Reply by cauldron

  • Avatar
  • cauldron
  • Member
  • Offline
  • From: Italy
  • Registered: 06-07-2022
  • Posts: 22

Re: Controller or DP with better action than VPC1

groovy wrote:
cauldron wrote:

KEY_SENSITIVITY, for each individual key you can specify a tolerance from -30% to +30% , positive if it sounds too soft and negative if it sounds too loud .

hmm, +/- 30% ... of what? ... of the time delta between two contacts?

To stay in your example: When an individual key has a time delta of 20 ms at pp, then the time delta can be varied by KEY_SENSITIVITY between 1.3 * 20 ms = 26 ms and 0.7 * 20 ms = 14 ms before it goes to the MIDI output converter of the Studiologic keyboard?


I can't know the implementation of this mechanism for the Studiologic keyboard.
The point was not to lose the dynamics of the MIDI velocity from 1 to 127.


I can tell you what I do with the VPC1 and the electronics that I replaced on the diode matrix. I do as you say. For example, the lowest octave sounds very soft and with a calibrated weight you can verify that the time interval is much greater than the keys of the adjacent octave. To simplify I use a linear curve to recalculate the velocity from 1 to 127 with the new min and max time delta values. If the average is 1ms (=127 vel) and 20ms (=1 vel) I could use 2.5ms (=127 vel) and 25ms (=1 vel), for example. In practice, I turn two wheels (delta time min and max) until I find a satisfactory setting with respect to the neighboring keys.
Of course there is also the velocity curve that then acts immediately after.

PunBB bbcode test

Last edited by cauldron (01-02-2025 23:02)

Reply by groovy

  • Avatar
  • groovy
  • Member
  • Offline
  • Registered: 12-06-2009
  • Posts: 722

Re: Controller or DP with better action than VPC1

cauldron wrote:

I can't know the implementation of this mechanism for the Studiologic keyboard.

Okay, then I know what you have done, but not what Studiologic implemented. I hope that the future will show, that each individual key of the new SL88 GT can be calibrated to MIDI velocity 1 - 127 when played ppp - fff.

Nice work, btw, cauldron! Back to the electronic & mechanical roots of decalibration. As an enduser, who only gets access to the MIDI layer, I would prefer to "filter" each individual key of the VPC1 with an individual, linear velocity curve - even if costs some native resolution. Hopefully Pianoteq Pro will have this 88-curves-feature sometimes, would be a reason to upgrade.

Reply by cauldron

  • Avatar
  • cauldron
  • Member
  • Offline
  • From: Italy
  • Registered: 06-07-2022
  • Posts: 22

Re: Controller or DP with better action than VPC1

Pianistically wrote:
cauldron wrote:

The controller VPC1 adds the MIDI offset at the end of the process, it may seem equivalent to the calibration (which is used at the beginning of the process)

Indeed the VPC1 offset is just a midi transform at individual process , but I am not  sure what you mean when you say that the calibration on the TP 400/W happen at the beginning of the process . To the best of my knowledge there is no modification of any physical characteristic of the keyboard when using touch sensitivity but it seems they apply positive or negative volume gain depending on the value chosen .  So with the VPC1 , you limit the range of output values and modify both volume & timbre if a non zero off value is entered as with the Sensitivity option you only change volume but keep the whole range of midi data. None of these workarounds are optimal imho . As you rightly mentioned if you apply a positive upset on a key on the VPC1 , you restrict the ability to play low velocities  , but on the Numa , given it is very easy to reach 127 , you will end with uneven volumes in f to fff range . I think that the most satisfying option is to use scripting in a DAW such as Logic PRO or Reaper and implement specific velocity curves with a s-Shape for the keys that need to be adjusted . Specific S-curves will preserve low velocities and high velocities as needed and will increase  responsiveness for the keys being adjusted . Another good option is to use Velpro software with allows  to enter  user defined velocity values at key level .


I agree with the various observations. Unfortunately the problem is that different products have different implementations and the configuration flexibility is not always so wide.
For example if I could use the Lachnit MK23 Studio keyboard ( http://www.flkeys.at/Products.html ) that uses the same Fatar TP-40W keyboard as the old Studiologic SL88 Grand but with optical sensors things would be much simpler acting after the midi event produced by the keyboard.
MK23 has a MIDI resolution of 4096 velocity CC#88 values. If 127 midi values are enough for you, you certainly don't lose resolution by mapping a portion of 4096 values on 127 values. It's different to create a correspondence of 127 values in another 127 values, there are some areas that will lose resolution with a granularity greater than 1.

For me too it was a problem not being able to attenuate the f/ff/fff dynamics of Studiologic NUMA X PIANO GT. I hope that the new SL88 GT MK2 can introduce a "SL Editor" of MIDI velocity curves like the old SL88 Grand did so that it can act only where needed and not on the entire dynamic range like the KBD_SENSITIVITY parameter does.

