I was wondering whether someone has put together a guide on how to change Pianoteq's parameters in a way that makes sense. I feel like every time I change something I just make it sound worse. I guess this is easier for a piano tuner, which I'm not, but I think it would help people out there trying to customise the sound to their liking. Is there anything similar out there that is still relevant for Pianoteq 9?
I also tend to keep my presets pretty close to the factory ones. I've never been able to really improve things with mic or EQ changes; every time I get something that sounds better in one way or in one register, it's at the cost of being worse somewhere else. But there are several things I consistently do that have paid off and I'll describe them.
First there's a post I want to link to but I can't find it. It was from a user from some years ago, named something like jfelice88keys, who had a guide to realistic-izing pianoteq pianos that went something like this:
- harden hammers a little bit in the grand staff to simulate the greater wear that this range gets in practice
- increase damping duration in the very lowest bass strings (big copper-wrapped monochord bass strings are harder to dampen effectively)
- set unison width to 0 for these same monochord bass strings
- turn the condition slider down a bit
- every day add a *very* subtle touch of randomness to a couple of the following per-note parameters: damping duration, strike point, unison width, detune, hammer hardness
Note, to do that latter thing in practice, what I do is, for each preset, I have a base version of that preset *without* any randomization, and then a subtly randomized variant of that base version. Sometimes, but not every day like he suggested, I'll delete that randomized variant and generate a new one, with different randomization, from the base version.
There was more to it than that, but those were my main takeaways that I've used in all presets.
On top of those things I like to:
- add just a little bit of blooming energy (see this post).
- soften the piano-level hammer hardness quite a bit (so that you can get a soft whispering pianissimo), also soften the mezzo hardness slider but not as much (so that there's not a jarring transition in hardness during play), and turn the forte hardness slider *up* a little bit to allow for a little more tonal contrast at the top end of the dynamic range
Those are the things that have consistently paid off for me. Except, with the new sombre presets, they don't need any hammer softening -- but they could maybe benefit from some hardening at the forte level.
Edit: I'll add some more about the hammer softening that I like to do. This was inspired by a time when, a couple years ago, I had the chance to briefly play on an acoustic Shigeru Kawai SK-EX in a showroom. This has been my one and only experience in life so far with a high-end acoustic grand. What amazed me was how gently whispering the sound could be when you just very very softly depressed a key, and then how much tonal variety there was as you struck a key with increasingly high dynamics. That was before I ever used Pianoteq. So since I found Pianoteq, I've sought to emulate that kind of tonal contrast in dynamics. It works better in some models than others. To do this, I usually adjust the "piano" level hardness slider to the point where it starts to sound TOO soft -- which is usually in the vicinity of 0.1 to 0.2 -- then turn it up just a little bit from there so that it's out of that weird zone.
- harden hammers a little bit in the grand staff to simulate the greater wear that this range gets in practice
This is genius!!
I also tend to keep my presets pretty close to the factory ones. I've never been able to really improve things with mic or EQ changes; every time I get something that sounds better in one way or in one register, it's at the cost of being worse somewhere else.
My experience is almost always the same; I usually return to the original preset and then (if I feel like experimenting) try to adjust it more conservatively or minimally. Mic positions seem especially problematic or tricky, maybe because of phase issues (attenuation or amplification of certain frequencies) among other things. I've thought that if I want to experiment with microphone placement and relative positioning, it might be advantageous to do that in a DAW with a frequency analyzer plugin active, in order to see in real time what the microphone movements are doing to the overall waveform and where resonances or cancellation seem to be occurring.
But there are several things I consistently do that have paid off and I'll describe them.
First there's a post I want to link to but I can't find it. It was from a user from some years ago, named something like jfelice88keys, who had a guide to realistic-izing pianoteq pianos that went something like this:
- harden hammers a little bit in the grand staff to simulate the greater wear that this range gets in practice
- increase damping duration in the very lowest bass strings (big copper-wrapped monochord bass strings are harder to dampen effectively)
- set unison width to 0 for these same monochord bass strings
- turn the condition slider down a bit
- every day add a *very* subtle touch of randomness to a couple of the following per-note parameters: damping duration, strike point, unison width, detune, hammer hardness
Many of these parameters can be randomized continuously (at each note-press on a keyboard) to a user-defined extent with the "humanize" function visible when right-clicking the parameters slider knob or button, as an alternative to the more static randomization available in the Note-Edit functionality.
On top of those things I like to:
- add just a little bit of blooming energy (see this post).
- soften the piano-level hammer hardness quite a bit (so that you can get a soft whispering pianissimo), also soften the mezzo hardness slider but not as much (so that there's not a jarring transition in hardness during play), and turn the forte hardness slider *up* a little bit to allow for a little more tonal contrast at the top end of the dynamic rangeThose are the things that have consistently paid off for me. Except, with the new sombre presets, they don't need any hammer softening -- but they could maybe benefit from some hardening at the forte level.
