Pianoteq - Looks Like new Modelled Piano

[aMUSEd]

Just the CPU optimisations alone are really impressive - Thinking about it, to model a piano properly you would need to model each string with mathematical precision and it's responses to being struck by a weighted hammer at different speeds and amounts, then model the complex and non linear interactions between the strings (all 88 or so of them), as well as the interactions with the piano body (including materials, size, amount of space, whether closed or open), and all sorts of other things too (different types of pedal etc) and all this in realtime with no glitching even when played very dynamically as the Piano is a very expressive instrument. And that they have done this and I can get over 200 sustained notes of polyphony even when playing very fast (as I like to) and still only use less than 50% of my P4 3.6 is nothing short of astonishing.

And with normal playing it barely registers.

[Yossarian]

I feel like Ziokiller - it's a great step forwards in technology and does sound great but I can't afford it right now

Same here, I'm afraid. It's on the list, though (that sucker's getting longer and longer... ).

/Yoss

[Krakatau]

Just the CPU optimisations alone are really impressive - Thinking about it, to model a piano properly you would need to model each string with mathematical precision and it's responses to being struck by a weighted hammer at different speeds and amounts, then model the complex and non linear interactions between the strings (all 88 or so of them), as well as the interactions with the piano body (including materials, size, amount of space, whether closed or open), and all sorts of other things too (different types of pedal etc) and all this in realtime with no glitching even when played very dynamically as the Piano is a very expressive instrument. And that they have done this and I can get over 200 sustained notes of polyphony even when playing very fast (as I like to) and still only use less than 50% of my P4 3.6 is nothing short of astonishing.

And with normal playing it barely registers.

I guess there is more than one string per key on a piano ...

....master bluedad, how much ?

[aMUSEd]

Is there?

[Johnny_Boy]

Similar conversation at the PianoWorld:
http://www.pianoworld.com/ubb/ubb/ultim ... 240/2.html

It has some comparion MP3 files using the same MIDI files.... (art vista, sampletek, pianoteq, giga pianoII, and a real piano)

[The Chase]

hmm heard the mp3's, sounds as realistic a piano as donald duck farting. this is crap.

Bullshit. I am sure that nay-sayers like you wouldn't be able to hear the difference between this and a real piano, especially in a mix.

i have been playing piano my whole life you dumb f**k, classicaly trained from the age of 5.

those mp3's sounds like a piano for one split second, then morph into something else the next, then something else. This does NOT sound like a PIANO.

for a QUARTER of the price of this i got some of the sampletekk stuff in their group buy that just blows this piece of overpriced 249 Euro crap away..what a larf. LMAO, 'm going to spend $500 AUD for a modelled piano that doesn't em, sound like a piano. NOT.

You're being drastically negative. There are no other synths that come this close to a real piano.

If anyone heard this in a movie or a coffee shop or restaurant, I highly doubt one would think, "Hey, this piano sounds synthesized!" If anything you might think it sounds a bit odd, but that's it.

[Cyrosis]

Personally I am using the Art Vista steinway samples, I have compared the 2 on identicle midi's and find the Pianoteq to be very clearly the loser as far as realism. It of course is still impressive for being fully moddled but as its been said, I dont believe it comptetes with some of the really nice sample libraries.

Of course it doesn't, nobody's saying you should delete all your sample librarries now that Pianoteq is here. But realize it's a V1.0, and the sound coming out of this is already very impressive, given it's entirely modeled. I don't know if you're fully aware of the beauty of the technology behind this, it's the first time* somebody does a PM piano. You probably know how challenging PM can be, designing a piano PM is even more... (I didn't think it would be possible this early by the way).

Above all, as aMused pointed out, CPU optimisation in Pianoteq is already very good, so it's really encouraging as for the future of this VSTi, plenty of CPU left to emulate more closely sympathetic string resonance, hammer action, dynamic phase variation....

This is groundbreaking stuff, and only the beginning. I'm glad somebody dared to try piano modelling. Keep up the good work !

* Ok, looks like Klaviermodell was here before : )

It is groundbreaking - I've been wanting something like this a long time (btw - I had the Reaktor ens Wakax mentions since it came out and there have been other Piano models such as in Tassman and String Studio 2 and also 6 String for Creamware was up till now probably the best but this is streets ahead in terms of playability, expressiveness and sound). I find it totally incomprehensible how people can blithely and dismissively compare it to sampled piano's that have been in a process of stepping over each other's achievements in terms of realism (and size) for many years now with something that is such new technology - this is a huge step forward for PC based music - some perspective is needed I think.

