you probably already mastered enough of each hand that you can start alternating hands straight away, or at least doing 3 – 4 repeats on each hand.

What you do not want to do is to spend a whole session doing just one hand.

I think that perhaps part of the confusion here is due to a lack of discrimination between two necessary stages of practice:

1.   Stage 1 – Investigative. During this stage you are simply trying to figure out what to play: the right notes at the right time, the size of the passage you can tackle efficiently, the best movements for the sound you want to produce. If you make your passage short enough, this stage does not need to take more than a few minutes – sometimes a few seconds. The end aim of this stage is to get all the technical/musical parameters figured out, so that you can go on to stage 2.

2.   Stage 2 is real practice. A lot of students stop practising when they get it “right”. But actually real practice does not end when you get it right: it starts when you finally get it right. The aim now is to ingrain on your subconscious what you have consciously figured out on stage 1. Sometimes people ask: “Ok, the size of the practise I should learn is the passage I could learn after 7 repeats. But if I did 7 repeats and learned the passage, why practise it? After all I learned it.” This argument – although very logical – confuses stage 1 and stage 2. They are different stages with different aims. Both are necessary.


So, to go back to the question of alternating hands, in stage 1 you must concentrate on each hand in isolation. But this should take only a few minutes. On stage 2 – which is the bulk of the session – you must alternate hands (unless you re doing something time consuming as repeated note-groups).

You answered this in another post, but I'm bringing it up here. How MUCH faster should I be playing a section? What if it's already a slow section that poses no difficulty, should it be played faster anyway?

Always faster than required. Even if it is a slow section.

This was one of Arrau’s favourite practice tricks. He believed (and I agree) in having a “reserve”. If you can play faster than necessary, you will be able to play the correct speed effortlessly and with great ease. As I have said now several times, in my personal opinion the overall aim of practising is to make a piece “easy”. This is of course a basic principle in sports training. I once met a sprinter, who trained by running with a rope attached to his waist and in the other end an old tire. However, once he removed the tire, he could run like the wind.

Here is another idea for fast runs: get some surgical gloves, cut the tips of the fingers (so you have the sensitivity at the tips), and practise wearing these tight surgical gloves. When you take the gloves off you will be amazed at how your fingers can fly.

What if you cannot play faster that required by the composer? Then practise as fast as you can, but always perform slower then you practise. The point here is the point I keep coming back to: Your performance should always look easy and feel effortless. If you are always at the edge of your ability, you will not be able to play well: chances of botching will be great, and you will be so stressed out by it that you will not be able to concentrate on the actual musicality.

Very slow pieces (e.g. Arvo Part’s Fur Alina, or some of Beethoven’ slow movements) which have huge phrases that must be upheld throughout the piece present specific problems that have to be tackled with specific procedures. Still they will benefit from the technical ease that comes from playing them prestissimo (try the first Movement of the Moonlight – It is quite hilarious – and it will show you that this is exactly what the 3rd mv is – or try, as Jacquies Loussier does – to play the prelude 1 of the WTC prestissimo. It is quite exhilarating – and if you can do it, palying it slow will become that much easier)

I am sorry to be a bother, but please do try to get around to answering my question in this thread. It is very important. 

http://pianoforum.net/smf/index.php/topic,4750.0.html

Sorry it took so long – I have answered it now.

Best wishes,
Bernhard

[bernhard]

Arg. Now I'm confused again. I thought a given practice session was intended to work on one thing, using 20 minutes to get it to the point of 7 repetitions, and continuing practicing that one thing for the remainder of the 20 minutes.. One thing being a RH passage, a LH passage, or a HT passage.. If you try to do more than one of those things during that time, aren't you reducing the effectiveness or have I still missed the point?

Yes, this is more or less correct.

Before starting a practice session you must decide what you are going to do in that practice session. This is the first priority. But how do you do it?

Here is my suggestion (which I believe at the moment of writing to be the most efficient way to go about it) in regards to a new piece you have never seen before and which is still impossible for you:

1.   Spend as much time as needed listening to CDs of the piece and analysing the score. As your experience grows you will be able to know straightaway how you should go about learning it. But at the beginning this phase may take quite a long time.  In fact most people fall back into old, inefficient practice methods because they get discouraged at this stage. But I will assume all that has been done.

