[LAD] ?==?utf-8?q? Potential MIDI headaches?

[Ralf Mattes]

Am Samstag, 19. Januar 2019 23:40 CET, Len Ovens <len at ovenwerks.net> schrieb: 
 

> 
> Cool. I do wonder where the sample sets are that actually have 127 samples 
> per note. 

No, those people don't use samples - we are talking full-size grand pianos here.
Those Disklaviers record and play back performances on actual concert instruments.
Think of it as a digital version of a Welte-Mignon player piano (the company has it's origins
here in Freiburg/Germany). The local museum does regular concerts with roles recorded by
famous soloist/composers from the first half of the last century.
IIRC the velocity resolution on those roles is higher than the 7 bits used by midi.
  
> Certainly Pianoteq might have a full range but most of the 
> electric pianos I have heard sound more like in "Bennie" than anything 
> that actually came from strings. I am talking about the people who walk 
> into a music store and buy an electric piano or other stage keyboard.
> 
> Now any of those people would prefer to sit down in front of an acoustic 
> piano, but none of them can afford (or are willing to afford) an electric 
> stage/home piano which actually sounds real. Remember that "most" people 
> would never think about using a keyboard controller to get sound from 
> their computer.

???
 
> In the case of keyboard synth combinations, where the signal path is 
> kb->midi->internal synth. MIDI 2 may show some improvements that even the 
> average person will notice. In time such an instrument may even be cheap 
> enough for "most" people. However, it seems to me that the synth in the 
> pianos I have seen does not even fully use the 128 velocity values 
> available now.

IIRC the Roli keyboards have better resolution and those are used quite often for
contemporary music performance.
 Some Casio E-Pianos seem to have high-resolution velocity values (14 bit), see
https://support.casio.com/storage/en/manual/pdf/EN/008/PX760_860_1500_160_AP260_460_MIDI_E_EN.pdf

> In terms of velocity vs. amplitude I would guess that 127 levels at 1db 
> per level covers more than most ADC's would show. At .5db per level the 
> range is still probably wider than the dynamic range available in a nice 
> quiet studio/sound stage... so I would hope that the range of timbre 
> differences makes a wider range of velocities worth while. I would like to 
> see a blind AB test where the same performance is rendered by the same 
> synth in both MIDI 1 and MIDI 2.

Not what our piano teachers say ;-)

Cheers, RalfD
 
> 
> --
> Len Ovens
> www.ovenwerks.net
> _______________________________________________
> Linux-audio-dev mailing list
> Linux-audio-dev at lists.linuxaudio.org
> https://lists.linuxaudio.org/listinfo/linux-audio-dev