Pianistically wrote:You can clearly the difference particularly in the opening when the resonance plays a very important role. Cheers
Oh wow, yes, it's visible to the naked eye without further processing. That matches my subjective impressions so far of a wetter voicing with more resonance, which sounds significantly more natural and spatial than when I boosted resonances in v8. Thank you so much for sharing!! I'm glad it's not the placebo effect
So far, I'm very, very impressed with v9. All pianos sound at least a touch better, but some have been transformed (The Blüthner blew my socks off!). The global hammer hardness is a much smarter way to parametrise this feature, experimentation is much smoother this way. And since the thunder pedal seems to actually extend the dynamic range (rather than just remapping velocities) it adds a lot of new expressions to the mix. What a slam dunk!
You are welcome. I'm quite impressed too. Enjoy !
]]>You can clearly the difference particularly in the opening when the resonance plays a very important role. Cheers
Oh wow, yes, it's visible to the naked eye without further processing. That matches my subjective impressions so far of a wetter voicing with more resonance, which sounds significantly more natural and spatial than when I boosted resonances in v8. Thank you so much for sharing!! I'm glad it's not the placebo effect
So far, I'm very, very impressed with v9. All pianos sound at least a touch better, but some have been transformed (The Blüthner blew my socks off!). The global hammer hardness is a much smarter way to parametrise this feature, experimentation is much smoother this way. And since the thunder pedal seems to actually extend the dynamic range (rather than just remapping velocities) it adds a lot of new expressions to the mix. What a slam dunk!
]]>Pianistically wrote:I ran a comparison in audacity of one midi file rendered with same preset and parameters with both Bösendorfer 280 VC and Kawai Shigaru EX . I mix and render the version 8 rendition after inversion and version 9 . The mix shows a non null audio file result
Out of interest, in terms of amplitude/energy, roughly what %age of the sound does the difference amount to? Obviously there isn't a 1-2-1 relationship between sound quality improvement and that measure, but I'm curious nonetheless
Hello Daniel , here is a screenshot of the same preset Shigaru Kawai Player V8 vs V9 . Midi recorded is moments musical op 16 Rachmaninoff intro, lot of sustain pedal and half pedals , so lot of resonance. Y axis in Audacity is the normalised amplitude of the signal so that it fits in a -1 + 1 range. You can clearly the difference particularly in the opening when the resonance plays a very important role. Cheers
]]>Pianistically wrote:I ran a comparison in audacity of one midi file rendered with same preset and parameters with both Bösendorfer 280 VC and Kawai Shigaru EX . I mix and render the version 8 rendition after inversion and version 9 . The mix shows a non null audio file result
Out of interest, in terms of amplitude/energy, roughly what %age of the sound does the difference amount to? Obviously there isn't a 1-2-1 relationship between sound quality improvement and that measure, but I'm curious nonetheless
I will post a graph tonight when I am home . But the delta is not negligible. I’m some portions the delta in energy is very visible which would to indicate more partials .
]]>In fact it ranges from -1 to +1. The more the value is below 0, the harder it gets to play fff (like if you had pulled down the velocity curve). The more the value is above 0, the easier it gets to play fff (like if you had pulled up the velocity curve). Observe BTW how the ppp pp mp... fff letters move while changing the offset.
--- --- ---
The velocity-offset feature is great. A handy, efficient way to create a customized soft/muted/dampened piano by moving the velocity-offset downward into negative values, optionally soundboard-impedance decay to lower values, Hammer-hardness to lower values, direct sound duration (conterintuitively) to higher values, and perhaps unison-width and condition sliders to higher values, and Action->Key Release Noise, Damper Noise, and Voicing->Hammer Noise to higher values (and possibly modifying Mute and Damping Duration, etc.), all features/parameters available in or from the main Pianoteq GUI window.
--- --- ---
]]>YvesTh wrote:Perfect, that's what I was hoping for. However, I have a problem saving these settings. Once I have defined my global curve and individual corrections, I save them as “personal velocity” preset (I have checked “enable global velocity curves.)” When I change instrument or presets, the global curve remains (ok), but the corrections in "note edit" disappear. If I reactivate my “personal velocity” preset they come back (so they are saved correctly in "velocity perso"). Is it a bug ? I would like to be able to keep these corrections for all pianos without having to use the freeze tool. Thank you very much.
Translated with DeepL.com (free version)
Indeed it can be saved in customised velocity curves but it is not saved as global velocity. The idea behind being that it is used for customising presets.
Okay, never mind, let's wait for v10 . Thank you very much for your prompt response.
