Now, while sequencing the Aria from Bach's Goldberg Variations this time, my musical advisor commented on the strength of the attack. I have heard this comment before when I was performing Beethoven's Piano Sonata, and I tried to deal with it in various ways, but I could not find a good solution. I think it was especially noticeable at the beginning of each measure in this piece, as the dynamic did not change too much.
In fact, the same note can have a different sound quality depending on how the piano is played (different touches). Pianists often use various techniques to shape their sound, such as lifting their fingers, playing with the pads of their fingers, or tapping the back of the keyboard. I believe this is a creative way to express nuances beyond just dynamics.
Unfortunately, the sound libraries I have do not offer such advanced functionality. I tried various methods, such as adjusting velocity and using half-pedaling or the damper pedal, to reproduce a similar sound, but I couldn't quite achieve the desired expression. The Italian piano brand FAZIOLI has a model equipped with four pedals, which seems to fulfill that role. Recently, VSL released a sound library simulating this model, Synchron Fazioli_F308 and once again, I find myself wanting it.
This time, using Aria as the subject, I compared my current piano library (Vienna Synchron Concert D-274) with two newly upgraded options: Pianoteq8 by Modartt, a physical modeling piano, and Ivory3 by Synthogy, a sampling-based piano. All pianos used were either sampled or modeled from the flagship Steinway & Sons D-274 (Hamburg) model. When comparing under identical conditions, I found that for this Aria, Pianoteq 8 performed best, followed by Ivory 3, and then Synchron D-274. Pianoteq 8 provided the best connection between notes, the smoothest response to dynamics, and the most clearly articulated polyphony.
I found that Synchron connects sounds sampled at different velocity layers, whereas Pianoteq generates each sound individually through calculations. Ivory 3, on the other hand, fills in the gaps between sounds sampled at various layers using calculations. This appears to be causing differences in the smoothness of the sound transitions.
Also, Pinoteq's response to the position, orientation, and number of microphones is quite impressive. I am really impressed by the calculation that creates these things. Especially in the binaural setting, it was as if I was sitting on a piano chair and listening to the sound as I played, and I had the strange sensation of the sound spontaneously cutting through beautifully. This is a setting that is not used for listening to music in a concert hall, so I used the 3-mic setting for this aria.
As for the attack, Pinoteq and Ivory3 have a function that allows the user to adjust the hammer's hardness. I have tried a few things, and it seems that this will allow me to control the strength of the attack, which has been an issue until now.