https://www.youtube.com/watch?v=i-jQsje3_3c

I hope someone donates €50,000 (~100 days of studio time) to Modartt so they would have enough time to build the sample library needed for the production of this model.

But PT pianos are not free of annoying defects either... it's just that those defects are not AS annoying because you can edit 95% of them out... something you will never be able to do with an end user purchase of samples.  Some of the pianos like the Bluethner can be modified and edited nearly perfect, others like the D4 and K2 will take much more work... but if you're willing to do the work you can edit them into gorgeous magnificent instruments that any audience, any studio musician or engineer will immediately recognize.

I can actually understand that. I haven't been satisfied with how PTQ's models sound from the day one. I own Stage version, and my current instrument is K2 model, and it took days of trying and error from me to achieve something I would call "relatively rich and clean piano sound". I had only a limited set of tools, so had to do everything by combining some ready preset with reverb and delay effects from equalizer. I didn't like how all ready presets sound out-of-the-box, they were either "plastic" and devoid of overtones/reverberation, or too "muddy" with them. May be it because I play in headphones all the time - I've got impression that PTQ is deliberately provides presets which lack reverberation component because Modart expects you'll play it back through some speakers and your room and sound system will complete the generated sound.

I would really love to hear your thoughts on the subject, about how to properly adjust models to sound more naturally and realistic, wdco. Even if it'll assume you must have Standard version to achieve that.

I also think there is a certain marketing benefit for Modart in it. Shigeru is definitely an excellent, unique and full of character hand-made instrument. But there are 0 opportunities at it on the market of VSTis. You won't find even a single sampled library of it, available for wide audience. There is only samples of unknown quality developed by Nord to be used in their own digital pianos, i.e. you can't use it unless you own their gear which supports it. Adding something really unique, and at the same time of such high quality, to their product would benefit Modart significantly.

click here: http://www.llpianos.co.uk/pianos-for-sale.html

There are a few end user sample sets that could provide source data without introducing the unnecessary flaws, defects, and sampling errors contained in most products... into the model.  But then again, we are seeing some of these defects in the PT models/patches themselves... hence their interest in finding edited fxp's that might be possibly used to update/reset the model. 

And their models are getting better.  The Bluethner ...harmonically speaking, requires far less tweeking than the Steinway which requires extensive editing and still you can't get all the problems out.  And the big problem with that much editing ...is that often you end up chasing your tail... much like what happens when you try to rescue a mix and after hours and hours of editing, you oversaturate yourself to the point of returning to the mix a few days later only to find it still isn't as right as you thought it was. 

It is remarkable how close the PT model is to the Steinway D I sampled in the 90's right down to some interesting details and this highlights the importance of starting with a source piano that is well prepared, well tuned (with someone who can tune right there while the samples are being taken), and with a mic setup that minimizes artifacts created by mic error. 

We tried averaging away the harmonic errors but it's a delicate process... averaging too much obliterates the "character" of the piano.  And we found that creating a harmonically perfect piano was actually less appealing in most cases than leaving the defects.  But both extremes were objectionable and so it's the delicate process of figuring out just how much to edit so you don't take it over the edge.

I would like to see more piano variety as well and fewer old clunky pianos modeled ...but it looks like these clunkers were probably an early on necessity to develop what the modeling software needed to address.  These pianos can be useful for cartoon tracks (not kidding) and documentaries like the Ken Burns Civil War series where clunky pianos add to the sense of time in which the documented events occurred.  So for film scoring, arranging and production, these old models can be a useful addition to your library but for today's live performance and music productions the peak demand is for an unclunky gorgeous piano.

We'll see what PT comes up with next... right now, I've located a well signatured Baldwin Concert Grand in a large warehouse... a tempting prospect.

Well, I 've posted items on this wish list here and there before. but since this thread is dedicated to new piano models:

1. A Baldwin grand. Don't know the models well enough to suggest one. The piano of Stravinsky, Bartok, Leonard Bernstein, Ray Charles, and Marian MacPartland. Sweet clear tenor and middle good for popular music and more.
2. A Steingraeber 212
3. More recently, an English make that I had not heard until recently: A Chappell, as in: https://www.youtube.com/watch?v=JAKPj5F7wBc. The resonances and upper partials sound are nicely in tune with the lower partials. Gorgeous old sound. Don't know if there is a bit of extra reverb added to that video.
4. A pre-war Mason and Hamlin.
5. A Steinway B. I'm not convinced that adjusting the string length on the D4 gets us to this sound, yet.
5. And specific pianos: I doubt that Electric Lady in New York would ever allow access to the Yamaha used by Norah Jones and others, but this is a direction that I hope Pianoteq will move towards--modelling specific pianos that are known from recordings. Doing so would raise the bar a bit--how closely can a Pianoteq piano come to sounding like a specific recording? A little terrifying, too, but it's already very close with the Bechstein when listening to the "Let It Be" demo.