2.   So you start your practice session by sight-reading (HT) through, say, the first 12 bars of the piece. This may be a laborious affair. You may make lots of mistakes, and be hesitant and slow. But it will give you an idea what you are up against. It will tell you straightaway if you can skip HS and go to HT straightaway. It will tell you where the most difficult bars are. It may even be possible to learn the whole piece in this first sight reading. But I will assume the worst. The whole thing was truly impossible, and at the end of the 12 bars (which took you a full 4 minutes to complete) you are drenched in perspiration, feeling utterly discouraged and thinking to yourself: “this is impossible – I will never be able to master this piece”. At this point, write these feelings in your practice diary. You want – in two weeks time when the piece has been mastered – to go back and read these words again. And see how wrong you were.

3.   Anyway, what are you to do? You must now decide what part of these 12 bars you are going to practise. HT is clearly out of question in this case. So start with the first two bars. Try the right hand. Repeat it 7 times. If anything is even worse than when you first started. So you know that even 2 bars is beyond your ability at the moment. This should have taken some 30 seconds – 1 minute. So move to the LH. To your surprise, by the 3rd repeat you had it mastered  (= memorised and feeling really easy and effortless). So why not increase it? you go for 4 bars, and although more challenging after 7 repeats you did master it. This took 1 – 2minutes. Now go back to the RH. Cut it in half. Instead of 2 bars, do just one. And wonders of wonders, it is completely mastered after 7 repeats!
 
4.   Now you are going to practise 1 bar on the RH and 4 bars on the LH. You still have some 15 minutes left on your session. So at this point start alternating hands. By the end of 2 minutes, there is really no more work to do on the LH. The RH is still giving you some trouble, so you must use all sorts of trickery: repeated note-groups, rhythm variations, etc. So work for a while (2 – 3 minutes) just on the RH, and when it starts to feel tired, stop and do the LH a bit. You still have 10 minutes left on you session. So go on and tackle the next four bars of the LH (they happen to be an almost exact repeat) and the next bar of the RH. So now you have mastered 2 bars on the RH and 8 bars on the LH. You still have 7 minutes left.

5.   So join hands on the first 2 bars. You will not be able to simply do it. You may have to cut the 2 bars in half, or you may succeed with the dropping notes trick. You may even have to do repeated note-groups. If it takes you 15 minutes instead of 7 and your practise session ends up being 35 minutes instead of 20, that is Ok.

6.   Now you have mastered RH bars 1 – 2, LH bars 1 – 8 and HT bars 1 – 2. Write this down in your practise journal and write down what to do tomorrow: another 8 bars on the LH (this may not be possible though – you will find out). Another 2 bars of the RH and have bars 1 – 4 HT.

7.   Come the next day, and you simply play RH bars 1 – 2. Perfect? Move on. Not perfect? Try repeating it 6 – 7 times. Perfect? Move on. No signs of improvement? Forget about your plans for this session and repeat the session of the previous day exactly without skipping any step. Do the same for LH and HT. If LH is ok, repeat the previous session only for RH and HT, and do a different LH section. In any case, the previous day you mastered the section by following those steps, so today you will as well. Except that it will take a fraction of the time: 10 minutes instead of 35 minutes. So you still have 10 minutes to tackle the 2 new bars on the RH. Maybe you will not be able to join hands to day. But that is Ok.

