Piano World Home Page Forums Our Most Popular Forums Digital Pianos - Electronic Pianos - Synths & Keyboards Digital dynamics

I don't know what your level of experience and knowledge is but with all due respect I think perhaps John Mortensen has a little bit more experience and knowledge about the music and the piano than you do? Or maybe not.

No, it does not.

I certainly can. This guy in video makes some very good points but at the same time very bad ones, tone/volume count is one of them. While practicing I often try to get that one level of tone I would really like and on good instrument you really can play with that. Another problem with the video is that he says speed is the only factor in playing, actually there is one more, hammer travel distance, if you won't hit key to the full stop for example the hammer will barely touch the strings, so with the same speed the strings will be affected a bit differently.

Nordomus and DazedandConfused. I suggest you listen to a few more of John Mortensen's videos. You both may learn a lot.

I wonder what Mortensen’s intent is? Is he saying, “I have defined the amplitudes of 10 discrete volume levels on an acoustic piano. Here is a chart showimg the ampltudes. I will now pick out of a hat a random number between 1 and 10, play that amplitude, record it, and you will see it matches the correct amplitude I defined.”

I take it his day job is a magician? Clearly not a pianist. Who has time for that s**t?

This is totally wrong - has been disproven by many published papers - and has been discussed many times at PW. I will now shout. Please forgive me.

THE VELOCITY OF THE HAMMER AT THE INSTANT IT STRIKES THE STRING IS THE SOLE DETERMINANT OF TONE, when all other variables - no use of sostenuto pedal, no previous notes still sounding, no repetition, no una corda are removed.

OK. We've got 8 levels for one note. For example, we need to play 3 notes. How many combinations can be created? 7+8+7, 6+7+8, 8+8+8, 8+7+6 and so on and so on. It is only for the 3 notes. And how many notes are there in an everage piece of music? - yeah, that is why musical variety is endless even with this 8 levels.

There is a saying "one note on the piano is no music".

It takes a lot of control to play a part in 10 clearly defined levels.
You don't necessarily have to be able to perform that.
But within your current playing, if you shape the music from tone to tone, you will have much finer differences in the levels.
You just can't get that steps without the context.

-rhodes74

All this reminds me of "Spinal Tap" --- Yeah, but it goes up to eleven..... ;-)

In Pianoteq, there is a "dynamics" setting from 0-100 dB. The standard modeled pianos are set around 40-50 dB. But personally, I enjoy playing the modeled pianos with the dynamics set to 100 dB. This gives me amazing colors and much better control at each level. Crescendos/descrescendos feel like they go forever.

Yes, many sampled piano have recorded sound for each of the 88 notes multiplied by the number of layers and the combination of pedals (una corda, sustain). We can also have release samples and overtones (sympathetic resonance).

The question really is how precise is the optical sensor (or many sensors) that detects a keystroke velocity on a given DP. I don't really know. I tend to think that the high-end DPs can detect all 128 dynamic levels supported by MIDI, but I'm not sure about low-end DPs. The only way to check it out is to record your playing in MIDI format on your DP and then to analyze MIDI file on a computer. If you see random key velocity values in 0-127 range, than full MIDI range sems to be supported, but if you see some quantization (like 4, 8, 12, 16, etc.) than probably the number of detectable dynamic levels is reduced on that DP.

How many dynamic levels are needed? Suppose you need to play a 4-octaves scale with crescendo. To play it with truely even crescendo you need 29 levels.


Besides I find very interesting what DazedAndConfused has said above:
http://forum.pianoworld.com/ubbthreads.php/topics/2773594/re-digital-dynamics.html#Post2773594


lol

Yes. I seem to remember in the publicity blurb that PHA50 outputs at 16bit to the modelling engine. I'll see if I can find a link.

Analogue television had an 'infinite' number of levels - so did analogue audio tape if it comes to that, but their digital equivalents have a far greater resolution (or dynamic range). Perhaps it's similar with analogue and digital pianos. I shouldn't be at all surprised.

DPs, as long as they're worthy of the name, have plenty of dynamic range. The MIDI standard of 128 levels is more than enough. But there's nothing restricting internal sound engines to that specification anyway. I think the newer Roland pianos have over a thousand.

8 bits is not good enough for classical piano in terms of either level or timing. I believe Yamaha use a much higher resolution for their ePiano Disklavier competitions for good reason. You can prove it yourself. Play a piece with plenty of rubato and record the audio out while recording midi data into a sequencer. Now run the midi data back to the piano and record the output. Compare the audio outputs they should be the same? They won't be. And one will sound weird and the other won't, just like every classical midi file.

Remember that dB is a ratio, so saying the Pianoteq has a dynamic range of 100dB is fine, but, relative to what baseline? Is the baseline is SPL (Sound Pressure Level), then the lowest level a human can hear is 0dB, but the noise level in an average home is about 50dB, so you need a minimum dynamic level on the piano to be heard above the noise.

The practical dynamic range of a full orchestra is 65dB - of a trumpet is 30dB - of a violin is 35dB.

The noise level of a baffled air exchange system in a concert hall is about 35-40 db.

Measuremens of live concerts in those halls show a dynamic range for a typical concert pianist of about 20dB.

Yes, see https://en.m.wikipedia.org/wiki/Disklavier, the Disklavier pro has 1023 levels of velocity.

Note that resolution is not equal with precision. On a Disklavier the playback could induce a near linear transformation of velocity.

About the timing, it’s more complicated. A SMF file is not limited. The time is expressed with a fraction of a quarter note, and the fraction can be choosen on a DAW. We can see 1/480 or 1/960 of a quarter note. Then a SMF record (floppy, flash drive...) has a quite high limit. On a MIDI cable, a single event needs around 1ms... and 2ms if you use an enhanced resolution. If you play a very synchronised chord, each notes will be sent one per 1ms or 2ms.

Thank you for making this point.And I'd like to add ...
Pay attention to the sound, not to the specs.
Sound makes art. Specs do not.

I cannot find a link right now, it might have been a video where I learned this, but millions is more accurate. There were 16,000 internal levels during the SUPERNATURAL sample + modelling days.



https://www.rolandcorp.com.au/blog/magic-touch-brief-history-rd-series

Um... not quite. Analog TV eventually runs into discrete photons and electrons. For digital cameras, there's a limit to the number of photons that it takes to saturate a photosite. According to Grass Valley, it's about 15,000. That means that a 14 bit number would be sufficient to count them. Of course real world equipment can't actually count them....