I'm a new PTQ user, and still testing the D4 and K2 models and presets, with great pleasure. As alternative pianos, like Jake Johnson, I very much like the real (New York) Steinway B and the Baldwin grand (the 7-foot model is the SF10) of approximately the same string scale. Alas the SF10 has not been in production for a long time, but it shares with the B a liquid yet clear sound throughout the midrange, but has a generally stronger bass than the B. Of course all of these pianos, the Steinway D included, are made individual by voicing and tuning, so we need to compare particular individuals of each of these pianos. However, the particular D's I have experienced seemed most suited to large halls, because they project a metallic brilliance that overwhelms smaller spaces. This is just my experience, of course, but I mention it because I was very happy to discover how close the PTQ K2 model comes to both the B and the SF10 7-foot pianos. I think K2 could be modeled into either one with some tweaks.

I'm not sure exactly how PT models are created but I would guess that they derive a model from a set(s) of samples... making some interpolations and assumptions, of course.  But to answer your question as to how to get individual or individualized pianos... right now PT provides the ability to edit the given models and by far it's better than dealing with sample sets like the Vienna Imperial in which the flaws users complain about are not fixable.

To answer the second part of your question ...if you're thinking of "how can we come up with our own models... say, from a sample set" ...the answer is that I don't know of any end user modeling software that can do that but that doesn't mean that something might not be in the works.  Sooner or later, someone will have a program where you can sample a handful of notes on any piano, and apply that to "a" piano model, which then will apply the control data you supply to the model.

Years ago my physics buddy from college and I had some success creating math models of piano spectra.  We experimented with imposing one set of spectral parameters on another one derived from a statistical analysis of Piano A and we imposed that on Piano B.  And we even tried to mathematically define the character of a piano but it was all preliminary and in the meantime I was designing piano sample sets in my business.

If you are meaning that you're looking for a piano that's different from the in-the-box PT patches... that's possible with the editing tools in PT.  I basically turned the Bluethner into something more like a Steinway in some respects while still maintaining some key Bluethner character.  I didn't set out to do this.  But it was the result of just trying to clean up the Bluethner for live play.

When I went to do the same thing to the D4, it was three times the work and yet stayed a "Steinway" despite the makeover.  The K2 makeover made one of the best stage pianos I have and very fun to romp around on... but it too was very difficult to get right as it looks like the developers merged some Steinway modeling along with some thinner Yamaha model material.

Not something likely to happen outside of  Modartt's facility )-;
Unless the math appears in the public domain, which is likely given the academic involvement - though the actual CODE remains proprietary.
In any case WAY beyond me.

The thread on PtqSpecProf a while ago suggested (to me) that one could basically subtract one piano sound (Call it pianoteqA) from another (Call it recordedpianoB) to get the difference, add that difference to PianoteqA and the result will be a new PianoteqB, which will be very close to recordedpianoB.
Grossly over simplified perhaps

Not very close, but closer...The spectrum differences are computed at a single velocity, and as you may know, piano spectrum varies continuously. This software helps in bringing the source spectrum closer to a desired goal but must be used also with some spectrum profile scaling and maybe removing bad notes and such. It is one tool, the other, everybody has, is a good pair of ears

The pro version gives to the user much more control and flexibility for modifying a given instrument, but the basic sound of the instrument cannot be completely changed, it is the result of many, many parameters that are unavailable to the end user. As an example, try to go from the Blüthner to a Vibraphone: impossible, but still the same pianoteq engine is at work. When the distance is less, for example going from the Blüthner to a Bösendorfer, some emulation can be achieved, but it is far from what a real Bösendorfer would be if Modartt decided to model one;

As an example, here is one done with PtqSpecProf from a while back (with Pianoteq 4):
http://www.forum-pianoteq.com/uploads.php?id=1642

If this is roughly how the development process goes it would SEEM that very close approximations to ANY renowned piano from a particular studio or concert hall could be developed fairly quickly at some reasonable cost.
It could probably be done from a sample library, perhaps even from a live performance - although separating out individual notes may take a LOT of very tedious effort.

I think my question is "HOW can we get individual piano simulations, are the tools available to owners of ptq-pro ?"

When I edit, I decide if the piano will be "special use" such as a track piano or do I need it for live/onstage play and general workability both in the studio and onstage.  If it's on stage, then step one is to find a patch that is closer to what I think I want than any of the other patches.  Part of that decision making process includes eliminating all the effects (even level and delay compensation and the mono feeds of the mics into the opposite channel).  The idea here is to get as close to the raw piano as possible by eliminating variables that have the potential to add more harmonic distortion than is really there.  Since you're looking to eliminate ambience for live pianos onstage... you want to move the mics in close.  The problem with close mic'ing is that the harmonic content of the piano becomes more erratic the closer you get.  This is why so many sample sets early on... didn't fit into mixes and it was because they were sampled with mics further away from the piano which brought in some ambience.  So played solo they sounded cool... and the harmonic content was generally more consistent.  But when mixed in context, they fell apart and kept bumping something else out of the mix whenever you needed to place the piano upfront. 

In all this virtual modeling, I am amazed at two things... how nicely suited most of the effects are for the in the box patches and secondly, that I never use them.  For studio, I will use the Lexicon PCM or will have a piano edited to reduce stereo width and thin out the bass a little so as to place it behind other tracks.

Not sure about posting fxp files... I will think about it.