Who cares about achievement when it would be spending 330$ on a piano emulator that doesnt sound like a piano! Achievement, breakthroughs, and impression have nothing to do with a product that doesnt deliver.

[The Chase]

Why are they calling it the first moddelled piano? Wasnt the electric piano technically a moddelled piano?

[ianweb123]

Do you suppose that when the original King Kong film came out in the 20s the audience thought " well King Kong was good but I reckon it will be a lot better when its all done using CGI".

There are those that will say that synth x never sounds like instrument Y.. I wouldn't mind betting that if you played them a recording of a real piano EQed up they would say that it doesn't sound like a real Piano..
Personally I feel that this is a very good instrument, does it sound like a real piano? I don't know, I'm nowhere near 'expert' enough to allow others to appreciate it.. Do I enjoy playing it? How can I .. It doesn't sound anything like a Steimway Model O - 180 Miniature Grand, esspecially after it was tuned by Stanley Harris on a tuesday afternoon..

FFS appreciate things for what they are, for once.

[aMUSEd]

Personally I am using the Art Vista steinway samples, I have compared the 2 on identicle midi's and find the Pianoteq to be very clearly the loser as far as realism. It of course is still impressive for being fully moddled but as its been said, I dont believe it comptetes with some of the really nice sample libraries.

Of course it doesn't, nobody's saying you should delete all your sample librarries now that Pianoteq is here. But realize it's a V1.0, and the sound coming out of this is already very impressive, given it's entirely modeled. I don't know if you're fully aware of the beauty of the technology behind this, it's the first time* somebody does a PM piano. You probably know how challenging PM can be, designing a piano PM is even more... (I didn't think it would be possible this early by the way).

Above all, as aMused pointed out, CPU optimisation in Pianoteq is already very good, so it's really encouraging as for the future of this VSTi, plenty of CPU left to emulate more closely sympathetic string resonance, hammer action, dynamic phase variation....

This is groundbreaking stuff, and only the beginning. I'm glad somebody dared to try piano modelling. Keep up the good work !

* Ok, looks like Klaviermodell was here before : )

It is groundbreaking - I've been wanting something like this a long time (btw - I had the Reaktor ens Wakax mentions since it came out and there have been other Piano models such as in Tassman and String Studio 2 and also 6 String for Creamware was up till now probably the best but this is streets ahead in terms of playability, expressiveness and sound). I find it totally incomprehensible how people can blithely and dismissively compare it to sampled piano's that have been in a process of stepping over each other's achievements in terms of realism (and size) for many years now with something that is such new technology - this is a huge step forward for PC based music - some perspective is needed I think.

Who cares about achievement when it would be spending 330$ on a piano emulator that doesnt sound like a piano! Achievement, breakthroughs, and impression have nothing to do with a product that doesnt deliver.

Because it does deliver - in spades. Firstly it does sound very good and no amount of denial can make me think otherwise, I've been playing this all afternoon and it's a breath of fresh air musically. Sure it's not going to sound perfectly like a piano yet but it's already much more playable and responsive than all but the largest multi-sampled libraries and they do tend to glitch if I play them fast. This doesn't and if they can do this in version one I for one would want to support them in developing this further instead of supporting more of the same sample sets that get bigger and use more resources everytime. Plus I'm not anal about it not sounding 100% like a piano - it plays more like a piano and more like a real instrument and that's what counts - it can sound like lots of other things as well which is more interesting to me anyway - the whole point of a good physical model is it's a starting point musically, not just an end in itself.

[Cyrosis]

Do you suppose that when the original King Kong film came out in the 20s the audience thought " well King Kong was good but I reckon it will be a lot better when its all done using CGI".

Do you think in the 20's they already had another method of creating amazingly real special effects that had been in use for many years? I think not.

[eduardo_b]

It's probably been in development for years - why should they give it away at a bargain price?

Because it's just too expensive for what many would consider a non-essential plug-in. It's not as good as sample-library plugs for the same money if the goal is authentic sound, and, as a modelled piano with effects, it's not going to be perceived by many as a worthwhile expenditure. It may be unique, but not necessarily worth more than half of what it's selling for.

