Topic: Ideal piano

I've been using Pianoteq Stage for several weeks, and the model I prefer is the K2, probably because it comes closest to my concept of an ideal piano. For centuries, piano makers have sought to compensate for the properties of materials that make the pianoforte flawed as a musical instrument. Those flaws include the inharmonicities of steel strings, necessitating stretch tuning, resonances of soundboards and wooden case that make the frequency response of the instrument uneven. Strings have been made longer to reduce inharmonicity, but without eliminating it. Besides that, the physical layout of the piano is less than ideal for projecting sound, and because the strings are spread out over a wide area, microphones used in recording need to be placed so as to minimize phase cancellation of sound waves.

Thus far, a primary goal of Pianoteq seems to be to create the sound of a real piano in all its glory and all its flaws, in the interest of realism, including inharmonicity and strings not tuned to perfect unison. I tuned pianos years ago, and I always tuned unisons as perfect and beatless as I could, knowing that they would go out of tune soon enough, without my giving them a head start. I would play the piano after tuning it, and for me, beatless unisons gave each note purity and power, and the overall sound of the piano was a delight to hear. The beating of equal-tempered intervals remained. Some people like pure intervals, but for me, the beating of equal tempered intervals in a chord is a pleasant sound, often compelling harmonic progression to the next chord.

The piano maker strives for an ideal piano without ever actually attaining that ideal. Compromises are made, and properties of materials are accepted as unavoidable. When I listen to some of the Pianoteq models, I sometimes hear the imaginary wooden cabinet resonating in a way that is not perfectly pleasant. The technology of Pianoteq opens possibilities that a traditional piano maker can only dream about. In the digital realm of sound design, it is possible to eliminate or drastically reduce the inharmonicity of strings, the resonance and uneven frequency response of wooden cabinets, phase cancellation, and other flaws of the real piano. While Pianoteq may have limitations in terms of creating the sound of a real piano, it can create a sound that approaches the ideal in a way that a real piano cannot.

Is it possible, using Pianoteq, to design a piano that sounds better in some ways than a real piano? Could Pianoteq be improved by introducing a new model that improves the sound in ways not possible with a real piano? Such a piano could be regarded as a digital musical instrument without a physical counterpart.

Re: Ideal piano

Hi Steven
Maybe it's time for you to upgrade ?
Have a look at the differences between versions :
https://www.pianoteq.com/pianoteq5
Only with the Standard version you can access some essential settings such as string lenght, sympathetic resonance or sound board impedance.
If it's not enough, go to Pro version, then you are able to manage all the parameters that allow you to create your ideal piano.

Re: Ideal piano

it is possible to eliminate or drastically reduce the inharmonicity of strings, the resonance and uneven frequency response of wooden cabinets, phase cancellation, and other flaws of the real piano. While Pianoteq may have limitations in terms of creating the sound of a real piano, it can create a sound that approaches the ideal in a way that a real piano cannot.

I'm not sure what you're aiming at, but to me it sounds like you're trying to make something that is not a piano.  All the things you're trying to eliminate are what make individual models what they are.

I'm not sure what you expect to replace a cabinet with.  Are you hoping to remove a cabinet from a model completely, because I don't think that's possible (or for me desirable) ?

From what I understand of pianoteq the PTQ files that are models contain model specific data (and code ?) which is used to generate the sound.  I don't think you can do what you want from any pianoteq version.  It would be equivalent to requiring Modartt to give you the means of creating your own models (as opposed to adjusting existing ones).  I don't think that's likely to happen.  It would allow anyone to release new models of piano and that sounds like a kind of financial shooting in the foot for Modartt, who clearly would loose a potential revenue stream (which their hard work created in the first place).

If you're using Stage you probably ought to try Standard (at least).  The microphone placement and use alone is probably something you'd value in your search for this "ideal".

The Pro model adds note-by-note editing which is probably the closest you'll get to the level of control you seem to want.  The Standard version lacks this level of control, but can fully use FXP files created with Pro, but you'd need Pro to do such detailed editing.  But again note that even on Pro you're just adjusting the model, not creating a new model.

