Beto-Music wrote:

It's not always easy to describe sounds in words.

What exactly do you mean by "granularity" in a sound ?

Do you have a example, a exagerated example, of any "granular" sound, to give us a good idea ?

Is it presente in the entire range (bass to trebble) or not?

I figured out the source.  The algorithm that simulates the resonance inside the piano body (different from the reverb) is not fully evaluated to a time step equivalent to the sample rate; it's definitely lower.  I'm sure this is because the real model is too slow to run in real-time.

The sound reflections are not fully solved to the timescale, so they arrive in little packets that are further apart than they should be, and this makes the sound grainy.

If I'm right, Modartt needs to lower the time step on this sim.  I don't mind if it isn't real-time.  It should be made an offline toggle or something.

It also sounds like the areas of the piano case that should be curved are partially faceted instead, perhaps to further decrease processing time?

The last sentence is "fascinating" (to ME).
I am quite sure the inside curve of a piano case can be represented mathematically and that sound reflections from such a curved surface could be represented.
I "assume" that the curved surface would be approximated as many small facets - the size and quantity being a design decision, perhaps based on computation power needs (& availability), but in any case finite. 

IF (if, if, if) Pelpix is right on this - and IF it is the sole cause then it might be POSSIBLE to get an acoustic piano to produce a similar effect, e.g. by placing some number of short straight boards around the curves.

Maybe, maybe not.

Trying to find out what PelPix means by granular sound, I used a couple of low notes using a modified U4 preset to bring out as much as possible the raw synthesized sound.

Here is the fxp and test for U4.

http://www.forum-pianoteq.com/uploads.php?id=1685
http://www.forum-pianoteq.com/uploads.p...und_U4.mp3

This example shows the first 20 seconds or so with a normal U4 Tall preset, sounding quite good to me, letting the last note decay, and then the "raw" preset playing the same notes, and made to decay much longer.

If you listen again to the first portion, you may now notice more of this raw, a bit granular sound far in the background.

Maybe PelPix can say if this is what he means.

Doing the same test with Black Grand samples or my keyboard's sampled piano sound, I don't hear such a texture. It's a bit hard to say in fact because my keyboard is a looped sample and playing the midi file to my GigaStudio Black Grand samples cuts the last note short...but I find a different tail to the sound, smoother and I hear only partials interacting in the decay.

I tried to minimize this partial interaction in the raw preset, using a flat temperament and no unison detuning to bring out the basic waveform.

My hypothesis, since PelPix hears it in all modeled pianos, is that it may be a side effect of producing a waveform with many limited resolution computations (even 64 bit floating point is limited)
Pianoteq's modeling is in final analysis a very, very, very, ... , VERY sophisticated form of additive synthesis where the final waveform is the result of a very high number of calculations.

Maybe this background texture is a result of a lot of very tiny errors accumulating, while a sample or a piano recording is one shot. No comparison also with a real instrument that doesn't go through amplification and produces a 3D soundfield directly by moving air.

By the way, in response to Jake, my soundcard can output 24 or 16 bits samples and I do hear a slight difference, 24 bits samples being slightly smoother by having finer steps. But since we usually output 16/44.1 wav or compressed mp3, this difference is only useful when listening live.

This textured effect has always been there I think, qualified maybe as the "pianoteq sound" by some, but the better the modeling of all the components of an instrument gets, the further this raw sound is pushed in the background, to the point where it doesn't bother me at all.

My hypothesis may be wrong of course, this is just a very limited experiment. But then Pianoteq's goal has always been real-time output and it does fine as such, so approximations are unavoidable in that case.

There it is!

Also, the second preset is a really good emulation of a CP-80, to my ears.  It sounds just like mine!

But Pianoteq sounds fantastic to me.

It is remarkable what they could achieve with the limited computational power of a pc.

I dont know the details of their model, but differential equations are often solved on a computer using time iteration methods like runga-kutta etc. All such methods introduce errors, especially if the time steps are too big.

I also hear some artificiality, particularly in the high notes, I think this could be down to the limited power of a pc.

But I have the highest regard for the Moddart team! Great work! Love your product!

And the performance of Pianoteq can only improve as computational power increases with Moore's law. This software would have been unimaginable on the 8 bit processors of the past!

Looks like we have been trolled, but expertly.

Bugger.

Looks like we have been trolled, but expertly.

Methinks you are right; I've re-read all the posts, and as Marcellus said to Horatio:  "something is rotten in the state of Denmark ".   (it is not Hamlet that says this in Shakespeare's play, "Hamlet".)