8.   Next day repeat the same procedure and adjust your learning/practising accordingly.

9.   And so on and so forth.

10.   Now, let us say that you mastered bars 1-2 HT on the first session, and bars 3-4 HT on the second session (HS you may have mastered far more bars). On the third session, dedicate the full session to join bars 1-2 and 3-4 HT (if you have overlapped there should be no problems: the first 2 or 3 repeats it may fall apart, but after 7 it should be fine – if it is not you need to go back and do more work on the previous sessions). There is no need for HS work at this level. As the passage you have mastered increases in size, most of your work will be HT. HS is only for the very small passages you tackle at the very beginning.
There are two very important exceptions to this guideline.

a.   Counterpoint music. In this case you must learn the whole piece HS. In fact, you should learn each voice (part) separately – if there are more than two voices – as in Bach fughes. The reason for that is musical, not technical: you must have a very accurate reperesentation in your mind of how each voice goes.

b.   Long, fast, running passages, as in Schubert’s Impromptu op 90 no. 2. In such cases you need to work a speed HS at least 30% faster than the final speed hands separately, and you cannot really achieve speed efficiently working straight on hands together. You must increase speed HS and then HT will increase automatically.

The problem I face here, is that I can see too many instances where the paradigm above may not apply. So you have to adjust and adapt this general idea. Nothing is carved in stone. But I trust this will give you the gist of it. The more you apply these ideas, the less doubts you will have, and you will get to a point where by just looking at the soccer you will simply know how to proceed. The beginning is slow though.


 

See, "but not before"  - how can HT in the same session that you started HS possibly be the right time with that said?

As soon as you mastered HS, move on to HT. Do not wait until you have the whole piece HS to start joining hands. However do not join hands if you have not completely mastered HS.

Ok, lets take an example. The following is a 6 bar phrase taken from the Malaguena. Lets pretend for the sake argument that this is the piece in its entirety.

 

Lets assume that my RH can handle the RH part of bar 1 into the first beat of bar 2 based on the 7 rule.  My LH based on the 7 rule can handle bars 1 and 2 since they repeat.  How would you break this into sessions?

Ok.

Let us also assume that I am a complete beginner with a dismal technique.

1.   My first step would be to analyse the score and notice that there is quite a lot of repetition. So how much of these six bars do I actually have to practise? On the left hand there are two groups: the first three chords (that takes care of bars 1 – 2 – 5 – 6) and chords 7 – 8 – 9 – 10 –11 – 12 (bars 3 – 4). In fact, All I have to learn are bars 2- 3 – 4. Now there are 9 chords there. I have two choices here: I can see if I can learn the whole progression after 7 repeats. If I cannot I just try four chords, and if that does not work, two chords. But there is a much better approach: just go straight on to the repeated note-groups. 9 chords is just at the limit of feasibility. There is one more thing I will do: When doing groups of two chords (the most basic groups), instead of doing the full E major chord (chord no. 1) I will do just an octave and miss the two middle notes. This simplification will allow me to concentrate fully on the accuracy of the skips – which is really the major technical problem on the left hand. As soon as I get the skips accurate, I will put the missing notes back . I am still doing just groups of two chords. This is a very tiring routine, so instead of moving on to three chord groups, I will stop, rest the left hand and move on to the right hand.

2.   Again there is a reasonable amount of material repetition on the RH. The first three chords are repeated 5 times: the third repetition with a different rhythm. After that 3rd repetition there is a group of anthore three chords which are again repeated. At this point, the two main difficulties I see in the RH are accurately skip between chords, and simply move without hesitations form chord to chord. This is an exuberant piece, so there is no place for doubts and hesitations. So I will concentrate in this moving from chord to chord to star with. For that purpose I will initially ignore the rhythm and just consider equal times for every chord (say, consider them all crochets). I will only play six chords:  Chords no. 7 – 8 – 9 – 10 – 11 – 12. (second half of bar 2 + bar 3). Mastering this sequence will imply mastery of the whole line. Again I willuse the same approach I used for the LH: Repeated chord-groups, and for the 2 chord groups I will start with octave rather than full chords. Once I get the accuracy sorted out I will complete the chords. By the time I finish 2 chord –groups, my RH should start to feel tired, so I will move on the the LH and do 3 chord groups.

3.   I will continue alternating hands as above: for every chord group I finish I change hands. Once I get the whole sequence of 9 chords in the LH and 6 chords in the RH and can move easily, smoothly and accurately form chord to chord without hesitations and at speed, I will use a metronome and make sure the rhythm is ironclad (this will be relaxed later – since the notated, metronomic rhythm is never the true rhythm – but for the moment, the sections are too small and devoid of musical context to allow for a true rhythm, so a metronome is quite important at this stage – unless you have an unfailing sense of pulse that allow you to do without the metronome).