StephenG

Re: Ideal piano

A while ago there was a tool, PtqSpecProf offered by a contributor Mr. Gilles. The aim of which is to "redefine the tone color of all the notes" I quote the author, "... cannot change the global behaviour of pianoteq of course, it just alters the tone color so as to bring it closer to another instrument".

One must have the pro version to use this tool. I think the idea is that if a sampled piano make/model is not too far off tone colorwise from a Pianoteq model, a skilled individual can use the tool, along with knowledgeable tweaking, to create a Pianoteq model that comes reasonably close aesthetically to the sampled model in question. This would indeed constitute the creation of a new model if (big if) close enough.

I find it of note that Mr. Brown spoke highly of the K2. Other posters have said that the K2 sound was reminiscent of a Fazioli.  I quote from the Modartt literature "The K2 Grand Piano is developed especially for Pianoteq 5. It is not based on any specific model but created from scratch by the Modartt virtual piano factory, combining the best elements of several source pianos."

The K2 model seems to be an attempt by Modartt to create the "ideal" piano as they saw it at the time. Certainly different from any other physical pianos but resembling the "best" of them.

The above seems to be along the lines alluded to by Mr. Brown. I am thinking that Pianoteq Pro may indeed have enough flexibility to create aesthetically pleasing changes so different from a particular model, so as to effectively and practically constitute a new noteworthy model/instrument.

There is commentary, details, and attempts at new models using PtqSpecProf . This tool is limited and requires a highly skilled individual to use it effectively after it generates Spectrum Profile information. If it were highly enhanced, would it affect Modartt's revenue stream,  I do not believe so. Apart minor tweaks using Pro I stay away from trying to imitate a particular sampled piano, because the effort, experience, expertise, technical skill set, and the "golden" ear required for that endeavor is extremely high and requires many years of honed ability. I am looking forward to upcoming models, and I am thankful for all the generous FXP contributors.

Re: Ideal piano

sjgcit wrote:

I'm not sure what you're aiming at, but to me it sounds like you're trying to make something that is not a piano.  All the things you're trying to eliminate are what make individual models what they are.

I thought I was clear in expressing the idea that the flaws piano makers strive to minimize can be dealt with in the digital realm to an extent not possible with a physical piano.

sjgcit wrote:

I'm not sure what you expect to replace a cabinet with.  Are you hoping to remove a cabinet from a model completely, because I don't think that's possible (or for me desirable) ?

No, just a cabinet that has a flat frequency response, instead resonating in some bands of frequencies but being dead in others. The model of a cabinet does color the sound in Pianoteq, but it seems to me the goal of realism may conflict with the goal of creating a beautiful sounding instrument.

sjgcit wrote:

From what I understand of pianoteq the PTQ files that are models contain model specific data (and code ?) which is used to generate the sound.  I don't think you can do what you want from any pianoteq version.  It would be equivalent to requiring Modartt to give you the means of creating your own models (as opposed to adjusting existing ones).  I don't think that's likely to happen.  It would allow anyone to release new models of piano and that sounds like a kind of financial shooting in the foot for Modartt, who clearly would loose a potential revenue stream (which their hard work created in the first place).

I am not suggesting that, but Modartt could create new models in which the imperative to emulate real instruments is relaxed in favor of creating new instruments that sound beautiful. The K2 seems a model aimed in that direction, and perhaps I could obtain what I want by tweaking the K2 model in Pianoteq Pro. A feature that interests me is the ability to adjust the harmonics or partials of the virtual strings. For example, sometimes a bass string growls or buzzes in a way that I think can be improved.

Re: Ideal piano

Now I've become much more interested in historic pianos as a result of using pianoteq - I'd hardly thought of them at all before.  I think this leads down a completely different rabbit hole than you followed.  You think of piano makers are striving to eliminate flaws, whereas I think of piano makers as having made instruments that lost something they once had.  I now find myself holding the view that period music was designed for the nuances of period instruments and often doesn't work as designed on modern instruments.  It's not that modern instruments deliver worse sounds, it's that they deliver different sound from the original intent, and I've come to appreciate the original intent a little more than I used to.

