Topic: PianoDisc versus Pianoteq

Hi All,

I've been looking at the PianoDisc system. https://pianodisc.com/prodigy2/


I know it is not the same sort of sytem as what Pianoteq offer, but they say that their system provides 1,024 levels of expression. Now, I can't find out what they really means, but I asume it is not simply a note velocity of 1 to 127, as what all digital piano offer.

Has anyone ever tried the PianoDisc system? I'm curious to see how a PianoDisc file works in a DAW.

All the best,

Rob

Re: PianoDisc versus Pianoteq

kenrob2037 wrote:

they say that their system provides 1,024 levels of expression. Now, I can't find out what they really means,

It means it can MOVE the keys of an acoustic piano with (nominally) 1,024 velocities, providing (nominally) that number of volume / timbre for each note.

Where do I find a list of all posts I upvoted? :(

Re: PianoDisc versus Pianoteq

dv wrote:
kenrob2037 wrote:

they say that their system provides 1,024 levels of expression. Now, I can't find out what they really means,

It means it can MOVE the keys of an acoustic piano with (nominally) 1,024 velocities, providing (nominally) that number of volume / timbre for each note.

Yes, but do you know how that translates electronically? I wonder what information it would have in a midi file?

Rob

Re: PianoDisc versus Pianoteq

Personally I believe MIDI 1.0 which has 0-127 velocity steps is already actually more steps than is required for natural realistic and extremely subtle or refined presentation of dynamics; fractions of a decibel per individual key are not the cause any limitation of realism in the sound, or emotional connection. I don't believe it is that factor that has any bearing whatsoever on the realism of midi piano -not from the perspective of player or a listener.

0-127 - It's an unnecessary technical limitation (if midi was being invented today), but not a real world issue - much as say 16 bit audio was never really a problem for realism of recordings versus 24 bit audio.

I suggest it's more about marketing to have more steps - because it's easy to do today, and more is always better on paper.

FWIW Midi 2.0 can now offer substantially more than this 1024 steps:

"With MIDI 2.0 Voice Channel Messages velocity is 16 bit.

The 128 Control Change messages, 16,384 Registered Controllers, 16,384 Assignable Controllers, Poly and Channel Pressure, and Pitch Bend are all 32 bit resolution."

Re: PianoDisc versus Pianoteq

Key Fumbler wrote:

Personally I believe MIDI 1.0 which has 0-127 velocity steps is already actually more steps than is required for natural realistic and extremely subtle or refined presentation of dynamics; fractions of a decibel per individual key are not the cause any limitation of realism in the sound, or emotional connection. I don't believe it is that factor that has any bearing whatsoever on the realism of midi piano -not from the perspective of player or a listener.

0-127 - It's an unnecessary technical limitation (if midi was being invented today), but not a real world issue - much as say 16 bit audio was never really a problem for realism of recordings versus 24 bit audio.

I suggest it's more about marketing to have more steps - because it's easy to do today, and more is always better on paper.

FWIW Midi 2.0 can now offer substantially more than this 1024 steps:

"With MIDI 2.0 Voice Channel Messages velocity is 16 bit.

The 128 Control Change messages, 16,384 Registered Controllers, 16,384 Assignable Controllers, Poly and Channel Pressure, and Pitch Bend are all 32 bit resolution."

Don't forget that velocity not only influence the db levels but also the tone .  What is important to consider is not the range at such i.e. 0 -127 but the level of possible step values with a given piano dynamic ( as per notation value) . For instance , if  you play a passage in a the p region around velocity 50 ,with good stability in your playing , with a good midi keyboard ( consistent midi output) the number of times you are going to hit the same velocity number and produce the same sound is high compared to the same passage played with intermediate values. It is particularly noticeable with quick repeated notes  that sounds unnatural in absence of round robin algorithm if you play with Midi 1.0 .

