Let's discuss Blüthner (Page 1 of 2)

During V4, i liked Bluthner very much, and was not convinced by D4.
With V5 D4 has made huge improvments and Bluthner.... not so much. The Bluthner has a "metalic" overtone that is too much "pianoteqish" for me.
Also, like always it's a bit of an illusion to think you can correct its defaults with the settings, only a new version can do that.

All in all the Blüthner piano still is my favorite Pianoteq instrument (and yes, I have started 'collecting' a few). In fact it was the reason I bought Pianoteq (that, and the flexibility of a modelling instrument). The Blüthner Model One is one of the most beautiful virtual instruments I know, and at least from my point of view quite without serious competition.

(Also let's not forget we are complaining on an extremely high level nowadays... and sampled instruments are neither more 'perfect' by definition nor is the approach without serious drawbacks.)

The problem with sampled libraries is that trying to make them 'perfect' is a losing game, actually. If you want responsiveness and low latency, you have to limit their size quite severely, which in turn limits detail quite drastically.

Some numbers: let's say we sample only with 44/16, which is physiologically as 'perfect' as it gets without unnecessary overhead (i.e., let's not start the 192/24 debate here), and we still get into trouble. We have 88 notes on a grand piano. The attack sample size has to be at least 30 seconds in the bass, 15 seconds in the treble notes before looping just to sound 'decent'. In other words that is still actually rather short (i.e. will lead to unnatural behaviour in very long chords, for example). But let's say we have attack samples of 20 seconds on average, and 10 velocity layers. This means 88x20x10=17600 seconds of audio, or almost 5 hours of high quality samples. Meaning... an uncompessed size of 3GB, or at least 300MB when quite heavy lossy compression is being used. And these are rather conservative estimates.

The problem with strong compression in memory is that it takes lots of time. Also most formats introduce quite high latencies all by themselves (e.g. MP3 and Vorbis about 100ms, FLAC still minimally around 5-10). On the other hand, you can't really expect to be able to hold 3GB of samples in memory. Reading them on demand from disk is also out of the question.

There are good reasons most 'hardware' pianos employ quite drastic measures to keep sample sizes reasonably small (and they can at least use ROM memory which incurs virtually no latency), like: lower sample quality, less layers, shorter attack sample sizes and such, which all lose lots of details. Then velocity morphing and DSP effects are used to mask that lack of detail. Even in the case of a very large sampled instrument with lots of layers, you still have to employ velocity morphing or you effectively end up with just 10-20 actual velocities, which is something people with good hearing will notice, and the larger your samples, the more work it is.

...and all these points still leave out the basic fact that certain details simply cannot be sampled properly: like sympathetic string resonances, for example. I.e. those are either actually 'modelled' even in a sampled library, or the result is just pathetic. That said, most current high-quality sample libraries are very cool indeed... but they are compromises, as every practical piece of software always is. And I would really be surprised if a sample library actually manages an effective latency of less than say, 50ms or so. Pianoteq's latency is certainly not zero, but I guess it might win that contest at least quite easily.

However that's what most software samplers are doing.
The software only loads in memory a few dozens kilobytes of the beginning of each sample to be able to play them while the hard drive is seeking for the rest. This works really well, and even better with a SSD.

With Kontakt Player and Vintage D, the asio driver is set to 2ms, but the actual latency reported by Kontakt is indeed 40ms+, which is probably one of the reason you feel less "connected" with software samplers.

Well remambered.

It's importante to be aware of each piano own characteristics before blame pianoteq.

It's strange... someones complained about bass.. other people about trebble, and other about midle range.

Everyone will compare pianoteq tone with the real piano models or even digital pianos they were used to play.

I really am into the Pianoteq Bluthner as well. The Bluthner Daily preset is my go-to piano.

Yet while we're talking about improvements, another un-piano like characteristic that I mentioned in another thread is a subtle phase-shifting effect that I hear when I play closed chords. With P4, I heard what someone else said was a whooshing sound when playing chords with the pedal down. This has largely disappeared with P5 but not entirely.

Outside of this, there is nothing else for me that strongly betrays the Bluthner as not a piano. I don't care if it sounds like a real Bluthner, but I do care that it sounds pleasing to the ear and real.