]]>Perfect, that's what I was hoping for. However, I have a problem saving these settings. Once I have defined my global curve and individual corrections, I save them as “personal velocity” preset (I have checked “enable global velocity curves.)” When I change instrument or presets, the global curve remains (ok), but the corrections in "note edit" disappear. If I reactivate my “personal velocity” preset they come back (so they are saved correctly in "velocity perso"). Is it a bug ? I would like to be able to keep these corrections for all pianos without having to use the freeze tool. Thank you very much.
Translated with DeepL.com (free version)
Indeed it can be saved in customised velocity curves but it is not saved as global velocity. The idea behind being that it is used for customising presets.
]]>YvesTh wrote:Philippe Guillaume wrote:In fact it ranges from -1 to +1. The more the value is below 0, the harder it gets to play fff (like if you had pulled down the velocity curve). The more the value is above 0, the easier it gets to play fff (like if you had pulled up the velocity curve). Observe BTW how the ppp pp mp... fff letters move while changing the offset.
Thank you, that's what I understood, but my question was more about how the shift occurred along the curve. For example, with offset=+0.5, will the velocities vary identically (5 becomes 10, 40 becomes 45, and 100 becomes 105), or will the curve be distorted (5 becomes 6, 40 becomes 45, 60 becomes 70, and 100 becomes 110) like if normal keyboard curve become slow keyboard curve?
Yes it will be distorted non linearly, as you say, similarly to the velocity presets called "moderately slow keybard" etc.
Perfect, that's what I was hoping for. However, I have a problem saving these settings. Once I have defined my global curve and individual corrections, I save them as “personal velocity” preset (I have checked “enable global velocity curves.)” When I change instrument or presets, the global curve remains (ok), but the corrections in "note edit" disappear. If I reactivate my “personal velocity” preset they come back (so they are saved correctly in "velocity perso"). Is it a bug ? I would like to be able to keep these corrections for all pianos without having to use the freeze tool. Thank you very much.
Translated with DeepL.com (free version)
]]>Pianistically wrote:for info , I ran a comparison in audacity of one midi file rendered with same preset and parameters with both Bösendorfer 280 VC and Kawai Shigaru EX . I mix and render the version 8 rendition after inversion and version 9 . The mix shows a non null audio file result , therefore all instruments benefit from changes introduced in the latest model , including the latest ones above.
The new Pianoteq 9 sounds phenomenal with a very rich sound and the latest UI improvements (new soundboard graphic design, velocity offset at note level , mics mixing, global hammer slider, and LED switches) are really well thoughts. Sombre presets are also dead gorgeous for cinematic or atmosphere renditions. I am really impressed by the massive amount of changes between version 8 and 9. Congrats to the Modartt team and all contributors !just trying to understand, so you were trying to do a null test between same settings of different models between PTQ8 and PTQ9?
One of the beautiful things of PTQ is that sound is synthesised in real time with subtle differences for each key press, so you wouldn´t be able to null one model in PTQ8 with the exact same model in PTQ8 (no typo, I mean the same version) - let alone changing versions.
If you look at the spectrum of a note as it decays you'll see that there is a lot of modulation in both amplitude and frequency, and no two notes - when played repeatedly - are exactly the same, so no chance of canceling each other out. and that is leaving the transient aspect aside.
on top of that, on average, PTQ9 plays a few 1/10s of dB louder than PTQ8, all else being equal, so no chance of a null from there too.but maybe I got you wrong. not trying to be confrontational or anything, just trying to understand your methodology and purpose.
no , this is is not the test I have done and apologies if I wasn’t clear enough .
I recorded a portion of Rachmaninov ‘moments musicaux’ and save the session in a midi file .
Then I rendered the midi file with Bosendorfer 280 VC in version 8 and 9 with same preset and same user parameters and compared the audio files in audacity .
Afterwards, I applied the same method with the Kawai Shigaru .
In both cases, there were differences in the audio files ( WAV 24 bits) produced comparing the same piano with the different engines .
The rationale for doing the test was in original post .
I need to stop thinking about this lest I lose my entire weekend to this
]]>I don't have the knowledge nor the skills/tools to do this myself, but to me it would be very interesting to hear what the difference between the versions sounds like, isolated. I also wonder if subtracting the ptq9 sound from the ptq8 sound is any different from subtracting the ptq8 sound from the ptq9. Would it be like listening to what they removed vs what they added, or is it the same. My gut feeling is that its the same thing, or it depends on how the substraction is done.
subtracting in this context, if I understand your question, means adding two sounds together but inverting the phase of one of them. in the digital world sound is represented by waves that oscillate between a maximum of 1 and -1 (or anywhere between). if you have the same frequency and amplitude for any given wave but invert the phase of one of them, adding then together will result in 0, so, silence. which one gets inverted is irrelevant. really easy to implement this in a DAW and check for yourself.
