I'm testing another DP at the moment. The Korg D1 via 5-pin MIDI-out and a Roland UM-One Mk2 USB-interface to my lightweight Linux-laptop. The D1 with the Roland interface seems to have a shorter overall latency than my former setup. At the same time it has a louder mechanical key noise than I'm used to. This has two consequences in my opinion. The pianosound "feels" a tad too early and the attack of Pianoteq is unpleasently covered / overlapped by the acoustic key noise. I could face it with sheer volume boost like a real piano does, but I tried another strategy ...

Increasing the latency of Pianoteq.

Normally I used that laptop with 64 samples Buffersize at 44100 Hz. Given that Pianoteq 9 has latency of 3 times the buffer size [1] this means a latency of:

64 / 44100Hz * 3 = 4.4 ms

Now I have chosen 128 samples Buffersize at 48000 Hz. This means a slighly higher latency:

128 / 48000Hz * 3 = 8.0 ms

In other words the latency is 3.6 ms higher now.

It is not too much to be unacceptable in my opinion, but distant enough to better separate the mechanical key noise from Pianoteq's attack sound.

[1] Pianoteq 9 README_LINUX.txt
"* with Jack or Alsa, use buffer sizes that are multiple of 64, as Pianoteq prefers those sizes. With the Alsa driver, the latency is 3 times the buffer size (a buffer of 64 samples at 44100Hz gives a latency of 4.3 ms)."

Warmest regards,

Chris

Absolutely if anyone hasn't tried what you described, it could be well worth it. For the sake of realism, it could be as important as a fitting velocity curve.

I like high latency numbers in Pianoteq for piano playing - PC/cable latency seems very low so I like to see 20ms in Pianoteq - and over time am fine with it for other music (like synths) too.

If anything some comfortable amount of extra latency and perceived "grand piano feel" has helped with musicality in my modern music types. Gives me a good neurological confirmation bias if this is set this way.

When playing some music requiring less kinetics, I find that it can be good to make the velocity curve a little quicker and play softer, rather than shortening latency.

Until I began to play piano, played some real grand pianos, got used to it, and coming from modern music in past years, I aimed for zero latency for some time in Pianoteq (can get 1.2ms)

Zero latency doesn't make sense to me, for physical piano playing.. that's some supposed baseline goal (since dawn of the MIDI age my music circles was awash with this talk). It was best for synth lines, bands and recording, playback and on - esp. simplistic pop lines and so on. But for solo piano, classical music etc. it's good to try different buffer sizes as groovy shows.

Would definitely encourage pianists to experiment with this - it can be a qualitative difference, in how realistic their dpiano feels.

But today I had an Aha-effect. I'm testing another MIDI keyboard at the moment. All other parameters in my Laptop/Pianoteq environment were unchanged.

I noticed, that the sound appeared faster than with my older keyboard. The latency seemed to be shorter now, but to a point that it felt a little "too early"!

So I increased the latency in small steps. Start point was my usual buffer setting in qjackctl: buffersize 64, periods 2, samplefreq 44.1kHz. Then I increased the periods to 3 and to 4. In other words I played Pianoteq with three different nominal latencies:

a) 2.9 ms
b) 4.4 ms
c) 5.8 ms

To my surprise the middle latency (4.4 ms) felt just damn right! Most natural sync between tactile, mechanical feedback and piano tone.

My guess is that the older keyboard had another timing caused by the geometry of the keys/contacts OR slower key scanning OR a slower USB MIDI interface (non-exclusive OR).

Assuming the new keyboard has a shorter over-all latency, I can use bigger buffersizes now.

I have not measured the differences between the two MIDI keyboards so it is possible I'm just a victim of the placebo-effect.

Anyway, the immersion is amazingly better with that sync of tactile and audio latency.

cheers