I thought I'd try out Galaxy's Vintage D and Synthogy's Ivory II:
http://www.try-sound.com/detail.asp/galaxy_vintage_d/en

Although these were played over the internet the connection was good and I could get the general sense of the software. I'd say the Ivory sounded better than the Galaxy, but both seems to have shorter notes than Pianoteq. The Galaxy sounded flat and a bit stifled. Ivory's sound was silky and smoother, but lack the depth of what I hear when ever I hear a real acoustic grand. Maybe the having the actual software loaded gives them more depth. When I hear acoustic grand pianos they all have bright and metallic quality in the sound.

Here are a few links, listen to the brightness of the Steinway in these tutorials:
https://www.youtube.com/watch?v=CC6bgBA8en8
https://www.youtube.com/watch?v=eMzdigll8jY

I don't know what these piano are, but they seem typical of the brightness and metallicness hear in real pianos:
https://www.youtube.com/watch?v=aKVLXDS4De0
https://www.youtube.com/watch?v=fAzwviInwoU

The Bluethner seems to need less tweaking, maybe this is because Modartt worked in collaboration with Bluethner to produce it. Where as the D4 v5 (Steinway) seems a good piano to create other sounds from. I tried tweaking the Bluethner, but always felt the pre-sets weren't really improved upon.

Its subjective.
Hear's an interesting talk:
https://www.youtube.com/watch?v=Un3p614XExc

Glad to hear, you tried this Steinways. I can recommend you to try Bosendorfer from Vienna Symphonic library too. http://www.try-sound.com/detail.asp/vienna_imperial/en

I think this examples are not good, as piano sound depends not only on instrument, but largely on player (though some bad players deny it)). Unfortunately we are deprived of opportunity to control our touch in all the aspects, when playing digital piano. And in this case we depend on software and its presets. Can them afford us woody sound, or not? Especially in classical music, which "love" woody sound.
Another example of Steinway D Hamburg, played by great pianist (his sound was often named "golden", after Genrich Neuhaus characteristic):
http://www.youtube.com/watch?v=BZYuHFYQLCQ
And you can find a lot of another classical examples of round and wooden Steinway's sound. Though, I understand, that other genres (f.e. jazz) enjoy brightness

What you say about the physical sound of the piano, that is, as measured as opposed to heard, is of course correct.  But those who assert the opposite, while they seem to be arguing for something magical, are also pointing, inexactly, to something real.  Sound as we hear it as human beings is is a complex perceptual phenomenon, and the control of touch and other parameters, for combinations of piano sounds, can have significant changes on sounds as we perceive them.  Pianists often talk about this as the creation of illusion; but "illusion" suggests that there is a kind of hearing available to us that is free from it, so that isn't quite right.  It is, however, a way of hacking our perceptual apparatus to achieve certain combinatorial effects -- for instance, the use of timing and contrast to call attention to different phases of the piano sound's decay.  When a pianist says that s/he makes the sound, not the piano, this is the skill that is meant, and it is really true, if by sound you mean, not the measurements made by simple unbiased non-biological instruments, but those made by our own highly biased faculties.  All the pianist really knows is that they have a certain sound in mind, and they make some small adjustments so that it comes back at them, more or less.  There is no way in practice to fully know what those adjustments are; they are deeply internalized.

I don't think this is very different than saying that a great painter can create a sense of color and light that is unique and personal, even though all the paint used came out of factory-made tubes.

Its true that there are a variety of standards, from absolute beginners to performing a piano concerto at the highest level. I wasn't really addressing someone's playing ability, I was simply saying what I generally hear in grand pianos. The clips showed what I was getting at. The range that they have, from a low percussive sound to an ear piercing shrill, from soft playing to the attack and every thing in between. The piano has the capability of being all these.
Thanks for the video clip . The recording was a bit muffled and most probably cut out some of the tonal range of the performance, but I could still hear a bright sound in that piano. I would imagine that the pianist in the clip had it tuned to the concert he was going to perform. I have different headphones (six) with varying frequencies. Some of them omit sounds in Pianoteq that are present in others.
When I tried the Galaxy and Ivory I thought they both sounded good. If you owned either of them and just played them at home you should be satisfied with them. Its only when you start to compare them with the detailed sound that acoustic pianos have, then you'll be disappointed, or not quite satisfied.
The piano software developers have all done a tremendous job in allowing us to have a grand piano in are houses. I think Pianoteq produces sound and depth very close to what real pianos produce. Though there's always room for improvement. I'm sure healthy and constructive criticism will always be welcomed.