I think I'm a little concerned you're going to eliminate the character of instruments.  The word "ideal" has always made me uncomfortable, I think.

However, good luck in finding the sound you want and I think Painoteq Pro is probably a must for you.

StephenG

Re: Ideal piano

I tried a while ago to discuss such an "Ideal" piano on pianoworld.
I consider it to be what the piano builder strives for, but knows they cannot achieve.

I agree; the limitations of materials, the case resonances, the bass/middle and middle/treble breaks, the soundboard, the incidental noises that can't be muffled however much felt is applied - - all lead to a much less than "Ideal" instrument.

However, adherents to tradition insist that a piano without all these artifacts just isn't a piano.
...and because these have been accepted over time as the definition of a piano - I accept that.   

The basic Karpus Strong equations produce such an "Ideal" string, but the wood and metal that physical pianos have to be built from add/subtract from that.
So, it is unfortunate, but what you have described is a synth - and there is nothing wrong with that, it just doesn't have the (largely unintended) characteristics of a "Piano".

I don't know HOW you would derive a synth in Pianoteq, there is probably too much "Piano" included at the very core of the model(s).

Re: Ideal piano

Thanks for your post Steven Brown, I enjoyed reading it.

Major misunderstandings possible: "Ideal", or even just "better" clearly only boils down to the manufacturers' specifications, and then the opinion. More sales don't necessarily mean better, but maybe just cheaper. Some think, "Wow, the sell a half million of these, must be good." where others might think "Smart strategy, they have cornered the school market for entry level models." In either case, their top tier instrument might have some kinds of theories applied.

You like the K2 - so that's your ideal to work from I can confirm that an upgrade will allow you to alter that sound in vastly different ways.

There is never anything objectively perfect or ideal in instrument manufacture. One man's trash is another man's treasure etc. Flavor is king.

Your idea of a better sound will kill it for someone else I love this kind of discussion and glad to read others posting about it.

You can however extremely effectively alter the sound in Pianoteq from heavenly on to creative synthetic sounds when adjusting things in non-piano minded ways.

Steven Brown wrote:

Is it possible, using Pianoteq, to design a piano that sounds better in some ways than a real piano? Could Pianoteq be improved by introducing a new model that improves the sound in ways not possible with a real piano? Such a piano could be regarded as a digital musical instrument without a physical counterpart.

I think most of us reading that would have to guess what you're saying and it poses a fun dilemma. What is "better"? Who decides?

This is not to be disrespectful of your question, and I don't think anyone else is either. It is fascinating once you begin to consider it more. I nearly posted yesterday with the below except that I thought Gaston gave a perfect answer, so you could explore yourself the possibilities. Since others have decided to also post, I hope some of this is enjoyable to consider.

How long is that string? Maybe string theorists can argue that out

Seriously 'though, when it comes to the available models I'm certain that some consider the default settings Pianoteq already too perfect by comparison to a real piano and thus proceed to add variations to stretch and unison width and way more. Users upload their take on their favorites, trying to replicate what they like most and so it goes.

Can there be one perfect model from which other models spring? Maybe Pianoteq wouldn't be able to release new models in that case.

After years of tweaking, I still return to the defaults for sanity checking because much tweaking in any direction over any time can amount to something similar to snow-blindness for the ears.

Think of this 'though:

Is one piano already considered perfect enough to base further perfection enhancements?

That's the essence of the problem. Maybe a glass piano can kill off a lot of woody variables and still maintain a variant of a piano sound. But, it would be a glass piano. Not "a" piano, or less a perfect piano.

The famous painting "Ceci n'est pas une pipe." or "This is not a pipe." aprox. 1928 by René Magritte for example is a painting of a pipe, so it's therefore not a real pipe. Seems simple statement but has deep roots in confronting what he named La trahison des images or the treachery of images.

Maybe you're posing a kind of La trahison des pianos or treachery of pianos!

