Topic by AKM

  • Avatar
  • AKM
  • Member
  • Offline
  • From: Taganrog, Russia
  • Registered: 03-11-2014
  • Posts: 511

Topic: Recent una corda implementation problem

Just unexpectedly discovered that the current Pianoteq una corda realization is not correct. If you set the Unison width settings to the maximum you can check that while the Soft pedal is On the unison detune is still the same as if the Soft pedal is Off. I was always very suspicious about the "left pedal" realization and now I see why. I'd love to see it to be capable of gradual behaviour with a continuous pedal used so there are true 2-corda and una corda modelled sound available. This is absolutely crucial for a serious piano playing.

Andrei Kuznetsov
https://www.youtube.com/pianopictures

AKM's Website

Reply by Philippe Guillaume

  • Avatar
  • Philippe Guillaume
  • Admin
  • Offline
  • Registered: 25-04-2007
  • Posts: 5,249

Re: Recent una corda implementation problem

A quite standard una corda regulation on a modern grand consists in following: the three strings are still struck when it is engaged, but the strings are hit differently, two strings being struck by a softer part of the felt (the felt lying between the grooves of the normal strike) and the third one only by the edge of the hammer. Hence the unison behaves the same way for what concerns the frequencies (they are unchanged) but not the same way for what concerns intensity and phase. This change of intensity and phase is what you hear on a real piano, and it is modelled in Pianoteq.

Philippe Guillaume's Website

Reply by AKM

  • Avatar
  • AKM
  • Member
  • Offline
  • From: Taganrog, Russia
  • Registered: 03-11-2014
  • Posts: 511

Re: Recent una corda implementation problem

Thank you, interesting. Just checked, on my old Bechstein it clearly hits just 2 strings when full 'on'.

Andrei Kuznetsov
https://www.youtube.com/pianopictures

AKM's Website

Reply by Philippe Guillaume

  • Avatar
  • Philippe Guillaume
  • Admin
  • Offline
  • Registered: 25-04-2007
  • Posts: 5,249

Re: Recent una corda implementation problem

Yes, that can happen of course too, but in such a case, the two strings are often struck by two shifted grooves (where the felt is harder), and I find it gives a less interesting sound. But that is in the end a matter of taste.

EDIT. Actually it is not only a matter of taste: the una corda regulation I mentioned above is a safe way to  prevent a premature and irregular wear that would come from three grooves being hit sometimes by three strings (normal playing) and some other times by two of them (una corda playing), ending with two grooves being deeper that the third one. Striking between the grooves gives a longer life to the hammers. Of course, you could suggest to shift even more (by one and a half inter-string distance) but then you increase the risk of striking some neighbouring note, oups, horrible!

Philippe Guillaume's Website

Reply by AKM

  • Avatar
  • AKM
  • Member
  • Offline
  • From: Taganrog, Russia
  • Registered: 03-11-2014
  • Posts: 511

Re: Recent una corda implementation problem

Indeed, very interesting. I clearly understand that it's impossible to implement all requests but personally I'd like it very much to be able to play with different Una corda behaviours - 2 strings, 1 string, soft/hard felt. To my taste recent una corda implementation is kind of not 'different sounding' enough. Sure you can tweak it with a slider but it is just more/less softer sound, not something slightly "other" that I actually expect.

Last edited by AKM (30-01-2016 01:07)
Andrei Kuznetsov
https://www.youtube.com/pianopictures

AKM's Website

Reply by LTECpiano

  • Avatar
  • LTECpiano
  • Member
  • Offline
  • From: Columbia, SC
  • Registered: 05-05-2011
  • Posts: 694

Re: Recent una corda implementation problem

An interesting result is to load the Erard 1922, model - then play in the range of one octave or octave and one half below the top C.  You will hear a distinct difference in the timbre by actuating the left most pedal.  I have not played an actual Erard of any vintage but this little demo will produce a delicate and  lovely music box like result.  I have an associate musician friend and stage performer who just bought the model B.  His favorite models:  the model B and his second most favorite:  the Erard 1922 which has been in Pianoteq for a few years now.

Have fun,]

Lanny

LTECpiano's Website

Reply by Philippe Guillaume

  • Avatar
  • Philippe Guillaume
  • Admin
  • Offline
  • Registered: 25-04-2007
  • Posts: 5,249

Re: Recent una corda implementation problem

AKM wrote:

Indeed, very interesting. I clearly understand that it's impossible to implement all requests but personally I'd like it very much to be able to play with different Una corda behaviours - 2 strings, 1 string, soft/hard felt. To my taste recent una corda implementation is kind of not 'different sounding' enough. Sure you can tweak it with a slider but it is just more/less softer sound, not something slightly "other" that I actually expect.

