Idea for Yamaha to have automatically more sustain on piano sounds

[bpsafran]

Hi Everyone:

Many complaints about the piano sound even on top of the line keyboards arise because arranger players are not pianists and do not use the sustain pedal properly.  The trick in sustaining the piano is to remove the sustain (move your foot off the pedal) in between chords but quickly put it back for the next chord.  This takes some practice.  I was thinking that when an arranger player plays chords (and/or styles) in the left hand and piano in the right, that the keyboard be programmed to automatically add sustain and remove it quickly and then add it again it when a chord change is detected.  This is not going to reproduce the pedaling of a piano virtuoso, but may help out the typical arranger player with their piano playing.  I think the programming should not be difficult and there can be a button to enable/disable this effect, since for fast songs one usually does not use the pedal.  Is there anyone here who can pass on this suggestion?

Regards

[mikf]

I don't think this would be practical or an improvement for a number of reasons. Pianists use the pedal in different ways depending on their individual style, and the music being played.
Mike

[bpsafran]

I agree that this suggestion is most applicable to slower music.  But it should be applied by a button push and one can always not push the button to use the feature.

[Toril S]

It could be a parameter in the panel sustain button.

[Keyboard Master]

To Me That would be a Wonderful Idea. If Yamaha adds such a feature. That would make a whole lot of use. Espacially if you play while sitting on the bed or floor and/or didnt have enough room for a sustain pedal. That's an idea to submit to Yamaha too for futire products. I totally Agree with this idea. Thabk you for sharing. 

[bpsafran]

Eileen: Can you or your contacts bring this idea to the attention of Yamaha?

[mikf]

Why don't you use the new site referenced in the forum specifically for this purpose.
However, I have to say, I cant see this ever being taken too seriously. I cant see it being worth the effort when you already can do it properly with a pedal. 
Mike

[jwyvern]

If you are a non pianist finding it difficult to handle the sustain pedal the keyboards already provide an aid not too different from what is being suggested here. Just turn on the Panel Sustain. The voice will then sustain beyond the release of the keys, the length of which can be controlled to match the tempo suitably in the voice editor settings. No pedalling required so I suggest that is checked out first  .

John

[SciNote]

I generally like to take care of functions like this myself, using the pedal in this case.  However, I can see the usefulness of such a feature.  For example, on the new PSR-E473, there are S.Art Lite voices that can be manipulated with the pedal.  A feature such as this "smart sustain" function would allow there to be a much more musically proper sustain function handled by the keyboard, while the player can use the pedal to work the S.Art Lite function.

This would be quite a bit different than just a simple panel sustain function, because the keyboard would need to recognize when you change chords with your left hand, and then cut out those previous notes, so that the previous notes don't ring on and clash with the new chord.  Imagine a song with chromatic chords, and let's say one measure is C major, then the next major is Db major.  With a simple panel sustain function, the notes of the C major chord would continue to ring on while you then play a Db major, which would produce major dissonance.

Even lower priced keyboards have the ability to recognize what chords are being played with the left hand, so this is a feature that should be easy to implement.  Maybe it could even be done with a software update on the higher-end keyboards.  To be realistic, the notes that are being cut out when changing the chord would have to have a quick -- but noticeable -- decay time, and not just cut out instantly, because it would normally take a piano a brief time to damp and silence the sustaining notes when the sustain pedal is released.   Maybe 0.1-0.2 seconds for the notes of the old chord to fade out when the new chord is played.

[Joe H]

The standard MIDI sustain message is just On or Off.  Higher end keyboards may support Half-damper pedal sustain which allows continuous variation of the sustain. If it's not programmed into the OS then a half-damper pedal won't work.  The Panel Sustain button I believe adjusts the release envelope which allows the Voice to sustain more... or less.

Joe H

[SciNote]

The standard MIDI sustain message is just On or Off.  Higher end keyboards may support Half-pedal sustain which allows continuous variation of the sustain. If it's not programmed into the OS then a half-pedal won't work.  The Panel Sustain button I believe adjusts the release envelope which allows the Voice to sustain more... or less.

Joe H

I don't see a reference to half-pedal sustain in either my post or the previous posts.  What were you referring to?  I did mention how such a "smart sustain" feature being proposed here would be best if it did not abruptly cut out the notes when going from one chord to another, but instead allowed them to fade out in about 0.1-0.2 seconds, but there, I am only talking about how I feel the envelope of the sustain (technically, it's called "release" with a synthesizer envelope, because it involves what happens to the sound when the keys are released) should be programmed should such a feature be implemented.  But this would just be a programmed envelope for this proposed feature, and would have no bearing on how an actual pedal would be used with the keyboard -- in fact, as I understand it, the purpose of this proposed "smart sustain" would be to eliminate having to use a pedal at all to get a more harmonious sustain effect without having sustaining notes from one chord ringing on and clashing with the notes of a new chord being played.