No matter how perfect to you,

Consider that there have been universally accepted standards for Martinis for many decades. It's was a perfect thing in the Mad Man jet age, to land at any airport in the world and order one drink you can rely upon. (Don't mention burger chains) In the right joint at least, they care about the standards, whichever of those you choose when you order one. People consider the original, or the variants to be better (Gin or Vodka, amount of Vermouth, shaken, stirred, olives or not etc.). It's human behavior, marketing and convenience and sometimes a disregard for history.

Even amongst the most popular and singular items, there are infinite variants and tastes and human variation as to how it's meant to be, or not, or then again, and so on, and so forth

Sticking with that metaphor, imagine a more perfect  Whiskey (or Whisky? Scottish? Irish? Bourbon? Take your pick.. malt, barley, rhy, corn, sour mash). Could it be possible there is no such thing as a perfect drink beginning with W. The inherent reason being that without all, or a subset of any of that imperfect infused peat and burnt oak character making it all blurry in the mouth, it would just be alcohol. Or exactly how much or a slightly different amount of peat or oak etc. could possibly be argued to be perfect? Esp. if everyone needs to agree, or at least believe it sufficiently that it becomes a societal norm, across the globe for it to become something worthy of the whole human historical record.

Something has to be sufficiently different, not necessarily a variant of perfection for it to gain that kind of traction.

Or what about wine. Imagine a perfect grape variety, and having agreed upon that, do we just throw that away to focus instead on a more pure alcohol flavor which is without that imperfect twang of wood dust and berry cinder hints, wood goat cheese aftertaste. Shudders. Myarrgh. Brrrbrrbr. There are wine companies who for decades try to flood the markets with such stuff of course, thus the importance of mass marketing and position in supermarkets. Still doesn't cut it as a real wine IMHO. There are numerous humerous names for it needless to say. It's popular, sells the most by volume. Cheap, convenient.

Whilst still in that pub environment, (hick.. prdn me) what of cigars? Could they be so much more perfect without all that smokey interference (Vaping?).

[Later that evening..] Vodka would be so much better if it had less flav.. << wait a minute.

The way I see it, for each person, that supposed ideal would be quite different by default.

Only given brainwashing levels of exposure to any single type of sound, could we all agree on a particular already given example of perfection which would totally stifle human growth IMHO.

Beethoven wanted a bigger piano. That was the direction for future manufacturing.

In much the same way, in some genres of popular music, when something becomes supposedly the best or near-perfect to an audience, omnipotent like a certain "swell / hit" sound in dance music, everyone jumps on that novelty and all of a sudden, everyone producing club copy/pastes their own whooshing swell/hit moments with a yobbo yelling "Yeah baby" until it finally becomes old-hat some centuries after humans finally grow bored of that kind of thing (likewise for a million production tricks since beginning of human awareness).

So similarly, like Beethoven wanting more keys, suddenly, a dance track producer must have that new swell and a hit and a party atmos. etc. and that famous dance piano sample or everyone thinks it's rubbish.

For centuries a certain level of trill would have been the accepted way and now, just you listen to all these show-offs trilling extra large and long, flourishing where there should be only the sound of God's disapproval.

Think of classic 70's R&B, Soul, Funk sounds - played in real-time by real musicians and recorded by all kinds of analogue recording studios - re-sampled for generations because humans still enjoy it and consider that is sounds all kinds of superior cool. You grow up hearing it, even if you didn't like it. If you're a musical person, chances are that you can at least appreciate its influence as you grow older. Swap out R&B for any given genre or style or era if that's more apropos.

Few of us grew up with an idealised piano model. Most who did would probably realise that it was probably one of the models in Pianoteq

In relation to the Pianoteq model, for many piano lovers, all of the characteristics to tweak, or a mix of such inferred imperfections in a given piano are utterly priceless.

Pianoteq does an amazing job to identify them and to make them customizable. The way a smokey old cabinet sound might blur or ampl. a range, adds character. To others it's not perfect, or "like" current tastes might dictate (as in, what's the most sampled/used dance piano jangle currently).

That's OK - you can tweak a lot of the imperfections out, to leave a more pure sound - but here may be the mythical cuttoff point you might be talking about?

Perhaps, perfect could be just sine waves. Synths from decades ago began down the path of allowing us to hear "perfect" wave forms. What you see on the oscilloscope is what you hear Mr. Moog.