Did you try the Model B une corda? I have the feeling that it offers a nicely 'different sound'. You can hear it on some demos on the Model B page, in Mozart for example, where the una corda is engaged at the beginning for a long period, same for Sriabin. I uploaded the two corresponding MIDI files (they are from the E-Competition) in the "Other files" section if you want to see when the una corda is engaged.

Philippe Guillaume's Website

Reply by AKM

  • Avatar
  • AKM
  • Member
  • Offline
  • From: Taganrog, Russia
  • Registered: 03-11-2014
  • Posts: 511

Re: Recent una corda implementation problem

@Philippe Guillaume
Sorry, I really don't hear it that much. I believe that it perfectly simulates the modern top grade grand pianos una corda behaviour, maybe it is because I'm used to not so perfect grands, more old ones. Also maybe because I'm using a consumer grade speakers and headphones I can't catch it.

@LTECpiano
Will try it in a sec, thank you. I was just about to ask what is it about that antique piano models. Is there a real single string Una corda among them? (I guess since the name is actually is)

EDITED: Overall I'm ok about the Una corda and sure there IS a difference just thought that maybe it is possible to suggest something to improve.

Last edited by AKM (30-01-2016 16:43)
Andrei Kuznetsov
https://www.youtube.com/pianopictures

AKM's Website

Reply by aandrmusic

  • Avatar
  • aandrmusic
  • Member
  • Offline
  • Registered: 18-04-2010
  • Posts: 265

Re: Recent una corda implementation problem

Philippe Guillaume wrote:

Yes, that can happen of course too, but in such a case, the two strings are often struck by two shifted grooves (where the felt is harder), and I find it gives a less interesting sound. But that is in the end a matter of taste.

EDIT. Actually it is not only a matter of taste: the una corda regulation I mentioned above is a safe way to  prevent a premature and irregular wear that would come from three grooves being hit sometimes by three strings (normal playing) and some other times by two of them (una corda playing), ending with two grooves being deeper that the third one. Striking between the grooves gives a longer life to the hammers. Of course, you could suggest to shift even more (by one and a half inter-string distance) but then you increase the risk of striking some neighbouring note, oups, horrible!

I find it interesting that you consider the hammers moving across to a string of the neighboring note 
(I know, you have experience as a technician of wooden/physical pianos)

Of course this could be the case in a PHYSICAL piano, but the virtual piano factory is able to space the strings of the neighboring notes... wherever they need to be.
They could be shifted until only one string is struck on a tri-chord, either by (simulated) groove or fluffy felt.
In fact their virtual position may only be relevant in relation to the soundboard and bridge - even then only their effect on the bridge and sound board seems relevant.

Reply by Philippe Guillaume

  • Avatar
  • Philippe Guillaume
  • Admin
  • Offline
  • Registered: 25-04-2007
  • Posts: 5,249

Re: Recent una corda implementation problem

aandrmusic wrote:

I find it interesting that you consider the hammers moving across to a string of the neighboring note 
(I know, you have experience as a technician of wooden/physical pianos)

Of course this could be the case in a PHYSICAL piano, but the virtual piano factory is able to space the strings of the neighboring notes...

Ah, yes, you are right of course ! For now we have developed the una corda pedal model for reproducing one position (the most common one I believe) but it would be indeed interesting to model all possible positions. Good suggestion for the future.

Philippe Guillaume's Website

Reply by SteveLy

  • Avatar
  • SteveLy
  • Member
  • Offline
  • From: Melbourne, Australia
  • Registered: 16-11-2015
  • Posts: 377

Re: Recent una corda implementation problem

I had a close look at the hammers and listened to the sound on our Kawai 6+1/2 ft grand and it's much like Philippe said: the hammers are moved over but still at least graze the leftmost string. Of course, this being an acoustic instrument and about 25 years old, there is variation and sometimes only two of three strings are hit.

IMHO Modartt/Pianoteq should take the same approach to Una Corda as they do to other parameters: push them further than physical limitations in a real instrument would allow: how far across the hammers are pushed, and how new or worn they are. OTOH, I think that much the same sort of effect can be achieved with the Unison width and Hammer hardness settings, so I don't see this as a feature that warrants urgent attention.

3/2 = 5

SteveLy's Website