Korg B2

That sounds good groovy - haven't looked at Korg for ages - but like some of those features, hope you enjoy

Just found, that the Korg B2 supports "quarter-pedaling" for sustain.

With the triple pedal unit Korg PU-2 I get four controller values:

0, 38, 74, 127.

I don't have a clue, how this is implemented technologically (?!)

Do you know other digital pianos, that are using 4 steps?

I have a Roland FP-10, which outputs only three values for sustain, 0, 127, and something around 60, can't remember exactly.  So literally supports "half pedaling", I guess.  This, or the 4 value setup in Korg B2, is simply a cost-cutting measure, and maybe more a purposeful "feature reducing" measure (to get you to buy more expensive model) than actually cost-cutting.  In any case, the splitting up of the analog signal into digital midi messages can be less precise if you only need to split the analog signal to three or four values, as compared to 128. 

I'm not often playing things that require better, but if I am I get my Yamaha Reface CP keyboard, which supports 0-127 sustain values, and run the sustain pedal through it.   Pianoteq doesn't care where the sustain pedal input is coming from.  In this case I have midi inputs into my pc from both the FP-10 and the Yamaha Reface, and Pianoteq can get midi messages from both.  I only do sustain from the Reface, which works seamlessly to alter the keypress midi messages from the FP-10.  In other words, it gives me a full-range "continuous" sustain pedal when playing the FP-10.

There are single-purpose solutions focused on converting a sustain pedals analog signal to "fully continuous" midi values ranging from 0 to 127.  This one looks good to me and is $54:  https://www.audiofront.net/MIDIExpression.php

Thanks, hesitz, for the info about the FP-10. That's why I'm surprised about the four discrete values of the B2 / PU-2 combi.

In my primary rig I'm using a continuous sustain-pedal with USB, that I built a few years ago -> Standalone continuous sustain pedal?

Since that day I'm spoiled and can't stand playing an on/off pedal any more . There is just a strange quirk with Pianoteq's Rhodes (and Wurlis too IRC) and this continuous pedal, sometimes it "yowls". I don't have this artifact with the Grand presets, here continuous pedaling works as it should.

With the "quarter-pedaling" the e-pianos presets work flawless and it seems 4 values are "just good enough" for a sustain pedal. The keyboard feel of the B2 reminds me a bit of a real Rhodes, that's why I needed another sustain solution. Another point is, continuous controllers are making much traffic on the MIDI-Bus and my B2-setup is based on a Raspberry Pi400, so I want to keep everything lean.

I have not opened the PU-2 pedal, but it seems to have steps of 0 Ohm, 5 kOHm, 10 kOhm and infinite at the two wires. So it is probably not a potentiometer inside. How could it be constructed?

I can see how the four values of B2 sustain are "just enough".  The three values of FP-10 seem much improved over on/off switch, but "not quite enough".

Regarding midi traffic, midi is so lightweight I can't imagine it's a problem on any RPi, much less an RPi400, which is actually pretty powerful.  So I doubt that's an issue; but I'm also sure there are ways to limit the granularity of midi messages being sent to Pianoteq, e.g., to 5 sustain messages/second, or 20/second, or whatever, not sure what would make you comfortable.  Haven't checked, but that sort of message limiting might be part of the software setup for the Audiofront convert device I linked.  Or, somewhere, I expect there is a ready-made or constructable utility app on Linux that you could route sustain midi messages through and have excess messages filtered out before forwarding only a subset to Pianoteq. But I highly doubt that's needed.  I think this is a case where the number of messages sounds like a lot to a human, but to the computer it's child's play; in any case, something to actually check in action on the computer before you decide it can't handle it.

I have no idea about the hardware implementation of continous or 4-discrete position sustain pedals.  With the FP-10, and the DP-10 continuous pedal I use, the DP-10 is a fully continous pedal; it's the hardware and/or software in the FP-10 that limits it to 3 midi values.  I sort of assume it's the same with Korg B2, or it would not function properly with non-proprietary continuous sustain pedals.  But if the B2 has a proprietary sustain pedal connector and is usable only with specific Korg pedal, then maybe pedal is implemented with multiple discrete values.  But my guess would be that it's same as FP-10, takes an actual continous analog input, which it splits into a very limited number of midi values.

Yes, indeed I did.