Endlessly changeable, there can only be favourites but I doubt anyone can say even their favorite synth is "the" perfect synth. They all do some things differently - yet, they can probably all play back "perfect" waves, syn/square/traingle and maybe some based on string theory

For anyone thinking about this, creating one's own idealized piano using Pianoteq is as close to possible as with any software out there. Perhaps you may need to upgrade to find some of the settings you want to tweak. The price comes without blinking if such tweaking is your bag.

You can make massive changes just using the main interface with point/click, like:

Lower or remove the hammer sound altogether, also bump up the hammer hardness - so, it sounds more pure.

Drop symapthetic resonance, duplex scale (aliquot strings in Bluethner) - so, it sounds even more pure.

Kill off that imperfect unison width - perfect.

Remove or lower Impedance and Cuttof and bump up the Q factor - you take out all that imperfect piano body distraction.

Flat 440k tuning with no stretching, no problem.

Add the direct sound up to 100 and maybe EQ/compress/reverb it to your liking and choose output (Stereo or Binaural maybe).

Now, you have way more pure tones.

It is no longer a piano though.

It can be maybe over-tweaked down to something like a "pluck" synth sound. It's strings to the ears. Like close mic'd harp, muted.

You can make it sound like bells, small or large. It's a tweaker's time-sink.

What's amazing to me, is that you can take your favorite Pianoteq piano model(s), and tweak cabinet resonance, hammer hardness, tuning etc. for months and years and still feel like you own about 500 amazing "real" instruments.

You can also "abuse" each of the settings alone or in concert and create quite viable and organic sounding impossible instruments.

It's more than just a couple of pianos to me - none of which can ever be perfect because there is no such thing but the ability to alter the sound of any given instrument in Pianoteq is practically limitless and then you can continue from there with all the effects in the DAW of your choice. Sky is truely the limit, as the expression goes.

I am a purist in some senses but also love cross processing everything.

It would be really enlightening to hear what you consider a more perfect sound particularly and maybe someone can help you tweak that out.

Pianoteq Studio Bundle (Pro plus all instruments)  - Kawai MP11 digital piano - Yamaha HS8 monitors

Re: Ideal piano

I can agree with the beer/wine/whisky metaphor.
Whereas the original intent may have been merely to STORE the alcoholic drink, the flavors acquired from the storage vessels became part (some would say now an ESSENTIAL part) of the booze.

Fermenting and storing in glass and stainless steel SUBTRACTS (more accurately, it no longer ADDS) those flavors.
{The short cut now used by chemical engineered plants is to add various wood shavings to the inert vessels in order to simulate barrel aging. }

A mathematically produced "instrument" is faced with having to artificially add in the analogue of the various tannins that alcohol extracts from wood or grape stems and skins.

If intoxication is the goal then sure, get straight rubbing alcohol.
If being a wine/whisky/beer snob is the goal - - you're kidding yourself, its ALL about getting a buzz

Re: Ideal piano

Interesting discussion. I, for one, am glad that Modartt has created one model of piano with an aim of overcoming rather than recreating the inherent compromises of piano design.

I'd like Mozart to continue to move the K2 forward with this goal in mind- what might an "ideal" piano sound like?

I believe it's quite appropriate to have models of extant pianos whose goal is to recreate the character of those pianos as accurately as possible, and have another model intended to see if a synthesized piano might in some ways be able to surpass that of a physical piano.

Re: Ideal piano

Qexl wrote:

It would be really enlightening to hear what you consider a more perfect sound particularly and maybe someone can help you tweak that out.

Thanks for your well-considered reply. To answer your above question, I'll start by saying the piano is my preferred instrument, the one that enables me to play with satisfying expression. It is the percussive attack, the exponential decay that is at first rapid but then trails off to endless degrees of softer until the key is released. It is the rich harmonic content of the sound which varies according to how hard or gently the key is struck, and according to the position of the key on the keyboard. I've never played a synth that affords that degree of expression, and the sound of a synth, though it may be pleasant at first, becomes annoying in the sameness of its sound. Taken to the extreme, a pure sine wave can drive one to madness. I always go back to the piano, and I would not change the settings in Pianoteq to make it sound not like a piano.

