How long can a piano sound?

NY Steinway-D only lasted 57 sec at 10 meter of string lenght...  Uhhnn...
Unles the test had a mistake, there is something wrong.

Yes, it's a good idea to remamber about such details.

About reverb, changing subject a little bit, and maybe I'm saying some foolish, but: Would be possible a advanced reverb software allow to listen from the perspective of the audience, like in a given seat of a large piano auditorium or hall?

That's exactly it. I repeated the test several times using identical miking for both.
I also did not understand the reason for this disparity. Please, if possible, I recommend that you do the tests yourself and confirm the result.

Philippe did a test two weaks ago or so, from a request, and found NY to have longer duration than Hamburg, despite less louder. But I don't think it was tested in maximum string lenght adjust.

Did you used headphones adjusted for loud sound?

Each piano model have different original lengths for strings.
It would be wise to set all to the same lenght before perform the tests.

The lowest note (A-1) is rarely the one that provides the longest duration. The notes providing the longest sustain are usually the bass notes with double or triple strings (on the Steinway D, double strings start from F0 (MIDI #29) and triple strings from Bb0). Providing a longer duration is one of the main goals of having multiple strings. If there was enough room, manufactures would equip pianos with double strings down to the lowest notes, the problem being that they are too thick for that. Historical pianos often do have double strings down to the lowest notes, which is made possible by their smaller diameter. The drawback is that they have less power.

Unfortunately I haven't a precise answer to that good question. I have the feeling that double wound strings have less sustain than single wound strings because of a possible increase of frictional energy loss in the superposed copper coils, but that needs to be checked. Double wound string are used mostly in upright pianos and baby grands.

Just my 2 cents: if I had to measure something, I would not trust my ears (ageing, I admit !), even with headphones, I would use a waveform, magnified to the maximum...

These original measurements, relatively speaking, are not too far off from what I noted in that recent thread about the difference in sustaining power of the NY vs. Hamburg Steinways. I briefly checked the other pianos back then and noted similar relative sustain as recorded here. The reason why I brought it up originally was to point out the great improvements in longer, more realistic sustain, but also that I felt the NY Steinway overall could use a good bit more.

The measurable sustain of the NY vs. Hamburg Steinway does vary note to note, and as was mentioned, can flip in certain cases. However the perceived sustain power of the Hamburg Steinway is fairly consistently much longer than the NY Steinway's, which loses energy and power much quicker. That's not only my subjective perception, but can be measured noting how soon the sustain has a significant dB drop. You can see that other thread for reference, but basically even compared to my smaller NY Steinway M grand, the sustain of the Pianoteq NY is much less, and loses power much quicker (the Hamburg is about on par with my actual piano in terms of sustain length and power, when volume matched).

For authenticity, I think the total sustain time is important, but even more important is the perceived sustain time, or the "sustaining power" as mentioned above with how soon a significant dB drop occurs. This all actually has a profound effect on how the sustain pedal is used, when the ear tells the player it's time to repedal (reverb influences this too when playing in a lively hall vs. a dry studio). That's one of the reasons students learning on keyboards are guaranteed to have issues when they play on a real piano... though Modartt seems to be well on their way to eliminating that problem. The first as far as I know.

You're right. Although I did these tests, out of curiosity, I am convinced that a smoother decay curve is more important for playability purposes than the total sustention time.

Still, the Pianoteq is way ahead, as sample-based digital pianos don't last more than 30 seconds.

Yes I've played around with this before, and tried it out again just now. Reducing only the Direct Sound Duration does have a strong initial influence on the sustain's power, but after time the difference seems to be less pronounced. In other words, the first 5 seconds or so are effected much stronger than about 20 seconds in, at least to my ears without sliding it too far. That does have some benefit, but it's more the drop in amplitude around 10 seconds in, and even more around 20 seconds in that feels off to me. Also reducing the Direct Sound Duration too far quickly gives an unnatural sound to the first few seconds of sustain (too much initial higher overtone sustain). Tricky to balance! I've actually found that increasing the Direct Sound Duration a bit, but also increasing the Soundboard Impedance a bit seems to balance things a little bit closer to what seems natural to me. But this too is easy to go overboard with, ending up with a less authentic piano sound.

I recorded a little MIDI example which I think shows the bass sustain differences of the various pianos in a musical context. On a good grand piano you'll typically hear a strong pedal tone like this sustain with decent power for a significant time. I think the modeled Hamburg matches what I expect pretty nicely. Again, I do think total sustain time is also very important, just that "sustain power" is more noticeable in most musical contexts. Here's the MIDI file:

https://drive.google.com/file/d/11mmmyC...sp=sharing

That would indeed be interesting. Sympathetic resonance of the strings to external/ambient sounds would be interesting too, but probably not high on anyone's wish list... except maybe some experimental/avant-guarde composers?

We have just released version 7.2: https://www.modartt.com/pianoteq#pane_changes, we hope you will enjoy it!

Don't forget you can increase the Soundboard Impedance in the Design section, which allows very long sustain.

Hi Beto-Music,
Im' not sure the new "longer perceived sustein" should be added as a new feature, I see it more as a small correction brought to the NY Steinway D Pianoteq model during its youth. As intensive the beta tests were, there always remain some little details that deserve adjustments (and sometimes even later, when mature, because some new modelling feature has been added, as recently done when releasing version 7). The process is always something like: "there seems to be some weakness here, well, let's see what the reference instrument tells... Oh yes, that is something that can be improved, let's try", and so on.

Thanks once again. So this makes sense, this was a improvement (get closer to the real one) and not a enhancement (beyond the real one).

Yeah, a nice improvement policy. For a example of enhancement, like I said above, it's the extended keyboard range, specially for historic instruments and some short key range instruments. Today we have option for original or extended range to choose, preserving originality.

I think this sums up (more or less) what dedicated users tend to think about Pianoteq. There's lots of software that could be described as 'useful' or 'enjoyable' - but Pianoteq is so much more than just software!

Had Philippe pulled a rabbit out of the hat again?

Nathan, please record a nice example comparing 7.1 to 7.2
:-)