And I did edit my post to make it clear that I was talking about the "panel sustain" that is available on many existing keyboards when I said that it is not the same as what is being proposed here.  The existing panel sustain simply adds sustain (or release) to the notes being played, allowing them to ring on for a certain amount of time after the keys are released, and these ringing-on notes could then clash with new chords that are then being played, such as when playing a C chord, then playing a Db chord.  In this scenario, the "smart sustain" being proposed by the original poster, as I understand it, would only allow the notes of the C chord to ring on until the next chord, in this case the Db chord, is played.  At that point, the keyboard would cut out the notes of the C chord so that they do not clash with the Db chord.

[mikf]

The OP proposed this in a fairly simple way only for piano voice. He suggests it would improve the piano voice which many complain about.
But the main reason most arranger players think the piano voice is too thin is not due to lack of sustain, but because they don't play like a  piano player would. They play singly note melody lines, when experienced piano players play multi note rh lines. And piano players know how to play those notes so they are not exactly simultaneous, giving a much richer tone. It is this combined with appropriate use of the sustain pedal that makes a really full piano sound. So it's about the whole playing technique not just sustain.
There may be other reasons why an automatic sustain function might have merit, some of which are mentioned above, I don't really know. But I doubt if it would cure the the thin piano sound some complain about.
Mike

[Joe H]

There is no such thing as smart MIDI.  MIDI is very limited because it is a 7-bit binary code written in hexadecimal numbers. 

Joe H

[SciNote]

Mikf -- That's a good point about playing multiple notes with the right hand.  I guess this could be simulated somewhat with autoharmony functions, but like you said, a real piano player may not always play those notes at the exact same moment.  Still, for someone who mainly just plays left hand chords (or easy-play/ "single-finger" chords) on the left hand and a single melody note on the right hand, the existence of something like this proposed "smart sustain" (my made-up name) feature combined with auto-harmonize could make the resultant piano output sound much fuller and closer to professional.  And I don't see why you'd even have to limit it to a piano sound -- it could be used as an optional feature on a variety of sounds.  I bet it would sound great with strings.

Joe H -- Like before, who is saying anything about MIDI?  As I understand the original post, all that is being proposed is a feature that, while playing chords using a piano voice, the notes of these chords would ring on after letting go of the keys -- similar to as if holding down a sustain pedal or using a panel sustain button.  But unlike just continually pressing down a sustain pedal or using a simple panel sustain button function, this proposed "smart sustain" would recognize when the keyboard player is playing a new chord, and then would automatically cut out the notes that were ringing on from the previous chord so that the two sets of notes don't clash.  This would simulate the action of a more experienced piano player, who, when wanting to have notes sustain or ring-on after letting go of the keys but not want them to continue when playing a new chord, would hold down the sustain pedal with his foot, but then quickly release the pedal and then press it again when changing the chord he is playing on the keyboard, thus silencing the notes ringing on from the original chord and allowing the notes of the new chord to sound and then ring on after the keys are released.  The only other thing I proposed above is that, when the keyboard would silence the notes of the original chord, don't have the sound cut out abruptly, but instead let the sound "roll off" in about 0.1-0.2 seconds for a smoother, more natural transition between chords.

None of this would require MIDI -- just simply internal programming of the keyboard to recognize chord changes (something most arranger keyboards are already capable of) and to stop any sustained notes (notes that are ringing on even though their corresponding keys are not currently being pressed) from sounding when the keyboard sees that the player is now playing a new chord.

[Joe H]

Bob,

I understand your argument but it would do you some good to learn about MIDI.  People who use smart phones may think this is possible, but it's not.

Joe H

[mikf]

Mikf -- That's a good point about playing multiple notes with the right hand.  I guess this could be simulated somewhat with autoharmony functions, but like you said, a real piano player may not always play those notes at the exact same moment.

Bob - it's not just about not playing the multiple notes together, it's also which additional notes. This is why auto harmony is not something that works well. It repeats full block chords on every note and that only works in special cases.
Mike

[metsasarv]

In 1989, I had my first arranger-type keyboard, the Fujitone3A. It didn't have a pedal socket, but it had two effect buttons - vibrato and sustain. I made as many music as I could. However, the worst keyboard anyone could want :-)

  As far as I know, such a sustain button should be found on the main panel of many PSR models.