What I am saying can be illustrated by the following analogy. Say a brilliant engineer named Polonius decided to go into the business of building a world-class piano, based on his ideas of what a piano should sound like. The piano that bears his name would not be a clone of a Steinway, a Bosendorfer, or a Yamaha. It would be a new instrument, may embody improvements over previous designs, would have its own distinctive sound, and if it's good enough, may receive acclaim in the world of music. And Pianoteq might undertake the development of a new model that emulates the Polonius. What I am saying is that Pianoteq could, and may have already done with the K2, build a piano from scratch, freed from the constraints and limitations of working with physical materials. Instead of trying to emulate a physical piano that already exists, build an aesthetically pleasing instrument that any listener would still identify as a piano.

To address the concern posted by sjgcit, I'm not trying to influence Pianoteq to abandon modeling traditional pianos. I am saying that I would welcome new models like the K2 that offer new variations and possible improvements in the sound of the piano. It may already be possible for me to realize that vision by upgrading to Pianoteq Pro and modifying the K2 as I like.

Re: Ideal piano

I'd like Mozart to continue to move the K2 forward

I'd also quite like Mozart to move any music forward, but I fear it's not going to happen. :-)

Steven Brown : I really didn't think you were suggesting Modartt stop developing new piano models, and I've quite a bit of sympathy for your goal - music would be very boring if people hadn't moved it forward all the time.  I'm just expressing a personal preference for them using their limited resources to refine what they have and add to the models.  I can actually see an argument for a more "abstract" model unrelated to existing pianos, perhaps simply for allowing for extra parameters (in Pro probably) like a detailed frequency response for materials being simulated which can be edited (although just saying it, it sounds complex to implement effectively).

StephenG

Re: Ideal piano

To delve into my idea of what the ideal piano would sound like, I would start by saying that as a composer, I explore the boundaries of tonality, using consonance, dissonance, conventional and unconventional harmonic progression, simple and complex chord structures. The ideal piano would enable a complex soundscape to be created and clearly heard without the muddiness introduced by the inharmonicity of strings and uneven cabinet resonances. The partials of low notes would tend to be in accord with high notes, rather than in conflict. The sound would be clear and full, not thin.

Last edited by Steven Brown (27-03-2017 00:15)

Re: Ideal piano

That sounds like a fine recipe for a piano

Some expensive architectural 3D rendering software I wish to own someday has a great feature along the lines of "add age" - the result being that you can take your rendering which would otherwise look too plastic/perfect and make it realistically grow some algae around edges of pools, dull colours, add cracks in paint where apropriate, puts soot under window sills etc., all tweakable of course. Great for answering the question "What might this building look like in 10 years?".

In reverse to this, Pianoteq can be imagined to begin at "aged" or at least "affected by reality" such as cabinet and all the params, so it might work to ask the question "What might this piano sound like with lessened influence from those factors".

Taking that idea, perhaps in terms of reverse-engineering a Pianoteq piano sound/model, maybe the theory you're evoking might apply here by being able to turn down settings like sympathetic resonance and all the others in concert. You can bring the sound to crazy states of differences in all areas - but you also can certainly pair things back whilst not eliminating the audible illusion of a given quotient for piano.

I'd definitely recommend upgrading Pianoteq so you can pull back on and push around those concepts.

K2 rings like a bell, to my ear and has been overlooked in my usage but I do play it from time to time and agree it could be a good model to work from in achieving your stated params.

One of my failings is to add too much extra flavor . It sounds great at the time but often has obvious problems in later listening. But that does work well sometimes, even after months passing to re-evaluate, so then I tweak it back where it's needed and when happy I just right click and save that settings to a folder of my choosing to access apart from those "nearly but no" variants.

It actually probably took me more than a year or so to get more serious about "less is more" and actually pairing settings back. Really, thanks to all for posting. This is a topic which can never be resolved, but it's quite fun.

I owe aandrmusic a nice buzz if we ever meet.

Best o' luck!