G

4.5 is indeed a huge improvement.  The only severe problem I can see appears to have to do with the resolution of the simulation.  Pianoteq sounds...granular.  It has since it first came out.  It sounds like the texture of a garage floor or a blackboard, but audibly.  It still had this problem even back when I first bought V2.0.  I find it so annoying that I haven't upgraded yet.

I will the moment they fix it.

Pianoteq has a higher pitch in its general sound than sampled software. It has more detail in its range. This sounds more comparable to a real piano to me. The sample sound always sound like their notes are shorter to me. Version 4.5 of Pianoteq has a nice straightness to the sound, which makes the notes sound cleaner and clearer. The 'Delay' setting in 'Effects' helps as well. The Pianos sound life-like too me.

I'm not hearing anything that sounds granular though. Maybe a parameter that tightens the strings could be added, as this might eliminate any noise that makes some unhappy with the product. Piano World forum comes to mind.

To Jake: I think a better solution is the FLAC format: it is able to deliver compressed but lossless audio files in 24/96 and can already be read by almost all PC's. CD & DVD's are almost non-existent for the young generation.

Can FLAC format files be played away from a computer, however--can a portable device or a car stereo play the file back? I was writing about DVD audio only because there are consumer level dvd players that can handle 24/96, so that if one were creating an album, the original recording would not be reduced in quality.

I remamber I had a speaker that get some weird vibration practically only with pianoteq YC5 add on in some few notes.
Strange how only YC5 making a metalic gride vibrate too much in the unhealth speaker.

By the way, in response to Jake, my soundcard can output 24 or 16 bits samples and I do hear a slight difference, 24 bits samples being slightly smoother by having finer steps. But since we usually output 16/44.1 wav or compressed mp3, this difference is only useful when listening live.

My hope is that DVD-Audio will catch on and that prices will fall. (And that iTunes will come to see that there is still an audience for  "hi-fi.") Here in the US, the large consumer store Best Buy sells DVD-A recordings for around $14-$30 US.

I don't know the percentage of current DVD players that support playing only audio files, however. And another problem is with car stereos--I imagine that no one wants to buy what looks like a CD, but will only play at home on something they associate with watching movies. I don't know if a DVD-A player even exists for cars. But let's hope for changes there. Maybe we have to create films as a background for the music?

Here is the fxp and test for U4.

http://www.forum-pianoteq.com/uploads.php?id=1685
http://www.forum-pianoteq.com/uploads.p...und_U4.mp3

This example shows the first 20 seconds or so with a normal U4 Tall preset, sounding quite good to me, letting the last note decay, and then the "raw" preset playing the same notes, and made to decay much longer.

If you listen again to the first portion, you may now notice more of this raw, a bit granular sound far in the background.

Maybe PelPix can say if this is what he means.

Doing the same test with Black Grand samples or my keyboard's sampled piano sound, I don't hear such a texture. It's a bit hard to say in fact because my keyboard is a looped sample and playing the midi file to my GigaStudio Black Grand samples cuts the last note short...but I find a different tail to the sound, smoother and I hear only partials interacting in the decay.

I tried to minimize this partial interaction in the raw preset, using a flat temperament and no unison detuning to bring out the basic waveform.

My hypothesis, since PelPix hears it in all modeled pianos, is that it may be a side effect of producing a waveform with many limited resolution computations (even 64 bit floating point is limited)
Pianoteq's modeling is in final analysis a very, very, very, ... , VERY sophisticated form of additive synthesis where the final waveform is the result of a very high number of calculations.

Maybe this background texture is a result of a lot of very tiny errors accumulating, while a sample or a piano recording is one shot. No comparison also with a real instrument that doesn't go through amplification and produces a 3D soundfield directly by moving air.

By the way, in response to Jake, my soundcard can output 24 or 16 bits samples and I do hear a slight difference, 24 bits samples being slightly smoother by having finer steps. But since we usually output 16/44.1 wav or compressed mp3, this difference is only useful when listening live.

This textured effect has always been there I think, qualified maybe as the "pianoteq sound" by some, but the better the modeling of all the components of an instrument gets, the further this raw sound is pushed in the background, to the point where it doesn't bother me at all.

My hypothesis may be wrong of course, this is just a very limited experiment. But then Pianoteq's goal has always been real-time output and it does fine as such, so approximations are unavoidable in that case.