[Joe H]

Just to clarify.  Our arranger keyboards are MIDI machines.  Its architecture is made up of several specialized sequencers.  The Operating System does not control how MIDI works and has nothing to do with how sustain works; which is a MIDI controller message. The sustain pedal on a real piano is variable.  To get a similar feel on a MIDI keyboard the instrument has to support half damper pedal sustain and uses a different pedal than the typical on/off switch sustain pedal. The half damper pedal provides variable sustain like a real piano when supported by the keyboard.

Joe H

[SciNote]

Joe, I don't know if you're just skimming these posts or what.  I don't know why you keep bringing up a variable sustain pedal and MIDI.  This proposed idea, which is not my idea anyway but sounds like a good idea, would just simply be a variation on the panel sustain button/feature that is already found on many keyboards.  I'm not going to go over all of the details again, but in a nutshell, instead of just sustaining all notes that are played, the keyboard would recognize when the player changes the chord he is playing (something that arrangers can already do), and then silence the existing sustaining notes so that the notes of the new chord do not clash with notes sustaining from the previous chord.  It has nothing to do with using a pedal -- variable or otherwise -- and in fact would be designed to provide much of the sound of using a pedal but without actually using one.  Whether such a function could or would be able to be implemented through MIDI using another keyboard controller, computer, or DAW is a completely separate discussion.

Am I missing something here?  If anyone else knowledgeable about the electronic operation of keyboards and MIDI can see the point that Joe is making and that I am missing, please clarify that for me.

[Joe H]

Bob,

With all due respect, to my knowledge what the OP wants cannot be done.  He simply needs to learn how to use the sustain pedal.  Maybe Yamaha could assign the pedal to control the Panel sustain button.  But I doubt that Yamaha could create a "smart sustain". MIDI is not capable of that.  As I already stated. Arrangers are MIDI keyboards. BTW... support for a half damper sustain pedal would be a real benefit for playing a piano Voice on the arranger. They work just like the sustain pedal on a real piano. The OP apparently plays piano. And I agree with Mike the Moderator, he is a piano player.  It's about technique not AI technology.

Joe H

[mikf]

Bob
I am certainly no expert on the functional details of the arrangers, but I think that Joe is saying that midi is not just a transfer facility for these keyboards. But that the internal functions of the keyboard are largely being driven through midi protocols. So what can and cannot be done might be limited by what can be done by midi commands.
Mike

[SciNote]

Yeah, as I thought about it more, that seemed logical -- to a point.  I can understand that, if you're playing MIDI tracks, once a keyboard is told to play a note, then it's envelope (and other characteristics) are already determined, so that you cannot just go back and tell that note to stop sounding.

But for live playing, I see no reason why a keyboard cannot be designed to monitor the notes that it is sounding through a sustain/release envelope and then be programmed to silence certain ones under certain conditions, such as when the keyboard recognizes a chord change.  These keyboards are controlled by computer chips that process, what, a billion instructions a second?  So tracking a couple dozen notes and silencing them should not be an issue.

[SciNote]

I've thought about this some more, and I think we're overthinking this.  You know, whether you're playing a PSR-E series, or a Genos, or anything in between, when you are playing with the sustain pedal pressed down, all the notes being played (at least on the part of the keyboard affected by the pedal) will ring on when you let go of the keys.  And then, when you let go of the pedal and press it again, whatever notes that were ringing on will now stop sounding, and since you've pressed down on the pedal again after letting off of it, any new notes that you now play will ring on after you let go of the keys.

So, I don't see any reason why any keyboard with a sustain pedal jack and chord recognition could not be designed to have a switchable feature that internally activates the sustain pedal function (as if a pedal was hooked to the jack and being pressed down), and then when recognizing a chord change, to then electronically perform the function as if the pedal was released and then pressed down again.  Again, all arrangers have programming to recognize what chord is being played with the left hand.  I would say that the keyboard could see when the player completely lifts his hand off of the left side of the keyboard, whether auto accompaniment is on or whether the player is just using a split keyboard.  That would be the signal to the keyboard to check out the next notes that are played with the left hand and see what chord is being played -- if it is the same chord that was being played before the player lifted his hand off the keyboard, then the keyboard acts as if a pedal is hooked up and still pressed down to keep the existing notes ringing on.  If, however, the player plays a different chord after lifting his left hand off the keyboard, then the keyboard acts as if a pedal is hooked up and that the player just lifted his foot off of the pedal and then pressed it down again, which would then silence the existing sustaining notes, but then allow any new notes played to ring/sustain on after letting go of the keys.