Pianoteq Studio Bundle (Pro plus all instruments)  - Kawai MP11 digital piano - Yamaha HS8 monitors

Re: Ideal piano

What a fascinating topic!  I greatly enjoy reading it and pondering over it.

I think in discussing this topic, it is very important to separate the notion of an objective ideal and individual preferences.

I think there may well be an objective ideal for a piano, but it is also probable that not everyone will like it.

An analogy is in audio loudspeakers. It can be agreed that objectively the ideal behaviour of a loudspeaker is a ruler flat frequency response from 20Hz (or even lower) to 20kHz (or higher), with even sound radiation characteristics and no phase / time abnormalities. 

Now not everyone likes this kind of sound - some find it too dry or "clinical".  Many prefer a "warmer", "tube-like" sound, or "character", which generally results from a skillful introduction of certain "deviation" from the objective ideal. 

It's perfectly OK for us to have our individual preferences.  But it DOES NOT in anyway change the fact that an objective ideal exists and that it remains a laudable goal to pursue it.

I acknowledge that the piano is very different from a loudspeaker because a piano produces sound of its own and a loudspeaker only REPRODUCES sound (hence an ideal behaviour is easier to establish), but perhaps one approach to an ideal piano is to consider the numerous separate components of a piano, then imagining the ideal behaviour of the these components (no need to consider any real world limitations), the perfect steel string, the perfect hammer, the perfect cabinet...etc. as well as a perfect harmonization of these components. It will still be a piano, not a synth, because of the inclusion of all the components, but it is indeed interesting to hear what such a piano will sound like (is this the K2?). Perhaps a separate ideal for different types of piano (large concert grand, baby grand piano, an upright piano...etc.)?

This obviously does not replace the equally interesting goal of modelling other existing pianos.

I did not particularly like the K2, but after reading this discussion I played it again and am liking it more and more. Such a pure and crystalline sound.

My two cents. Thanks again to the original poster for such a fascinating topic.

Last edited by lo134 (31-03-2017 19:17)

Re: Ideal piano

It seems that everyone is listening for different qualities in pianosound. Reading this very interesting topic I think, that demands on making pianosound (Ptq) is enormous, gigantic…..But remeber, we have so many different speakers and they affect what we hear or not hear, which pianosound is good or not…So, to get the ideal, best possible, complete… piano - we should also have speakers that are made espescially and only for pianosound, ideal pianosoundloudspeakers. Do such speakers exist? Or maybe that kind of technology does not exist yet? Well, just some more thoughts.

Re: Ideal piano

Qexl;

I think most of the interesting aging effects that you mentioned are available in open source software, e.g. Blender.
While you save up for the ridiculously expen$ive software package you might want to explore that.

I think most of the piano'ness of pianos may come down to the soundboard and the string strike point(s).
The strike point is largely responsible for initializing the harmonics, the soundboard is basically a set of pass filters.
Case, lid and other parts come into it, but as 2nd and n'th order effects.

BTW, I am a "buzz free" individual.
Buzz adds coloration and distortion to the human experience - IMO, etc.

Re: Ideal piano

I was reflecting on the original post again, here are a few thoughts.

Many of the compromises in a physical piano may come from parts having to share other parts, e.g. same soundboard, same bridge, same harp, same case, same string material, only a few string wire gauges, etc.

There might be an "ideal" string wire gauge, string length, strike point, sound board size/shape, etc. for each and every note.
So, in a pianoteq "ideal" piano it may be possible to optimize for each note separately, being free from the physical constraints of only one sound board, bridge, etc.

The question might be whether 88 individual ideal piano/88 could be modeled and then mixed.

I don't know how sympathetic resonances would figure in this.
I doubt that you would want all the soundboard/88 parts to volunteer to vibrate, although you might want to selectively allow some cross talk.

Last edited by aandrmusic (01-04-2017 19:09)

Re: Ideal piano

I had set this thread aside for time to read it and now have completed reading it (to date).