Is 0-127 a major limitation ? No.  One could say the same if you compare an MP3 audio file encrypted at 320 kb/s vs  a WAV file. The devil is in the details , and human ear perception about frequency changes is   high enough ( higher than perception of decibel levels changing) to justify Midi 1.0 specifications changes. This has been one of the driver to release 2010 Midi 1.0 Extended velocity Addendum ( The famous CC#88  with 16k level) prior to Midi 2.0 specs (Midi 2.0 is much more than just 65k levels of velocity)

Re: PianoDisc versus Pianoteq

Pianistically wrote:
Key Fumbler wrote:

Personally I believe MIDI 1.0 which has 0-127 velocity steps is already actually more steps than is required for natural realistic and extremely subtle or refined presentation of dynamics; fractions of a decibel per individual key are not the cause any limitation of realism in the sound, or emotional connection. I don't believe it is that factor that has any bearing whatsoever on the realism of midi piano -not from the perspective of player or a listener.

0-127 - It's an unnecessary technical limitation (if midi was being invented today), but not a real world issue - much as say 16 bit audio was never really a problem for realism of recordings versus 24 bit audio.

I suggest it's more about marketing to have more steps - because it's easy to do today, and more is always better on paper.

FWIW Midi 2.0 can now offer substantially more than this 1024 steps:

"With MIDI 2.0 Voice Channel Messages velocity is 16 bit.

The 128 Control Change messages, 16,384 Registered Controllers, 16,384 Assignable Controllers, Poly and Channel Pressure, and Pitch Bend are all 32 bit resolution."

Don't forget that velocity not only influence the db levels but also the tone .  What is important to consider is not the range at such i.e. 0 -127 but the level of possible step values with a given piano dynamic ( as per notation value) . For instance , if  you play a passage in a the p region around velocity 50 ,with good stability in your playing , with a good midi keyboard ( consistent midi output) the number of times you are going to hit the same velocity number and produce the same sound is high compared to the same passage played with intermediate values. It is particularly noticeable with quick repeated notes  that sounds unnatural in absence of round robin algorithm if you play with Midi 1.0 .

Is 0-127 a major limitation ? No.  One could say the same if you compare an MP3 audio file encrypted at 320 kb/s vs  a WAV file. The devil is in the details , and human ear perception about frequency changes is   high enough ( higher than perception of decibel levels changing) to justify Midi 1.0 specifications changes. This has been one of the driver to release 2010 Midi 1.0 Extended velocity Addendum ( The famous CC#88  with 16k level) prior to Midi 2.0 specs (Midi 2.0 is much more than just 65k levels of velocity)

If we are talking about tonal differences rather than velocity sound pressure level then I would suggest respectfully that there is far less than 127 perceptible changes within each piano key. It's great for a sample library to max that out (because it's there to be done).

Almost sufficient tonal differences are even heard at around just 12 layers of velocity. Extra levels are certainly nice (and I am a fan of physical modeling). Bring that up to 20 or 30 layers and you are talking extreme tonal subtleties, it is nice to know that you've got more but it's academic really - unless something went wrong in the sampling you wouldn't be able to tell the difference between say 115 and 116.

Also we don't just listen to one key at a time when we're listening to music, it's a blend of notes and timing - we don't get to concentrate on the individual keys tonal shift per individual key, mostly because music doesn't work that way anyway!

I have Midi 2.0 and MPE keyboards the audibly perceptible differences I believe will come more with say the likes of extreme pitch bend and removing the merest hint of stepping in rotary encoders being used to control synth parameters than in individual piano keys, which are percussive.

Re: PianoDisc versus Pianoteq

kenrob2037 wrote:

Hi All,

I've been looking at the PianoDisc system. https://pianodisc.com/prodigy2/


I know it is not the same sort of sytem as what Pianoteq offer, but they say that their system provides 1,024 levels of expression. Now, I can't find out what they really means, but I asume it is not simply a note velocity of 1 to 127, as what all digital piano offer.

Has anyone ever tried the PianoDisc system? I'm curious to see how a PianoDisc file works in a DAW.

All the best,

Rob

Hi Rob,

I'm also a Rob... piano technician and PianoDisc installer. I have an older Pianodisc IQ player system on my Baldwin 6'3" grand piano and the ProRecord MIDI key sensor rail for quiet time playing with a hammer stop rail. When you say 'PianoDisc' are you talking about the physical player/record system or the data + audio files the piano plays?