4.   Now that I have the RH and LH sequence of chords mastered, I will learn the whole line HS. It should be straightforward, since it is just a matter of memorising the sequence in which the chord sequences I already mastered occur. One trick I may use at this stage to avoid hesitations is to repeat each individual chord several times – while I repeat them, I have time to think where to go next. This way, there is never a break or any hesitation – although this is rarely necessary if one has done the repeated note-groups routine.

5.   I should by now have the RH and the LH at speed, memorised and moving forward without hesitations. This may have taken the whole 20 minute session – or perhaps a bit more. So I will have a 5 minute break.

6.   Now I will start a new session to join hands. In this particular case I will start the session by playing through both hands (separately) a couple of minutes (alternating them). If there is any problem, then I will not do HT. I will do something else in the session, and start form scratch next day. But if everything is still under my fingers (it should be), then I will just go straight for HT by doing dropping notes. First the RH in full dropping the LH notes, and then the other way round: the LH in full dropping the RH notes. This should take about 5 – 10 minutes.

7.   This passage is now large enough to allow for a musical rendering, so the reminder 10 minutes of the sessions HT, I will try to put the musicality into it. I will be working on the rhythmic accents, on the articulation, on trying to give a sense of exuberance to the sound. Ideally I should know in my mind exactly the sound I am after, and just keep repeating and comparing the sound I am producing with the one I want. Usually if the representation in one’s mind is clear, the fingers always comply.

Next day, I will sit at the piano and go straight for this section hands together. If it is perfect, that is it, all I have to do now is keep playing it and refining my musical concept of it – no further practice (as described above) is really necessary. On the other hand if I cannot remember a thing, then I will have to go back to the very beginning and start from scratch not skipping any steps – not matter how powerful the temptation to do so. In fact it is when you start skipping steps that you start to do wrong repeats and all sorts of mistakes creep in and become habits. So it is important to keep a journal of all the steps you took – at least in the beginning – with experience you will not need such detailed instructions. Rest assured that if you follow all the steps it will take a fraction of the time to achieve mastery again.

Most likely, the next day the passage will be somewhere between these two extremes (completely forgotten and totally mastered).If so, observe carefully where the problem lies. You may be able to tackle it in isolation and not have to go through all steps again. Ultimately the final arbiter is your progress. As long you are progressing you are doing the correct thing. If you stop progressing or if everything start to consistently fall apart, then you must change what you are doing.

I hope this helps.

Best wishes,
Bernhard.

[bernhard]

2. Although I'm still a *bit* unsure about how to know when it's too early to bring HT, I'm now trying to do HS and HT work within the same session, it seems to be working OK even though I don't quite understand how according to these methods I can be putting HT before each HS session has had at least an overnight learning period. But none-the-less, it does seem to work, especially when I am conscious of alternating hands.. more on this to come I hope.

Hands separate practice is important to isolate technical problems so that you can work on them without the added difficulty of co-ordinating both hands.

Hands together practice is important for dealing with co-ordination problems – you don’t want to do that if are still besieged by technical problems.

Hence, only do hands together once you have sorted out all technical problems with HS. Then you will be able to concentrate solely on co-ordination (not an easy task even when you have mastered HS).

At the same time, if you have no technical problems go straight to HT. There are many pieces that a beginner would have to do HT for wekks and a more advanced pianist can go straight to HT simply because he has mastered the necessary technique.

This means that HS is not obligatory, in fact do not waste time on HS unless you need to. However, if you need to never skip HS. Why?

Besides the advantage of separating technical and co-ordination problems, HS may be necessary for a completely different reason. It is with HT that you will develop “hand-memory”. And whatever goes in hand-memory stays. So you want to make sure that you never practice the wrong things with HT. Mistakes in HS are easily erased. So investigative technical work should always be done with HS. Only when you have all parameters sorted out should you proceed to HS.

