In a mathematical model, it is possible to do what is impossible in a real instrument, for example to limit string inharmonicity, controlling the quality and quantity of overtones for a better sound, and to model an ideal microphone that picks up all strings equally, without phase cancellation. I think Pianoteq could be a better instrument if it did not try to hide the fact that it is not a real piano and allowed itself to reduce the negative aspects of piano sound while accentuating the positive aspects. By negative aspects, I refer to inharmonicity of steel strings, thinness of sound in the lower register resulting from a weak fundamental and strong overtones, including inharmonic overtones, and phase cancellation resulting from microphone placement. To me, the problem with Pianoteq is that in their avid pursuit of realism, they seem more successful in reproducing the negative aspects of piano sound than in reproducing the positive aspects. One thing a sampled piano has to get absolutely correct is the tuning of the instrument, because there is little to nothing than can be done to correct mistuning, especially of unisons, after the recording is made. I prefer the NI sampled pianos over the Pianoteq modeled pianos, and of the sampled pianos, I prefer the Grandeur, which seems to be a better recording than Alecia's Keys, and the piano seems to be in a better state of tune. A better method would be simply to ask "which is your favorite?" and from that poll, select the two highest ranking pianos for the runoff. Otherwise you are giving equal rank to the two choices in the case where the person prefers one over the other. Mic position on pianoteq! Apparently the stereo image is different because of the mic position.Īsking "what are your two favorite pianos" is not a statistically valid polling method, unless the person is asked to rank their two choices. Use what you like, if you can, use it all! I even play with piano samples 32Khz (less than 1MiB, korg O3R from 1996) it sound's fine for rehearsal in my home when I am lazy to open the computer. More importantly, I had different reactions when playing and hearing to different sound. *But* when I played live I adapted to the sound even for "bad" pianos. garageband on iOS) pianos vs pianoteq (pianoteq is the only top-of-top piano instrument I have) and I realised the same midi file was ridiculous when played on different instrument. I tried the same with various cheap sampled (ex. It's a mistake to compare the *same* performance (midi file) to *different* intruments. To my knowledge this is only possible with pianoteq (Physical Modelling). But I prefer pianoteq, I like the harmonics (unnoticable in this example). With this midi file I liked the Grandeur.