A Steinway D Hamburg?

Hello, kalessin! Sorry, if I was too harsh in my words: sometimes it's just no time to use more smiles or to write IMHO one more time

Glad to get acquainted with you! I am not physician, but musician and (my second education) psycho-physiologist (the perception of music is my theme). As for the piano teacher, meeting different pupils on one instrument in one day - it's not surprising for me that one piano can have different timbre. I hear it every day. Moreover, this effect is preserved even with one pupil - when his arms are at the begining and after a year, when we made a lot of work about sound and his arms.
And good pianist can show, how many paints and variants of timbre can be extracted from one instrument. Of course, we can't strongly change the sound and of course we largely depend on piano we have: it's sound, it's quality, it's opportunities (as one have great variety of paints in itself, and one is blankly). But though our influence is not too big, sound can be changed.

Magic? Placebo? No, I don't think so. I am not physicist, as I said, but understand that sound is complex and can be achieved by different ways: by the speed of an arm, by the deepness of touch, and more complex - we say "by the weight of an arm". And all this become a hammer characteristics. Combining this parametres in different ways, we achieve changes in sound. In my opinion, understanding and hearing of this effect depends on musical experience: how many piano music you've heard (jazz, academical, pop and etc. Different pianists, piano models), how many acoustic pianos you've played and how long, what did you played (improvising a real piano - good way to find it). Though, I am convinced, there are still good pianists, who will not share my opinion. And it's normal, and i'am glad that you think (and has your own evidence) different.

Small addition: hitting the strings by an arm is not what only jazz players do - yes, academical composers use it too, and for more then century

Oh, wasn't it Steinway D Hamburg?
My sorry for you too) Of course examples were not bad, I also just wanted to show, that in different genres sound standard differs. And we, who play classical music, sometimes become mad about woodness))))

But guys, let's discuss Bluthner!

I half remember an anecdote about Cortot.  Apparently a pupil was appalled at the state of Cortot's piano (a Pleyel, I think) and could not get anything pleasing out of it.  When Cortot sat down to demonstrate a passage, the pupil was amazed at the beautiful sounds coming out of it. Now, Cortot was an exceptional pianist and he was very accustomed to his own piano, but it does indicate that there is more to creating a beautiful sound than appears possible from the mechanics.  When I come to think of it, the piano is technically a percussive instrument and it can of course be played as one, but composers from JSB onwards have written music for fortepiano/piano requiring a singing tone and the voice is not a percussive instrument.  So we are all in some way creating an illusion, right from our first lesson.

Interesting,

The problem is the gap between what pianists say or feel and the fact that the piano is indeed a mechanical instrument.
My piano's teacher speeks about colors, tones... I don't understand all what he says but I think he is probably right: there must be a lot more than only the mechanical aspects. But all these subtle nuances represent maybe 5% of the piano sound, and require a high level to be expressed, IMO... I don't know...

The fact is that I can't imagine in what a subtle nuance with my finger (other than speed or strenght) could result in any effect on the hammer

Hm. You mean, they recognised the piece, and associated that with Mr Johansson. I very much doubt that most 'people on the street' can do more than that, as actual musicians are a minority. And if they really could recognise at least 3 different players, then there of course is the whole rhythm and dynamics thing that no one really doubted is an important factor. In other words: I would find it quite believable that a pianist has a certain acoustic 'thumbprint' that is recognisable... but it probably is recognisable completely independent of the actual piano. And since the recent discussion was started over the question whether the pianist can actually modify the sound of the instrument just by hitting the keys differently, we are talking about different things.