Last edited by vorpal (03-12-2024 16:22)
1929 Baldwin C 6'3" grand with ProRecord module
Pianoteq Pro 8.4 iPad Mini + USB Cable

Re: PianoDisc versus Pianoteq

vorpal wrote:
kenrob2037 wrote:

Hi All,

I've been looking at the PianoDisc system. https://pianodisc.com/prodigy2/


I know it is not the same sort of sytem as what Pianoteq offer, but they say that their system provides 1,024 levels of expression. Now, I can't find out what they really means, but I asume it is not simply a note velocity of 1 to 127, as what all digital piano offer.

Has anyone ever tried the PianoDisc system? I'm curious to see how a PianoDisc file works in a DAW.

All the best,

Rob

Hi Rob,

I'm also a Rob... piano technician and PianoDisc installer. I have an older Pianodisc IQ player system on my Baldwin 6'3" grand piano and the ProRecord MIDI key sensor rail for quiet time playing with a hammer stop rail. When you say 'PianoDisc' are you talking about the physical player/record system or the data + audio files the piano plays?

Hi Rob,

I'm interested in the data. Is it represented the same way as the data that comes from a digital piano?

What I mean, does its MIDI data have velocity numbers? If so what are the prameters of the velocity? Are they measured from 1 to 127? And, does varying velocity triger tonal variances?

Also, are the piano sounds sampled as Pianoteq has done with their Physical modeling?

Thanks,
Rob

Last edited by kenrob2037 (03-12-2024 22:38)

Re: PianoDisc versus Pianoteq

kenrob2037 wrote:

Hi Rob,

I'm interested in the data. Is it represented the same way as the data that comes from a digital piano?

What I mean, does its MIDI data have velocity numbers? If so what are the prameters of the velocity? Are they measured from 1 to 127? And, does varying velocity triger tonal variances?

Also, are the piano sounds sampled as Pianoteq has done with their Physical modeling?

Thanks,
Rob

Well, first off, PianoDisc files are not MIDI. They are digitally encoded audio files that are processed by the Pianodisc solenoid board controller which drive each individual piano key and the sustain pedal. You can't 'play' them on a file player as they are not a music file but a set of instructions for the mechanical player mechanism that tell the solenoid when and how hard to strike and how long to dwell and release.

It's a proprietary format that Pianodisc has been using since the 90's when I was installing the first units that used 3.5" floppy discs in the player controller. They moved to CD-ROMs (PianoCD) and now to wireless bluetooth to transmit the songs to the Pianodisc CPU. That's what my PianoDisc IQ system and the newer Prodigy systems use. It's essentially a music song library on an Ipad that plays wirelessly to the player piano.

It doesn't have a piano 'sample' to compare to as it's more like a set of instructions for a CNC machine. the 1024 levels of expression are the amounts the solenoids can be controlled to. It's not a digital piano with a modelled or sampled sound, its a controller for a mechanical playback system that can be added to pretty much any piano.

Now, MIDI files can be converted to PianoDisc files, I've used a couple encoding apps over the years. The output is an .mp3 audio file that plays the piano through the player controller, but it is no longer a MIDI file. You can't reverse an encoded PianoDisc .mp3 file to MIDI as it's now an audio file. It does not work as a 'music' file and can't be imported into a DAW or audio player, all you'll hear is digital high-pitched noise.

So to sum up, PianoDisc is a mechanical playback system that uses proprietary music data files.

1929 Baldwin C 6'3" grand with ProRecord module
Pianoteq Pro 8.4 iPad Mini + USB Cable

Re: PianoDisc versus Pianoteq

Hi Rob,

Thanks for this, most enlightening.

I was assuming it was like other systems I have seen on the market. Some of the others are VST based software, as you can choose different piano sounds and record a MIDI file.

Your system is more like the old player pianos, which were a self-playing piano with a pneumatic or electro-mechanical mechanism, but your is digital/mechanical. It must have been a fascinating journey developing your product.

I'm going to a piano shop next month in another state to view one with the PianoDisc system.

All the very best,

Rob

Last edited by kenrob2037 (03-12-2024 23:48)