In other words, the keyboard would just pretend that a pedal is hooked up to the keyboard, and then whenever it detects a chord change with the left hand, it would automatically/electronically activate the same procedure it would do as if the pedal was released and pressed down again.  Since the sustain pedal and its jack is just a simple 2-wire switch connection, at the very least, the keyboard would just have to send a signal to a relay or a transistor that electrically connects and disconnects these two wires when it detects a chord change as I described above to simulate a pedal being pressed and released.

[tyrosman]

TO Add more sustain all you have to do is to Adjust the release with in the voice edit

[Joe H]

I don't think this would be practical or an improvement for a number of reasons. Pianists use the pedal in different ways depending on their individual style, and the music being played.
Mike

I agree, using the sustain pedal whilst playing piano is part of the art and technique of piano playing.  It has to be learned just as we learn to play chords and melody on the keys.

Joe H

[SciNote]

TO Add more sustain all you have to do is to Adjust the release with in the voice edit

Of course you can add more sustain/release that way.  I can even add release to the voices in a registration on my PSR-E433.  But what we're talking about here is more complex than that.  If you just simply add release to the tones, then those tones will continue to ring on as you play different chords, so that the notes that you play with a new chord could seriously clash with the notes that were ringing on from a previous chord.  Imagine playing a C chord, and then while those notes are ringing on after you let go of the keys, you then play a Db chord -- with the notes of the original C chord still sustaining or ringing on!  Unless you're going for a C11b9b13 chord, it's probably not the effect you want!

Now, an experienced piano/keyboard player will use the sustain pedal to avoid these clashes by using the pedal to sustain the notes until he changes chord, at which point he can let go of the sustain pedal briefly to silence the notes of the first chord, and then hit the pedal again before playing the notes of the second chord so that they will sustain without clashing with the notes of the first chord.

So the original idea posted here was, could Yamaha (or any manufacturer, for that matter) design a sustain function, for less experienced players or for situations where using a pedal would not be practical, with a sort of artificial intelligence to automatically silence the existing sustaining notes, but then still sustain new notes played, when it detects that the new notes represent a chord change from the previous notes that were played and sustaining?  And in my opinion, the answer is yes.

All arrangers have programming that determines what chord is being played with your left hand -- that's how the fingered-chord auto-accompaniment works.  There is no reason why software in an arranger cannot monitor and track what chord is being played with each new note played with the left hand, and then when it sees that the current chord is =/= (unequal) to the previous chord played, to set a procedure in motion that would simply and electronically do the same function as releasing the pedal and then pressing it again.  Most sustain pedals are just simple 2-contact SPST switches -- at the very least, all the programming would have to do is trip a relay or transistor at these contacts (at the sustain pedal jack in the keyboard) to simulate releasing and pressing the pedal.  In reality, the programming would likely just activate the procedures in the arranger's programming that accomplishes this function, but either way, I see no reason why it would not be possible.

[EileenL]

I think it is more a case of learning how to use a sustain pedal rather than expecting a keyboard to do it for you. We must expect to do some things ourselves as that is part of the fun learning new things.

[Lionel N]

Hello,

the idea is interesting, not necessarily for players who are unfamiliar with the sustain pedal, but also for those who are.

I work on a mod in order to provide stereo to the unfortunately mono PSS-A50. Stereo brings pleasure when playing.
Sustain also brings pleasure when playing and enhances the rendering.

The (entry level) PSS-A50 does not provide any sustain pedal connectivity. Whereas the (mid-level) Reface series provide it.
These keyboards are intended to be used at any place ("or even sitting in the passenger seat on a long road trip").
Therefore, you won't connect a pedal, but it would be fun to have a viable sustain feature.

The PSS-A50 has a "Sustain" button on the front panel, but using it usually results in a mess because everything is melted together. The chords that match, but also the one that don't match.

If I had to work on an "enhanced sustain" mod for this keyboard, I would probably take advantage of the ideas provided in these discussions.
The basics would be of course to detect the chords (via the midi connectivity). And then, enable/disable the reverb (also via midi). This is not a real sustain pedal bahaviour, but it could help.
But when you start to implement something, a lot of new questions or orientations are coming. For example:
The automatic chords detection used for the bassline rhythm section should also be implemented on the right hand. Maybe some difficulties to identify what is part from the left hand and from the right hand...

