Reply by dhalfen

  • Avatar
  • dhalfen
  • Member
  • Offline
  • From: Baltimore, MD, USA
  • Registered: 25-05-2008
  • Posts: 469

Re: Sample pianos - R.I.P.?

katie wrote:

The fact is for whatever reason, physical modeling is not there yet.  Period.

Oh, my oh my, where to begin...

Well, in terms of capturing the sound of specific pianos _exactly_, physical modelling would very likely "fall short," unless the attempt is to model the samples.

But in terms of creating an actual piano _instrument_, and being _playable_ as such, physical modelling has proven to me, in the form of Pianoteq, to be the superior method, with superior results.  The interaction of the various modules involved in creating the Pianoteq sound (and the various instrument models which underly this sound) proves to be the epitome of the phrase, "The whole is greater than the sum of its parts."

The sum of samples, on the other hand, is the sum of samples.

As with an acoustic piano, I can adjust my own performance to the instrument in Pianoteq.  (But, unlike a "real" OR "sampled" piano, so many additional details can be manipulated to better effect in Pianoteq.)  With the sample libraries, I do _not_ have this freedom.  Many players' solutions would be to throw effects and mastering tools at the sound, but I find that idea to be _deeply_ unsatisfying.  (Though I will admit that I like oddball compression for "artistic" purposes, and I'll throw overdrive/distortion in the mix for much the same reason.  >:^)

katie wrote:

So yes, the high end piano samplers are superior, whether some of us want to accept it or not. And the libraries are gettiing bigger all the time to the point that some of them are 70GB or more for just ONE piano, not a collection of pianos.

Which is a laughable trend.  Soon, you'll have 1TB libraries consisting of super-high-resolution samples of every imaginable velocity layer, at various levels of pedalling.  ZOMFG.  Where does the madness end?  It doesn't, because the thinking behind sampling is all about REproduction, not source production.  You can increase the pixels in an image to the millionth degree, but the 2-D picture that results will still _never_ be a painting on canvas (with all of its attendant textures and subtleties).  And an audio recording will _never_ wholly replicate a live performance.

My experience thus far:  I _know_ that I forget I'm "at the computer" when I'm playing Pianoteq.

"Our developers, who art in Toulouse, hallowed be thy physical-models.
Thy version 4 come, thy new instruments be done, in the computer as it is in the wood!"

Reply by Ecaroh

  • Avatar
  • Ecaroh
  • Member
  • Offline
  • From: Finland
  • Registered: 21-06-2009
  • Posts: 459

Re: Sample pianos - R.I.P.?

... And what makes me still wonder is this: How those traditional manufacturers (Roland, Yamaha, Clavia) can make their quite good piano sounds with their ridiculously small memory capacity (compared to the software libraries)?

Take for example Clavia Nord's piano library: their biggest pianos are about size of 64MB. (That doesn't tell the whole story - they have some kind packing and algorithms to make it better. Does somebody know more about that?) But for example Clavia's pianos  - Yamaha C7 and Steinway D and ESPECIALLY their Rhodes e-pianos - sound very good and they are even better than Giga level Ivory in some respects. It's a pity that Clavia doesn't have better velocity, dynamic range, EQ etc. adjustments to make their pianos more usable.

So in my opinion "size doesn't matter" that much when we speak about sample pianos. It's the overall quality of the whole package. Example from Clavia shows (at least to me) that with quite small memory capacity you can have a good piano and vica versa: big size doesn't prove anything.

P.S: Those who aren't familiar with clavia's piano sounds can check it here:

http://www.nordkeyboards.com/main.asp?t...nformation

Ecaroh's Website

Reply by Piet De Ridder

  • Avatar
  • Piet De Ridder
  • Member
  • Offline
  • Registered: 07-05-2007
  • Posts: 927

Re: Sample pianos - R.I.P.?

I'm sorry but I don't quite agree with some of the previous.

For most people - highly discerning musicians and ignorant laymen alike - the sound of sampled piano's (still) does provide the most familiar-sounding results and immediate satisfaction. We may deplore that fact, but it's a fact nonetheless. It's all very well, here in our cozy and ever affirmative cocoon which the Pianoteq forum is, to look down on sampled instruments and to congratulate one another on our refined and elevated taste in pianosounds, but go out into the big world (I mean: visit *any* serious music-related forum) and ask a thousand people what their preferred virtual pianosound is and I bet you that over 95% will go for a sampled instrument.