This really is déja vu - just as with Synful, people are so focused on the small shortcomings that they can't see how close we are already and what potential there is.

It's about value for money. How useful will this plug-in be for most, and what else are they wanting to buy when it comes to software music tools. This "piano" is priced with others in the market that really are pianos. Good isn't great, which is why Synful gets the same opinions from many. Potential for what, and for how much? That's what determines this market. After the early enthusiastic adopters have paid their money, the sales curve will drop as the rest of us hang onto our money. Cut the price in half, and I might be interested -- and it' still more than $150 even then.

[ianweb123]

Do you suppose that when the original King Kong film came out in the 20s the audience thought " well King Kong was good but I reckon it will be a lot better when its all done using CGI".

Do you think in the 20's they already had another method of creating amazingly real special effects that had been in use for many years? I think not.

Well apart from the fact that they could have used footage of a real gorilla, hey hang on, thats just like sampling.. using recordings of the real thing..

so yes I had already considered that reply

[Cyrosis]

Personally I am using the Art Vista steinway samples, I have compared the 2 on identicle midi's and find the Pianoteq to be very clearly the loser as far as realism. It of course is still impressive for being fully moddled but as its been said, I dont believe it comptetes with some of the really nice sample libraries.

Of course it doesn't, nobody's saying you should delete all your sample librarries now that Pianoteq is here. But realize it's a V1.0, and the sound coming out of this is already very impressive, given it's entirely modeled. I don't know if you're fully aware of the beauty of the technology behind this, it's the first time* somebody does a PM piano. You probably know how challenging PM can be, designing a piano PM is even more... (I didn't think it would be possible this early by the way).

Above all, as aMused pointed out, CPU optimisation in Pianoteq is already very good, so it's really encouraging as for the future of this VSTi, plenty of CPU left to emulate more closely sympathetic string resonance, hammer action, dynamic phase variation....

This is groundbreaking stuff, and only the beginning. I'm glad somebody dared to try piano modelling. Keep up the good work !

* Ok, looks like Klaviermodell was here before : )

It is groundbreaking - I've been wanting something like this a long time (btw - I had the Reaktor ens Wakax mentions since it came out and there have been other Piano models such as in Tassman and String Studio 2 and also 6 String for Creamware was up till now probably the best but this is streets ahead in terms of playability, expressiveness and sound). I find it totally incomprehensible how people can blithely and dismissively compare it to sampled piano's that have been in a process of stepping over each other's achievements in terms of realism (and size) for many years now with something that is such new technology - this is a huge step forward for PC based music - some perspective is needed I think.

Who cares about achievement when it would be spending 330$ on a piano emulator that doesnt sound like a piano! Achievement, breakthroughs, and impression have nothing to do with a product that doesnt deliver.

Because it does deliver - in spades. Firstly it does sound very good and no amount of denial can make me think otherwise, I've been playing this all afternoon and it's a breath of fresh air musically. Sure it's not going to sound perfectly like a piano yet but it's already much more playable and responsive than all but the largest multi-sampled libraries and they do tend to glitch if I play them fast. This doesn't and if they can do this in version one I for one would want to support them in developing this further instead of supporting more of the same sample sets that get bigger and use more resources everytime. Plus I'm not anal about it not sounding 100% like a piano - it plays more like a piano and more like a real instrument and that's what counts - it can sound like lots of other things as well which is more interesting to me anyway - the whole point of a good physical model is it's a starting point musically, not just an end in itself.

I am not trying to persuade you otherwise, and I see you have your reasons, even if I dont understand them fully, but the entire ground for me is that it doesnt sound like a piano, so I have no reason to see it as a ground breaking technology to emulate one.

[Cyrosis]

Do you suppose that when the original King Kong film came out in the 20s the audience thought " well King Kong was good but I reckon it will be a lot better when its all done using CGI".

Do you think in the 20's they already had another method of creating amazingly real special effects that had been in use for many years? I think not.

Well apart from the fact that they could have used footage of a real gorilla, hey hang on, thats just like sampling.. using recordings of the real thing..

so yes I had already considered that reply

yes, and im sure that crossed their minds, but they wouldnt beable to get a real gorilla to fight a T-rex, or climb a building holding a woman, fend off planes etc. BUT with samples, I can make them play a wonderfully realistic piano tune better then all the alternatives.