However, as long as your HS is mastered, you do not need to wait until the next day to join hands. Quite the opposite, by joining hands you increase the chances that next day it will be perfect and you will not forget it, since it will now be engraved in hand memory. Does that make sense?

3. Repeated Note Groups.. HOLY COW.. I tried them on this passage of fast cascading downard pentuplets (that's what you call a 5 note "triplet" right?)

I call then “quintuplets”. But I could not care less about terminology. Call then whatever you want.

 

It took me 57 minutes to get through the whole set of groups. The next day, MAN! the runs were so easy! It's almost hypnotic running through the breakdown of an entire set, but you really nail every single transition!  I did find myself questioning "how many repeats do I do for each group?" I basically started each group very slowly to really feel the note, then if the chord attack worked, I'd start there (infinitely fast) and then slow it down (this goes to Chang's "parallel sets" and "phase degree") and I'd just kinda repeat slowing down, speeding up until it felt mastered.. No set time or number of repeats. Some groups were easy, especially where the chord attack worked, some were hard when it involved shifting my hand to the left..)

Yes, Repeated note-groups is the single most powerful routine I ever came across (I thought I had invented it, but no, someone else had already come up with it long before I touched a piano ).

As such, it overrides the 20 minute rule. If you are going to do it, you must finish it in on session no matter how long it takes. The results are so staggering that it is worth it.

In my experience the ideal number of units is 7. That takes between 30 – 45 minutes to complete. 5 units is really the smallest group one should do, and around 12 is the maximum (or you will end up doing it the whole night!). Sometimes however – depending on the passage there is no way round it: you must do it on large number of units. There are two ways to save time:

1.   Do not use single notes as units. You can use whole bars (if the bars are mastered). Or you can use beats (likein your example of the pentuplets above, you can consider each p-entuplet as a unit, so now you have to deal with 3 units, rather than 15.

2.   You don’t need to complete the sets. Just go to the point where there is some overlap between the first and last group. For instance, in the case of 15 units, when you get to 8 note groups, the first group: 1-2-3-4-5-6-7-8 overlaps with the 8th: 8-9-10-11-12-13-14-15. There is so much overlap in the intermediate groups that you may be able to play the whole passage straightaway. So when you reach the overlap size try to do the whole section. You may be able to do it, and in this way save a lot of time. If you cannot, do 9-note groups and try again. Usually you will not need to go all the way to 15 note groups.

I also did an experiement with another repeated note group set on this:
 

That's image shows the first of 4 repetitions (it's cut off, but that dotted line at the top is an 8va) - it repeats dropping quickly through 4 octaves. I actually broke it into groups of 12 23 34 etc... all the way up to the 1234567. I asked myself "since it repeats, do I have to do each group all the way down the 4 octaves or is it enough to go through one octave?" since my arm position moves and therefore the hand position ultimately shifts a bit, I chose to do the whole 4 octaves - anyway, since LH and RH are the same, I did each group HS (slow, infinite fast and in between) and then HT at speed..  It took me just over 3 hours to get through the entire set of groups..  My hands were not sore at this point (probably because of the way I alternated them), but I don't believe a set of groups that takes that long is appropriate (somebody correct me if I'm wrong). I tell ya, the next day though, I was able to play the entire cascading run with HT close to speed with very few flubs, something I couldn't come close to doing before working the repeated note groups. The movements and maintaining that octave shape of my hand were just so easy, suddenly I didn't really have to "think" about each of the jumps..  Also, since each of these was an octave, there was no possibility of "chord attack" or "parallel sets" between each eigth note, at least not how I understand them.  Question: do repeated note group exercises ever need to be repeated or are they a one time benefit thing?

Yes, you may not need to do this for repeated passages even if they are an octave apart. (But no harm will come from doing it).

Also these days I reserve repeated note-groups only from truly impossible passages. On the other hand, if takes 3 hours, so be it. the saving in time compared to any other method is so ludicrous that it is worth every second.