Are all these people wrong? I don't think so. And suggesting that they are, is rather arrogant and foolish, it seems to me. You see, for many of the situations where a virtual piano is desired or required, the 'snapshot realism' of sampled piano's is indeed the safest, least complicated and, yes, most earpleasing solution. These instruments may lack a lot of the sonic sophistication and musical 'immersability' that Pianoteq has to offer, but what they offer instead is a simple, one-dimensional, convenient and non-negotiable believability. And for many people, that's all they need: a simple tool that allows them to easily create the illusion - no matter how shallow - of a real piano.

We can talk all we want about the amazing subtleties and mindboggling possibilities of Pianoteq, but for many people, all of that is totally irrelevant, I'm affraid. And it gets even worse: the average DAW-based musician isn't even particularly interested in Pianoteq's unique power and qualities. No, they simply want to press (or program) a few keys and hear a convincing pianosound coming out of their monitors. Beginning and end of story, as far as they are concerned.

In order to fully appreciate the uniqueness of Pianoteq, it seems to me, one also has to be sensitive to what Pianoteq offers BEYOND mere emulated sound: the joy of 'connecting' with an instrument, the pleasure of being able to dig inside its sound, the inspiration provided by the organic responsiveness of its timbre, ....  Not everyone is interested, capable or prepared to take that extra step. (And we shouldn't blame them for it.) The fact that playing Pianoteq is often a far more musically satisfying experience than playing a sampled instrument is of no consequence to many people. All they wanna know is: can I believe its sound? A question that is, paradoxically, both extremely superficial as well as completely to the point.

_

Last edited by Piet De Ridder (19-12-2009 15:01)

Reply by Ecaroh

  • Avatar
  • Ecaroh
  • Member
  • Offline
  • From: Finland
  • Registered: 21-06-2009
  • Posts: 459

Re: Sample pianos - R.I.P.?

Very true indeed.

And if we go little more abstract the whole conversation is quite meaningless if we don't specify the different situations and needs for a digital piano. Just to name few:

1. Live gig
2. Recording studio
3. Practicing (or making fun) at home
4. All the musical genres

Personally I might prefer good old hardware piano for 1, Ivory for 2, PTQ or acoustic piano for 3 and so on. In other words there is no a single solution for me for all kinds of situations and therefore I cannot say "X the best". Sometimes I can say "X is the best in situation Z" but even this statement is not constant: sometimes X feels better than Y and vica versa. At moment best thing to me is to have three or four different options.

Last edited by Ecaroh (19-12-2009 16:00)

Ecaroh's Website

Reply by Glenn NK

  • Avatar
  • Glenn NK
  • Member
  • Offline
  • From: Victoria, BC
  • Registered: 09-10-2008
  • Posts: 1,008

Re: Sample pianos - R.I.P.?

Piet De Ridder wrote:

We can talk all we want about the amazing subtleties and mind boggling possibilities of Pianoteq, but for many people, all of that is totally irrelevant, I'm afraid.

And it gets even worse: the average DAW-based musician isn't even particularly interested in Pianoteq's unique power and qualities. No, they simply want to press (or program) a few keys and hear a convincing piano sound coming out of their monitors. Beginning and end of story, as far as they are concerned.
_

Good points - but why are they so easily satisfied?

Having spent many hours on another music forum that is specific to one software, but certainly not instrument specific (as this one mostly is), I found that many non-pianists don't recognize or want a good piano sound, but as you stated, want the sound to "fit in the mix".

Never having played anything but a piano, and being a strict soloist, "fitting in the mix" is terribly important.  I want a big, shiny, black nine foot European (or Shigeru Kawai) grand that responds like a real big, shiny, black nine foot grand.

Every acoustic piano I've played has had it's own unique characteristic sound, but for the most part, the good ones all have a similar enough touch to be acceptable.  And after I've played any piano for a while, I can work with the sound.  But it must respond properly.  This is the key element in Pianoteq.  The sound is important, but if it didn't respond correctly, it would be worthless.  To the "average DAW user" playability and responsiveness are meaningless.


Second rambling:

So many of us (including me) have been "dogs chasing our tails", working feverishly to get the "perfect" sound from Pianoteq.  Good grief, there is no such thing - even the big, black, shiny, nine foot grands all have their strengths and weaknesses (and they may not "fit in the mix" either).

What I'm slowly realizing about Pianoteq, is that its greatest weakness may be the ability to tailor the sound.  This permits and tempts me to try endless combinations of impedance, hammer hardness, strike point, detuning, and on and on.  The result is that I'm distracted from learning to adapt my playing to the piano I'm sitting at.  This is extremely counter-productive.

Glenn

__________________________
Procrastination Week has been postponed.  Again.