Reply by Pianistically

  • Avatar
  • Pianistically
  • Member
  • Offline
  • Registered: 12-07-2023
  • Posts: 174

Re: Controller or DP with better action than VPC1

cauldron wrote:
groovy wrote:
cauldron wrote:

KEY_SENSITIVITY, for each individual key you can specify a tolerance from -30% to +30% , positive if it sounds too soft and negative if it sounds too loud .

hmm, +/- 30% ... of what? ... of the time delta between two contacts?

To stay in your example: When an individual key has a time delta of 20 ms at pp, then the time delta can be varied by KEY_SENSITIVITY between 1.3 * 20 ms = 26 ms and 0.7 * 20 ms = 14 ms before it goes to the MIDI output converter of the Studiologic keyboard?


I can't know the implementation of this mechanism for the Studiologic keyboard.
The point was not to lose the dynamics of the MIDI velocity from 1 to 127.


I can tell you what I do with the VPC1 and the electronics that I replaced on the diode matrix. I do as you say. For example, the lowest octave sounds very soft and with a calibrated weight you can verify that the time interval is much greater than the keys of the adjacent octave. To simplify I use a linear curve to recalculate the velocity from 1 to 127 with the new min and max time delta values. If the average is 1ms (=127 vel) and 20ms (=1 vel) I could use 2.5ms (=127 vel) and 25ms (=1 vel), for example. In practice, I turn two wheels (delta time min and max) until I find a satisfactory setting with respect to the neighboring keys.
Of course there is also the velocity curve that then acts immediately after.

PunBB bbcode test

Hats off !! Great work .

Reply by groovy

  • Avatar
  • groovy
  • Member
  • Offline
  • Registered: 12-06-2009
  • Posts: 722

Re: Controller or DP with better action than VPC1

Pianistically wrote:

For me too it was a problem not being able to attenuate the f/ff/fff dynamics of Studiologic NUMA X PIANO GT.

... wait a minute ... you are talking about a problem with the internal sounds of the Numa X Piano GT?

Reply by cauldron

  • Avatar
  • cauldron
  • Member
  • Offline
  • From: Italy
  • Registered: 06-07-2022
  • Posts: 22

Re: Controller or DP with better action than VPC1

groovy wrote:
Pianistically wrote:

For me too it was a problem not being able to attenuate the f/ff/fff dynamics of Studiologic NUMA X PIANO GT.

... wait a minute ... you are talking about a problem with the internal sounds of the Numa X Piano GT?


Yes, I tried the internal acoustic piano sounds of the NUMA X PIANO GT keyboard but I have the impression that it can also happen with external sounds.

For classical use the most useful is VintageGrand in my opinion. There is a lot of variety of sounds but I think they were designed more for jazz or light music in general (I preferred the CFX sound of Yamaha P515 and WhiteGrand of Clavia Nord which seem more natural and fluid).
This is to say that perhaps Studiologic has chosen to make a velocity curve that allows you to play f/ff/fff without too much effort. Maybe it is not a problem but a choice and I think I could share it for that musical genre.

The real problem is that the new SL88 GT MK2 does not have internal sounds so I expect that you can customize the velocity curves otherwise it is a problem for me (KBD_SENSITIVITY is not enough and acts on all the velocity dynamics). However, NUMA X PIANO GT is an excellent instrument, very versatile and competitively priced, but for classical use only I think I would have preferred to use a Yamaha P515/P525 with the CFX sound.

Reply by groovy

  • Avatar
  • groovy
  • Member
  • Offline
  • Registered: 12-06-2009
  • Posts: 722

Re: Controller or DP with better action than VPC1

Thank you, I am just interested in the use case: SL88 GT as plain MIDI Controller for Pianoteq. All that matters for me is the quality of the "normal/factory-default" velocity-curve called Medium, that you mentioned in post 46, and whether the KEY_SENSITIVITY parameter has influence on the MIDI output at all and if, what it does MIDI-wise. Maybe we know more in March ...

Reply by cauldron

  • Avatar
  • cauldron
  • Member
  • Offline
  • From: Italy
  • Registered: 06-07-2022
  • Posts: 22

Re: Controller or DP with better action than VPC1

groovy wrote:

Thank you, I am just interested in the use case: SL88 GT as plain MIDI Controller for Pianoteq. All that matters for me is the quality of the "normal/factory-default" velocity-curve called Medium, that you mentioned in post 46, and whether the KEY_SENSITIVITY parameter has influence on the MIDI output at all and if, what it does MIDI-wise. Maybe we know more in March ...


Unfortunately I had to return the NUMA X PIANO GT but tomorrow I will receive the old SL88 Grand which is not necessarily to be discarded even though it has been produced since 2015. I can't do the checks you ask me (in any case that KEY_SENSITIVITY parameter acts on the entire dynamics from pp to ff as if it were a sort of volume wheel)

The TP-40W keyboard of the SL88 Grand is heavier to the touch than the TP-400W of the SL88 GT MK2 (or NUMA X PIANO GT).

If it can be useful to you I can answer your question for the old SL88 Grand.