Qexl, I think that your long first post does an excellent job of 'selling' Pianoteq Pro and discussing what is possible.  I believe that I understand Mr. Brown and appreciate where he is coming from in his original requests.  I hope by now that he has acquired Pianoteq Pro and started to tweak the settings:  getting perfect unisons, long string length to remove inharmonicities, adjusting Q and soundboard resonance, etc., and even playing with these on a note-by-note basis to get what he wants.  Perhaps instead of the K2, which has always had an 'odd' sound to my ears, he might try starting with the Bluethner, which seems to be the crowd favorite for tweaking, especially if you judge by the large number of Bluethner presets that simulate other pianos: Grotrian, Steinway, Fazioli, "American" grand, "Austrian" grand, "German" grand, etc.  And, with the note editor, all kinds of teaks are possible, to simulate all kinds of characteristics in the physical world, from modifying the registers between single, double, and triple-string unison placement on the range of the instrument, to various tweaks of the elements that make up the cabinet and the material response of it.  When I first got Pianoteq, I was sure that a piano-tuner would have a field-day playing with it, as so many modifications are so easy in Pianoteq and so hard in the real world (varied unison tuning, hammer voicing, damper changes, temperament, etc.).

Myself, I like going the opposite direction - I like to take a too-perfect-sounding Pianoteq instrument and add 0.75 - 1.00 of the 'weathering' slider and make it sound more real my making it sound more imperfect.  In my book, the reality comes from the quaintness of mild blemishes and patina on the synthesized instrument.

I find Pianoteq fascinating.

- David

Re: Ideal piano

dklein wrote:

Myself, I like going the opposite direction - I like to take a too-perfect-sounding Pianoteq instrument and add 0.75 - 1.00 of the 'weathering' slider and make it sound more real my making it sound more imperfect. In my book, the reality comes from the quaintness of mild blemishes and patina on the synthesized instrument.

I find Pianoteq fascinating.

Like the 'Condition' slider, I wonder if it's now about having extra tools within the software (Pro version) in order to create a desired piano. The problem would be what tools?

My wish list:
I'd like to see the 'Humanize' feature added to the 'Hammer Hardness' section because I think it would allow better blending between the sections 'Piano', 'Mezzo' and 'Forte.' I'd like to hear a more string sound in the pianos. Not just Pianoteq but every digital piano I've tried on my system. I feel like this could be achieved in this section, but I don't know.

Also in the 'Delay' and 'Reverb' sections, there's a 'Tone' control. I'd like a more elaborate version of this.

The ability to combine more than three effects in the 'Effects' section.

Pianoteq is such a good bit of software.

Re: Ideal piano

Cheers aandrmusic and dklein,

Pianoteq is one of those pieces of software which screams buy it all. I resent that in most other software but appreciate the value here.

The tweaking is endless and thoroughly enjoyable along the way and it seems to be possible to learn something of tangible value every day as the saying goes.

sjgcit stated that an interest in antique pianos comes thanks to Pianoteq which is priceless. My leaning was towards period instruments, and I've come to appreciate newer pianos because of this outrageously good software.

DonSmith, I keep forgetting to find a post mentioning the ability to paste a random seed number somewhere to influence what the condition slider changes I'll have to search the forum for that next, and thank in advance the person who mentioned it. Easter eggs all over.

BTW, there are tools for humanizing the hammer hardness - click "note edit" and click to each of the 3 hammer hardness interfaces (for piano, metzo and forte)..

I just draw a line or select "randomize" and then you can further randomize, rescale, smooth, reset or draw a particular hardness or sofness per note.

In the first year of using Pianoteq I kept discovering the depth of already available tweaks and it's certainly endless.

For tone control, there are 2 different types of EQs to play with and for reverb, look into loading WAV Impulse files - another level of richness.

Most of these features are mentioned on the forum - thanks to all those who so graciously posted helpful things over the years. You can "use" Pianoteq for a long time before some of these features surface. The forum is full of excellent advice and tips if you search it.

Every day it's hard to just play without attempting at least a handful of tweaks to suit mood - it's as close as I have come to finding the perfect piano for each situation and I do hope Steven Brown is busy tweaking

Tweak away!

Pianoteq Studio Bundle (Pro plus all instruments)  - Kawai MP11 digital piano - Yamaha HS8 monitors