What is stopping you guys from Pianoteq building an orchestra (I am yet to hear a realistic strings on par in quality with Pianoteq....)
and what about brass, a full orchestra that unlike those deeply sampled libraries does not use ug Gigabytes of storage...
Just wondering
because all those instruments are not keyboard based / played ? ;-)
Modartt is a small team of experts focussed on piano and closely related instruments.
For realistic brass and woodwinds (and viola, the only string for now) check out http://samplemodeling.com/en/technology.php - e.g.:
trumpet: https://www.youtube.com/watch?v=Gp6wV0nuphk
trombone (old from 2010): https://www.youtube.com/watch?v=HqnWX0338fM
I don't like the fact that it only works on Windows/Mac and it's rather pricey. It's not like Pianoteq where you buy the software and add instruments at much lower cost. Each instrument pack is sold as a separate piece of software and is priced accordingly.
because all those instruments are not keyboard based / played ? ;-)
Hello John,
I completely concur with Luc on his reply. If one were going to model a bowed string, then the player would have to use his/her virtual bow on an individual stringed instrument. Likewise, if one is a brass player or a reed player, or a double-reed player (oboe, bassoon, contrabassoon), then that person would need to encode his/her embouchure into a single instrument. Then the following variables become important: lung/wind pressure, size of one's mouth, shape and tension of lips, mouthpiece geometry, reed characteristics, -- to say nothing of which fingering will yield a given note. (I am also a brass player, and am well aware there are usually two or more fingerings to yield the same tone -- but different placement on the harmonic series.) Finally, one would have to deal with "massed strings", etc.
Enough of my rambling; I hope you get the picture.
Cheers,
Joe
(...) Likewise, if one is a brass player or a reed player, or a double-reed player (oboe, bassoon, contrabassoon), then that person would need to encode his/her embouchure into a single instrument. Then the following variables become important: lung/wind pressure, size of one's mouth, shape and tension of lips, mouthpiece geometry, reed characteristics, -- to say nothing of which fingering will yield a given note. (...)
I doubt many people, if any, are able to distinguish between these features when listening to real players — I certainly have never been able, from merely listening to their recordings, to gauge which of the two, Heinz Holliger or Lajos Lencsés, has the larger mouth cavity or what shape their lips are — so I don’t really see why these aspects (even if essential to the sound generating process in real life) should be a part of the virtual model, other than make it unnecessarily complicated.
The ‘keyboard’-thing is not so much the problem, in my view. (Certainly no more so than it is with virtual harps or guitars.) It is — as always — the complexity of living sound (the very thing which stops modeled pianos from sounding real), the countless ever-changing details and the unmodelable wide range of musical expression (timbrally and dynamically) which allows a player to deliver 100% context-aware performances, that makes modeling wind instruments such a challenging, near-impossible affair. At least, at this moment in time.
Having said that, at least one of the SampleModeling instruments — their maiden release “The Trumpet” — is, in my opinion, a work of art. Wonderful achievement, which I don’t see bettered any time soon. The other brass are pretty good too but, for some reason, not quite as magical as the trumpet. (All personal opinion, this.) The woodwinds are generally considered less of a success, and I agree, though it’s still a very useful collection of virtual instruments to work with. The saxophones are surprisingly capable too. (If you don’t expect too much, that is.)
The viola however, while certainly another technical tour de force, is — still in my view — more of a miss than a hit. Too much timbral problems, to my ears, to seriously consider this as a ready-for-prime-time subsitute for a good sampled viola, let alone for the real thing.
Looking at what’s available today and at what’s been attempted in recent years, I think that we’ll have to wait quite a bit longer for convincing modeled strings (solo and sections) than for more-or-less believable modeled wind instruments.
_