I uploaded a little audio with Avdeeva playing Brahms on Youtube. 2 Fantasien. One of them was in the beginning of this topic before.
Intrested in your comments about sound and realness/unrealness of Bluthner. Did the sound of original Bluthner modell become better after modifications, or not? I tried to reach atmospere of classical recording by tweaking PT and using different plug-s in Cubase. Do you like this sound, or it seems not real/"tasty" enough? All opinions are appreciated!

http://www.youtube.com/watch?v=OJV7yUxxp-Y

Kridlatec,

Your recording sounds very good. There's a sense of formality--the piano is kept slightly away from us--but it still has force, and the piano is entirely convincing. I can't imagine anyone hearing anything other than a real piano.

Just wanted to chime in on the debate as to whether a player can change the timbre of a note via any means other than velocity. This comes up regularly on the piano teachers' facebook groups (I fall firmly in the "No" camp), and is always in relation to the "digital vs acoustic" argument.

The main point I want to make is that no one (in the "Yes" camp) seems to take into account the fact that skilled players can also make any digital piano sound much better than a lesser player can, and that this is largely a matter of technique.

I think that players who think they can influence the timbre of a single note independently of velocity are being misled by the way they listen to the sound. What I have seen is that these players have learned to listen for the sweet-spots in the piano's timbre, not by how loud the sound is, but by how the harmonics bloom. This makes good sense, as it encourages the player to judge their dynamics by quality of tone. I suspect that the way this is perceived from the resonance of a real acoustic grand is mostly impossible to recreate with amp+speakers, and as such, the player's expectations will not be met.

Another thing I have seen is that some teachers with this anti-digital attitude encourage their students to set the volume of their digital pianos to full, as they think that this more closely matches the sound of a real grand. I'm sure most of us would agree that this is very likely to end up making for a very unpleasant experience for home players, as the highest settings are only really necessary for playing in larger rooms, such as school halls. This, again, suggests that players who are privileged enough to teach/play on grands have not had the opportunity or inclination to spend sufficient time with a digital for their expectations to adjust accordingly.

Yes, Clive, I think it does this very well - but would still require such a player to adjust, as the effect of an entire grand piano, resonating as one entity (but also interacting with the room acoustics), cannot be effectively reproduced by a couple of cones and tweeters, IMO. The other problem - that of the sheer volume of sound that a grand creates - is also difficult to address for the same reasons as I outlined above, namely that to achieve that sense of immersion from speakers, you will likely end up with them being uncomfortably loud.

In last patch 5.1.3 Bluthner was revoiced. I think, attack became more clear. And shorter, more precise notes in treble range. For me Bluthner model became better once more. What do you think, guys, do you hear changes?)

You don't play with your fingers, you play with your mind.  It's the brain that counts, not simply the fingers.  The brain controls the movement of the fingers, which in turn make the instrument "sing."

Greetings!
I took the opportunity of the black Friday discount and upgraded from standard to Pro. I also updated Pianoteq from v5.1.2 to v5.1.3...
I brought Pianoteq's internal frequency to match my external soundcard freq (96KHz). Then I played the Blüthner and I was blown away for the difference in sound. For me it was as if I installed a new piano. I wasn't expecting that!
Now, I'll have to make more experiences to find out where differences resides, probably this weekend I'll have more answers. Was the Pro sound engine? the matching of the frequencies? the modifications introduced to v5.1.3? 
Did anybody else just updated and perceived the same (big) difference in sound?
Regards.

Yes, I think that was the update to v5.1.3. I already noticed from other posts that the sound engine is the same for all Pianoteq versions. And since I´m nearly 50 years old, I know that my ears don't have the sensibility to notice the differences in the upper freqs.

Funny the human mind, I really do feel differences (using headphones) when using different internal sampling rates... I know it is subjective but I do notice a difference (even in a blind test)...

Anyhow, for me, Pianoteq's Blüthner is still my go to piano and it continues to get better and better!

Kudos to Modartt!!!!

As mentioned above, when you have twice or four times as much samples in the same amount of time, you can represent very short transient changes (like attack of a drum hit, or the noise when a hammer hits the string) that much more precisely.

I guess, two reasons are right. I know, I heard few times from other users, that changing frequences doesn't change the sound. I definetely can't agree. I don't know, if soundcard/drivers/musical pitch or individual anatomy influence this process, but I hear great (great) difference between 41/44 and 96. In my case it's so. And no one will dissuade me
And yes, patch 5.1.3 (I wrote above about it) made Bluthner sound more strict and shaped. In my opinion, attack became better.