Those are great suggestions. Two adjustments I often make, that are complementary to each other (I usually do both rather than just one) for some of the presets is to slightly lower the impedance of the soundboard while giving a higher value to the Direct sound duration (a higher value (counterintuitively) seems to shorten the initial sound that is sent to the soundboard).
Edit: I'll add some more about the hammer softening that I like to do. This was inspired by a time when, a couple years ago, I had the chance to briefly play on an acoustic Shigeru Kawai SK-EX in a showroom. This has been my one and only experience in life so far with a high-end acoustic grand. What amazed me was how gently whispering the sound could be when you just very very softly depressed a key, and then how much tonal variety there was as you struck a key with increasingly high dynamics. That was before I ever used Pianoteq. So since I found Pianoteq, I've sought to emulate that kind of tonal contrast in dynamics. It works better in some models than others. To do this, I usually adjust the "piano" level hardness slider to the point where it starts to sound TOO soft -- which is usually in the vicinity of 0.1 to 0.2 -- then turn it up just a little bit from there so that it's out of that weird zone.
Seems like an excellent bit of tweaking to experiment with.
I also tend to keep my presets pretty close to the factory ones. I've never been able to really improve things with mic or EQ changes; every time I get something that sounds better in one way or in one register, it's at the cost of being worse somewhere else. But there are several things I consistently do that have paid off and I'll describe them.
First there's a post I want to link to but I can't find it. It was from a user from some years ago, named something like jfelice88keys, who had a guide to realistic-izing pianoteq pianos that went something like this:
- harden hammers a little bit in the grand staff to simulate the greater wear that this range gets in practice
- increase damping duration in the very lowest bass strings (big copper-wrapped monochord bass strings are harder to dampen effectively)
- set unison width to 0 for these same monochord bass strings
- turn the condition slider down a bit
- every day add a *very* subtle touch of randomness to a couple of the following per-note parameters: damping duration, strike point, unison width, detune, hammer hardnessNote, to do that latter thing in practice, what I do is, for each preset, I have a base version of that preset *without* any randomization, and then a subtly randomized variant of that base version. Sometimes, but not every day like he suggested, I'll delete that randomized variant and generate a new one, with different randomization, from the base version.
There was more to it than that, but those were my main takeaways that I've used in all presets.
On top of those things I like to:
- add just a little bit of blooming energy (see this post).
- soften the piano-level hammer hardness quite a bit (so that you can get a soft whispering pianissimo), also soften the mezzo hardness slider but not as much (so that there's not a jarring transition in hardness during play), and turn the forte hardness slider *up* a little bit to allow for a little more tonal contrast at the top end of the dynamic rangeThose are the things that have consistently paid off for me. Except, with the new sombre presets, they don't need any hammer softening -- but they could maybe benefit from some hardening at the forte level.
Edit: I'll add some more about the hammer softening that I like to do. This was inspired by a time when, a couple years ago, I had the chance to briefly play on an acoustic Shigeru Kawai SK-EX in a showroom. This has been my one and only experience in life so far with a high-end acoustic grand. What amazed me was how gently whispering the sound could be when you just very very softly depressed a key, and then how much tonal variety there was as you struck a key with increasingly high dynamics. That was before I ever used Pianoteq. So since I found Pianoteq, I've sought to emulate that kind of tonal contrast in dynamics. It works better in some models than others. To do this, I usually adjust the "piano" level hardness slider to the point where it starts to sound TOO soft -- which is usually in the vicinity of 0.1 to 0.2 -- then turn it up just a little bit from there so that it's out of that weird zone.
Thank you so much! I'll be trying this tomorrow
Many of these parameters can be randomized continuously (at each note-press on a keyboard) to a user-defined extent with the "humanize" function visible when right-clicking the parameters slider knob or button, as an alternative to the more static randomization available in the Note-Edit functionality.
I tried this last night and it made the piano sound worn. I didn't adjust the Humanize parameters, I just enabled it everywhere when available. I didn't dislike it though, it gives it more realism even if it sounds worn.
I've a preset right now where I'm using Humanize options.
If I may my advice is at first experiment. Use afterwards just where maybe you think they're really needed.
I tried this last night and it made the piano sound worn. I didn't adjust the Humanize parameters, I just enabled it everywhere when available. I didn't dislike it though, it gives it more realism even if it sounds worn.
The control points in the "Prob(ability) Density" section of the "Humanize" function can each be adjusted individually by the user, to narrow (or increase) the range of the randomization, as well as to move the center point of the range (the point nearer which most of the continuous automatic randomization of the parameter will occur).