The Clarity Challenge

Agree with Beto-Music about the Model B.

Also where you're at at your digital piano controller is usually not an ideal place to listen to sound from speakers. Typically keyboards are set up against a wall. A lot of the "muddiness" of the bottom end has to do with your ears being too close to one or more walls of the room. You'll get low-frequency phase enhancement and cancellation effects there.

Good sound system (amp + monitors), and well set up acoustic space (placement of speakers and listener, possibly anti-reflection surfaces on the walls esp. on those behind and close to speakers) can help a great deal. Unfortunately, usually such a setup requires a fair bit of room and otherwise somewhat inefficient use of space.

Nevertheless IMHO, all else being equal, in a less than ideal acoustic environment (very small space) the new model B sounds the clearest of all the acoustic pianos Pianoteq has to offer so far.

I prefer not to use headphones and, more importantly, I don't have good quality ones (have a fairly expensive noise cancelling Panasonic pair, but sound quality is only mediocre). I agree that that would be a good way to go for making comparisons. Something I should invest in at some point.

I considered the issue you raise, but did not want my post to go on for too long. The OP will surely already know if that's the case. I recall a recent thread about Pianoteq sounding much better when listening to pre-recorded tracks than when playing. I expect that's a fairly common problem. There have been some interesting suggestions, including placing speakers on the floor at an angle (like foldback monitors).

What you say about A vs B comparison being largely equipment/environment-independent is true up to a point. But if the equipment/setup is too much of a weak link (like my headphones or PC speakers or e.g. listener standing next to a wall at a live music venue), it's no longer reliable for comparisons.

When playing a DP-Pianoteq I have some kind of idea in my mind that I don't actually play a real instrument but actually "playing a record". Though at the same all this ideas just vanishes and you just play music. Need to say I have an extremely modest consumer grade speakers/headphones which I don't blame ever.

You may want to check this: http://thesubpac.com/

I think I found the right timbre: the Model B with binaural perspective and with less mids (using timbral EQ) is quite impressive, thank you! I think I will soon buy it, but before I will experiment with the timbral EQ with other presets, like the Bechstein: I never tried it and it is quite revelatory. Lowering the mids fixes the nasal timbre that I hate, and slightly raising the bass improves a lot the lower notes of the Bechstein.
There is still an element missing: what I call the raindrop effect. If you listen to the high notes in the video, they are thick but not muddy, they sound like raindrops, and I'm really struggling to obtain this effect.

Yes, I know that, but I don't like so much KIVIR's Bechstein, while the Kremsegg Bechstein is probably the most interesting piano in all Pianoteq, and I aim to make it the perfect instrument for me. For now, the problems are the lack of the raindrop effect that I mentioned before and the too weak bass. Lowering the mids in the "effects" EQ does improve, but if I raise too much the bass it becomes muddy, and muddiness kills counterpoint, and I mainly play Bach! The K1 doesn't meet my taste, while the K2 could be one of the best if not THE best piano in Pianoteq, but it sounds too electric and has this annoying overtones and nasal timbre that makes it "uncomfortable" for me, because I have extremely sensitive ears: I litterally have an headache if there are overtones that I don't like or the sound is too high, so it's quite a pain to "tune" every instrument. I find that I dislike a lot a "perfecly tuned" instrument: raising the conditioner slider to 70 and doing a little random detuning makes everything very natural and easy to listen, and it also improve counterpoint clarity, because every single key feels more individual and unique. I also play in Werkmeister III temperament, even using non-historical temperaments, and set the diapason to 436 hz for modern instrument, 430 or 415hz for pre 1830 ones. And the fact that you can actually choose different temperament and tune your instrument as you like makes Pianoteq the best VST out there.
I'm sorry, but I really can't spend 200€ for speakers now: I use a pair of Symphonized NGR wood, which are really cheap earbuds (I don't like big headphones, they just don't feel "right" to me) that cost me 30€ on Amazon, but they feel a lot better than a lot of 100+€ earbuds, at least to me. Perhaps is because of the wood, but I really love them!

All vintage models of pianoteq, at the time, sounded a bit better, in terms of naturality, than the modern pianos released in the same time.  The Kremsegg models sounds quite natural. I imagine modern pianos have some ellement that makes modelling just a bit more complicated. But the Bechstein 1899 have crossed strings and apparently a modern harp, anyway sounds great.

Be sure to choose the right velocity curve, to ensure for example that the mf you play it's really a mf that will be rendered by the software.
For exemple the controller Kawai VPC1 have a sensibility that needs different adjusts to get the best from different piano softwares:

What a coincidence, I was watching the comparison of pianoteq model D with a real Steinway D, from 15 months ago, and the despite of almost perfect sound, one detail that was missing was something that I imagine sounded like a water drop sound near the attack, perceptible in Chopin Nocturne op.9 N.2 . 
We both imagine the same metaphor.

https://www.youtube.com/watch?v=2l4aAFD-moY

If I remember well the model D got some few refinements after this vídeo comparison was made.

Keep in touch with the forum. You know how regognize the good points of pianoteq, and also have good ears and perfectionismo to see the few points where it can/need still improve.

It seems to me that with the sampled instruments it's as if, however you bash away, it never sounds too bad. Maybe it's to do with the compression inherent in the sampling process. I don't know. Sampling is a dark art. With Pianoteq the connection between what you do at the keyboard and what comes out is much more apparent - more like an acoustic piano but, then again, not quite the same. Except that no two acoustic pianos are the same either of course. You have to learn to play each and every one, just as it seems to me that you have to learn to play the different pianos in Pianoteq. When I first started a couple of weeks ago I was making the most dreadful racket but I'm starting to get them to sing now.

Yes, there's something going on here isn't there? I noticed that Synthogy's demo policy is to provide mp3s of people playing Ivory II. I didn't even bother listening to any of them because I know from experience that it will tell me nothing about what it's like to play the sample sets. And I'm starting to give up hope anyway that any sampled instrument will offer the sort of connection that I refer to above.

Personally I don't like using headphones - even really good ones - with digital pianos, not though because of what I'd describe as the loss of 'physical impact' (how loud do you play?) but because you cease somehow to be in the room. I wonder if it's got something to do with the way in which the sound of the instrument changes as you move (your head) - or something like that. Whatever it is, it seems very unnatural to me.

But 'real speakers in real rooms' are indeed a challenge. Sadly, I suspect that the truth is that you need very good speakers in a very good room and it might just be cheaper and take up no more space to go out and buy a reasonably new second-hand Steinway Model B or even D, or whatever else tickles your fancy. (Although I'd concede that if you want the range of large grands that Pianoteq offers it is going to get seriously expensive and space-consuming.)

For the digital version, anything short of something like a pair of ATC SCM25As (my personal favourite) in a properly acoustically treated environment simply isn't going to do something like Pianoteq justice. I'd suggest that problems with 'boxiness' and 'unpleasant overtones' are much more likely to be due to your equipment and environment than to Pianoteq.

One interesting thing to try, if it's possible for you, is to take your gear outside. I've done this a number of times in various circumstances just to get more of an idea what my room is doing to the sound. The answer is always that it's doing 'a lot', and what it's doing is unwelcome. But, here in the UK at least, it's too wet and cold I'm afraid to become an 'al fresco' pianist.