]]>for info , I ran a comparison in audacity of one midi file rendered with same preset and parameters with both Bösendorfer 280 VC and Kawai Shigaru EX . I mix and render the version 8 rendition after inversion and version 9 . The mix shows a non null audio file result , therefore all instruments benefit from changes introduced in the latest model , including the latest ones above.
The new Pianoteq 9 sounds phenomenal with a very rich sound and the latest UI improvements (new soundboard graphic design, velocity offset at note level , mics mixing, global hammer slider, and LED switches) are really well thoughts. Sombre presets are also dead gorgeous for cinematic or atmosphere renditions. I am really impressed by the massive amount of changes between version 8 and 9. Congrats to the Modartt team and all contributors !
just trying to understand, so you were trying to do a null test between same settings of different models between PTQ8 and PTQ9?
One of the beautiful things of PTQ is that sound is synthesised in real time with subtle differences for each key press, so you wouldn´t be able to null one model in PTQ8 with the exact same model in PTQ8 (no typo, I mean the same version) - let alone changing versions.
If you look at the spectrum of a note as it decays you'll see that there is a lot of modulation in both amplitude and frequency, and no two notes - when played repeatedly - are exactly the same, so no chance of canceling each other out. and that is leaving the transient aspect aside.
on top of that, on average, PTQ9 plays a few 1/10s of dB louder than PTQ8, all else being equal, so no chance of a null from there too.
but maybe I got you wrong. not trying to be confrontational or anything, just trying to understand your methodology and purpose.
]]>Philippe Guillaume wrote:Wow, what a difficult question! I guess you are speaking about the condition parameter? It is programmed in terms of parameters variations, not in years, and how fast it varies depends on the piano brand and model, the climate, etc. Sorry for the cloudy answer
Oh no, not at all! I was just interested in the details of Pianistically's experiment of subtracting the v8 audio from the v9. They said they've seen a difference and I wanted to know the weight of the difference.
In general, I appreciate that this is a reductive question, as the v9 improvements will be the result of careful parameter balancing. I'm just asking out of pure, unproductive curiosity .
Edit: Ah, I now see that writing '%age' must be confusing - I meant this as 'percentage', not 'age'. Sorry!
I don't have the knowledge nor the skills/tools to do this myself, but to me it would be very interesting to hear what the difference between the versions sounds like, isolated. I also wonder if subtracting the ptq9 sound from the ptq8 sound is any different from subtracting the ptq8 sound from the ptq9. Would it be like listening to what they removed vs what they added, or is it the same. My gut feeling is that its the same thing, or it depends on how the substraction is done.
Anyhow. I have not yet been able to test the new version as Im waiting for the iOS version to be released. When listening to youtube videos, I can honestly not tell the versions apart in any way, so Im really hoping the changes are noticeable when playing, and also when not wearing headphones. When playing without headphones, tiny details in the sound make practically no difference. Its more the overall sound quality and the character of the piano that shines through, and I never play with headphones. Im doubtful this version will have a significant impact for me personally, but I will gladly pay for it anyway given that I want the development to continue. Lets hope Im wrong. I can hardly tell piano libraries apart sometimes when watching youtube videos, and then when Im actually playing the plugins myself, the difference is huuuge.
]]>Wow, what a difficult question! I guess you are speaking about the condition parameter? It is programmed in terms of parameters variations, not in years, and how fast it varies depends on the piano brand and model, the climate, etc. Sorry for the cloudy answer
Oh no, not at all! I was just interested in the details of Pianistically's experiment of subtracting the v8 audio from the v9. They said they've seen a difference and I wanted to know the weight of the difference.
In general, I appreciate that this is a reductive question, as the v9 improvements will be the result of careful parameter balancing. I'm just asking out of pure, unproductive curiosity .
Edit: Ah, I now see that writing '%age' must be confusing - I meant this as 'percentage', not 'age'. Sorry!
]]>Pianistically wrote:I ran a comparison in audacity of one midi file rendered with same preset and parameters with both Bösendorfer 280 VC and Kawai Shigaru EX . I mix and render the version 8 rendition after inversion and version 9 . The mix shows a non null audio file result
Out of interest, in terms of amplitude/energy, roughly what %age of the sound does the difference amount to? Obviously there isn't a 1-2-1 relationship between sound quality improvement and that measure, but I'm curious nonetheless
Wow, what a difficult question! I guess you are speaking about the condition parameter? It is programmed in terms of parameters variations, not in years, and how fast it varies depends on the piano brand and model, the climate, etc. Sorry for the cloudy answer
]]>