Like (usually) every consumer projects, innovations are not implemented on the low (entry) cost products, but at first on the high-end products, and then, once the development cost is amortized, it can be provided on the lower-end products.

So we arrive again at the same question: why implement a function that has (most probably) no use for experienced players that will buy the high-end products?
The answer is maybe part of the products like the Reface series.

Lionel.

[SciNote]

You bring up some interesting points and ideas.  Of course, this discussion started as a suggestion to Yamaha for a feature to include in future keyboards, but the idea of adding this to an existing keyboard is interesting.  You mention using MIDI and the reverb function with the PSS-A50, but I don't think reverb would come close enough to a sustain sound.  Couldn't the PSS-A50 accept some sort of MIDI command to activate sustain on notes?  Otherwise, since it seems that you're already opening up the keyboard and changing wiring/components to add stereo sound, I wonder if you could just get to the contacts of the sustain button directly.  Imagine having a "magic box" that hooks up to the MIDI port of the PSS-A50 -- It seems that is what you suggested as far as having something recognize the chords that you're playing.  So, this "magic box" would read the notes being played on the keyboard and then use a program to determine what chord is being played.  Whenever it detects a change in the chord, maybe it could send a signal to a transistor or relay to manipulate the contacts of the sustain button on the PSS-A50, so that it would normally engage sustain, but then momentarily disengage it, and then re-engage it, when it detects a chord change to silence the sustaining notes of the previous chord.

Of course, there would likely be more to it than that, such as accounting for melody notes that are not a part of the existing chord being played, but are just passing notes that are not intended to change the entire chord of the background.  And obviously, trying to implement something like this would involve opening up the PSS-A50, taking apart circuitry, and soldering wires to contacts and circuit boards -- so, this would void the warranty, and could possibly damage the keyboard if not done properly.

Beyond that, of course it is best to learn how to use the sustain pedal.  But not everyone who plays these keyboards has mastery of all keyboard playing techniques.  Otherwise, why have easy-play/single-finger chords on a $5500 Genos?  An "enhanced sustain" feature like this would be a great way to get a nice, full, sustaining sound, without having various dissonant notes mashed together, and without needing to use the pedal.  And also, like you said, there could be situations where hooking up and using a sustain pedal just wouldn't be practical, regardless of the skill-level of the keyboardist, so that such an "enhanced sustain" function would be useful there, as well.

[Lionel N]

If we consider that the sound is generated when you press the key (note on), and disappears when you release the key (note off), pressing the sustain pedal acts like if the "note off" is not taken into account; means, the sound will be progressively attenuate and will disappear.
Releasing the pedal stops all the remaining sound, except the ones from the keys currently pressed.

So when the "sustain button" is on, it's like if "note off" were note taken into consideration.
On the PSS-A50, the push button for sustain is a push, not a mechanical on/off. Means, press one time and sustain is activated untill you press another time and it will be disabled.

@PJD raised, in his post (http://sandsoftwaresound.net/pss-a50-midi-mod/) :
Our Japanese blogger considered adding a sustain input. The PSR-F50 dedicates SWLL pin 53 (PORTC0) to sustain. Unfortunately, the PSS-A50 has other ideas and SWLL pin 53 mutes the headphone output instead. You could put an external switch in parallel with the front panel sustain switch, but it toggles sustain and, thus, it doesn't behave like a true piano sustain pedal.

He also explains in another post :
Pressing the Sustain button has the following behavior:
-Sends new release time when sustain button is pressed.
-Release time messages are sent on both channel 1 and channel 2.
-Turning sustain off resets the release time.


I'm not sure that a hardware approach (transistor or similar) is the most appropriated (and maybe not easy to know the current state on or off for the feature, since there's no feedback -led or any-).

If we consider the synthé as a black box, once the key is released the internal synthé considers the not is off and the sound dissapears. So it's no use that the "magic box" plays the sound again, because the sustain effect will not be the expected one.

Some options to try :
-the reverb, as I wrote above
-or, when a chord is detected, the "magic box" (thru midi) plays the same notes in parallel as the one from the keyboard, but doesn't send the note off. There's the risk to hear twice the same notes because of the lag of the "magic box" and midi messages
-or, (in case the sustain button / sustain function is on in the PSS-A50), try to send a note off for the notes not currently pressed
-or, (in case, try to send a message "resets the release time." (as mentionned by @PJD)
-or any other...

For sure, the constrains tring to implement this with a "black box" (closed system) is more limitating that implementing it in a "white box" (in the Yamaha SW).