And yes, usually it is a one-off thing. Some passages may be so difficult (for you) that you may need to repeat it the next day and even the next – but I myself never met any passage that needed more than 3 repeats to be mastered. And 90% were mastered like forever in the first go. The whole idea is not to do more work than strictly necessary. Pianists can be very insecure (just watch Mosis! ). Even if you neglect a piece for a few years, if you learned it this way, chances are that you will not forget it and if you do, going through the same process (in a fraction of the time) will bring it back.

And yes, there is no fixed number of repeats for each group: you must repeat until you master (= becomes easy). And that is the beauty of it: you will be practising more the bits that actually need more practice, and practising less the easy bits – exactly as it should be.

Also, you just had the answer to why you should alternate hands: you can practise for 3 hours without tiredness or risk of injury.

I do not think I have anything else to teach you.

Wonderful!

Best wishes,
Bernhard.

[mound]

However, as long as your HS is mastered, you do not need to wait until the next day to join hands. Quite the opposite, by joining hands you increase the chances that next day it will be perfect and you will not forget it, since it will now be engraved in hand memory. Does that make sense?

YES! Finally it does.  If I get the HT work from the end of yesterday correct first thing today, then I never have to repeat the HS sessions that it took to get there, even if that means they were only executed that first day, because I got HT right the 2nd day. THAT IS THE POINT! The point which has alluded me until now. I've kept thinking that something about the "psychology" of this whole approach requires me to learn and relearn the HS sessions before ever attempting to put them together, so that HS would be long term mastered and then HT would see a similar process. As if HS and HT are two completely unrelated entities.  So the test on day 2 is not weather each HS session is mastered, rather, if HT from yesterday is mastered. It is only if HT fails on day 2 that I need to start from scratch with HS, otherwise, there is no need to do any further HS work, because HT already works, so right to being musical with it.  Why did it take so long for this to click for me?? Hopefully this dialog helps it click for others in a fraction of the time.

Most likely, the next day the passage will be somewhere between these two extremes (completely forgotten and totally mastered).If so, observe carefully where the problem lies. You may be able to tackle it in isolation and not have to go through all steps again. Ultimately the final arbiter is your progress. As long you are progressing you are doing the correct thing. If you stop progressing or if everything start to consistently fall apart, then you must change what you are doing.

Gotcha!

Yes, Repeated note-groups is the single most powerful routine I ever came across (I thought I had invented it, but no, someone else had already come up with it long before I touched a piano ).

As such, it overrides the 20 minute rule. If you are going to do it, you must finish it in on session no matter how long it takes. The results are so staggering that it is worth it.

I'm going to officially give you full credit Bernhard So I am correct in my experiment - I can apply the repeated note group technique to any passage I feel the need for, as long as I have the time to complete it in one sitting. From other descriptions of the process elsewhere in these forums, I had the idea it was reserved to single note arpeggio type runs. This technique is staggering indeed! Even moreso than all the other topics of this method. You can spend 3-4 days trying the organization and session planning and 7/20 rule, comparing to "the old way" etc. and ultimately see the benefits in the end. But 50 minutes on a well defined repeated note group exercise and tomorrow you're like "HOLY CRAP!" For anybody reading who hasn't tried it, find yourself a tough passage and a free hour and go at it, it'll blow you away.

I do not think I have anything else to teach you.

You sir, have completely, and utterly blown my mind  I thank you!

Oh, just an aside, perhaps food for thought.
1. I'm sure everything you described above exists elsewhere in these forums.
2. I've been pretty exhaustive in trying to seek it all out and "order it" in my mind from this web/circular format.
3. Yet the very core of the method alluded me until 2 weeks of experimenting on my own, and several days of back and forth dialog.

Why? Perhaps some incentive for you to begin on a book?

oh and ps. I'm not sure there is need for you to worry about that Sinfonia PM I sent you last week unless you feel there is more to add specifically because of the style.


I'll keep this thread updated with my thoughts/progress.. My spreadsheet of course will act as a daily logger and "tomorrow planner" just as described originally.

Phew. I feel much better now   

-Paul

[mosis]

Mound,

3. Repeated Note Groups.. HOLY COW.. I tried them on this passage of fast cascading downard pentuplets (that's what you call a 5 note "triplet" right?)

It took me 57 minutes to get through the whole set of groups. The next day, MAN! the runs were so easy! It's almost hypnotic running through the breakdown of an entire set, but you really nail every single transition!  I did find myself questioning "how many repeats do I do for each group?" I basically started each group very slowly to really feel the note, then if the chord attack worked, I'd start there (infinitely fast) and then slow it down (this goes to Chang's "parallel sets" and "phase degree") and I'd just kinda repeat slowing down, speeding up until it felt mastered.. No set time or number of repeats. Some groups were easy, especially where the chord attack worked, some were hard when it involved shifting my hand to the left..)

Can you please detail how you went about using repeated note groups to conquer this passage? I am facing a passage of similar difficulty, and I am not sure if I am going about it the right way.

[mound]

the cascading quintuplets? I did it exactly as Bernhard has described numerous times elsewhere.

if I number the notes 1 2 3 4 5 6 7 8 9  10 11 12 13 14 15

I did these groups:

1 2
2 3
3 4
4 5
5 6
6 7
7 8
8 9
9 10
10 11
11 12
13 14
14 15

1 2 3
2 3 4
4 5 6
6 7 8
7 8 9
8 9 10
9 10 11
10 11 12
11 12 13
12 13 14
13 14 15

1 2 3 4
2 3 4 5
3 4 5 6
4 5 6 7
5 6 7 8
6 7 8 9
7 8 9 10
8 9 10 11
9 10 11 12
10 11 12 13
11 12 13 14
12 13 14 15

1 2 3 4 5
2 3 4 5 6
3 4 5 6 7
4 5 6 7 8
5 6 7 8 9
6 7 8 9 10
7 8 9 10 11
8 9 10 11 12
9 10 11 12 13
10 11 12 13 14
11 12 13 14 15

1 2 3 4 5 6
2 3 4 5 6 7
3 4 5 6 7 8
4 5 6 7 8 9
5 6 7 8 9 10
6 7 8 9 10 11
7 8 9 10 11 12
8 9 10 11 12 13
9 10 11 12 13 14
10 11 12 13 14 15


see how that breaks down? Then, for each group (that is, each line above) you use the various techniques to play through them, repeating the number of times necessary to master it.  Look at Reply #3 in this thread for details on how to practice each group:
http://pianoforum.net/smf/index.php/topic,2449.msg21217/topicseen.html#msg21217

So you can see how it could take so long to get through the entire "set" of groups, so make sure you have time before you start, because you have to make it through the entire set in one sitting.. (be sure to alternate hands and use any possible time savers as described by Bernhard that are suitable for your particular passage)

(oh, and that set isn't something that you have to "write out" or anything, you basically are starting with a "group" of 2 notes, mastering that group, and then starting on the 2nd note and doing the next group of 2 notes (thereby having overlap) and repeating that until you've made it through the whole passage. Then you start with 3 note groups, and do the same thing.. See how you end mastering every single note and every single transition between every single note?

Who woulda thought!



cya!
-Paul

[mosis]

Hmm, I was thinking that you would play one parallel set, then the second, then the second and third, then the first second and third.

The passage I am referring to is the run on the last page of Chopin's 9/1 Nocturne (the dodecatuplet). There are 5 very logical parallel sets that may be played. Should I just practice jumping between these two, are work on every single not seperately? I'm not sure that will be necessary.

Essentially, I am wondering when the chord attack should be used, and then repeated note groups should be used. :/

Oh, and Paul, did you end up playing notes 1-15 at the end of that session? Isn't that the ultimate goal?

[mound]

Hmm, I was thinking that you would play one parallel set, then the second, then the second and third, then the first second and third.

Judging  from what Bernard said above, you could if each parallel set was already mastered.

Oh, and Paul, did you end up playing notes 1-15 at the end of that session? Isn't that the ultimate goal?

yup!


-paul

[steven]

paul,
it's this the same to learn a piece HT ?
Let say our piece is 20 mesures.
You do as above like 1-2; 2-3; 3-4,...
In fact the same as above but not with notes but with mesures ?

thx
Steven