What's new

Why can't they get solo violins right?

I posted a version of this elsewhere, but I think this is a pretty damn good solo violin (my sloppy performance notwithstanding):

Audio Modeling violin.

[AUDIOPLUS=https://vi-control.net/community/attachments/violin-mp3.18440/][/AUDIOPLUS]
 

Attachments

  • Violin.mp3
    1.6 MB · Views: 247
Last edited:
To me best like embertone stuff is fine based demos but generally i think is related how expressive and how keyboard instrument violin is.(its fully becouse pitch is stepless for example and you can play at least 13 common ways i think)
 
Hmmmm.. but you play your fake violin on keyboard right? Violin is hard instrument to emulate due too many degrees of freedom... It takes a heck of editing to turn a keyboard played violin into something that sounds like a violin, regardless how good the patches are.
And the Joshua Bell violin is a great sampled library.

To get properly sounding violin you need a modeled violin and a controller that can work the model such as roli or something even weirder. But there is the issue...current modeled violins don't have proper "wood" sound and ambience of sampled violins and sampled violins can't work the necessary parameters... so we need some sort of hybrid violin and played on special controller. No money for such a niche, especially when good violin players are everywhere.

And then there is the question why? Why do you want a fake violin so badly to sound like the real violin that even violin player could not tell?

Many violin libraries are good enough to be acceptable by majority of listeners. Listeners usually don't bother with questions - is it real or is it fake violin.Just is your music good or bad.
If you need a precise violin for solo, just hire a violin player.
 
Last edited:
No, I've heard those on youtube, and I can still hear a major difference between Joshua Bell, Spitfire - and a real violin.
I wouldn't call a youtube video (esp if you are listening on PC speakers) the best of light to judge.

I haven't looked at them in depth yet, but suspect you can get close to the real thing. Obviously the real thing is best, but that can be said of any instrument.

But to make a virtual violin (or any solo instrument) that sounds exactly like a real performance by a real player is impossible.
I disagree. Assuming that is by "exactly" you mean "even the best of trained ears can't tell the diff." We just aren't there yet. Tech has already gotten beyond what many considered impossible years ago in this regard. There's no reason to think that just because we aren't there means we never will be. And I admit that notion doesn't wow me and part of me hopes that's wrong. We like to think the real thing will always be a cut above...and for me for reasons other than the sound, it is. I will never go to a concert and watch some guy or band playing iPads or otherwise not playing real instruments (I can live with some backing stuff on a MIDI keyboard). But in terms of the actual sound? The gap has already shrunk considerably. Hard to say timelines, but I would be very surprised if it did not continue to shrink...

And then there is the question why? Why do you want a fake violin so badly to sound like the real violin that even violin player could not tell?
?? Why would you NOT want that?
 
Assuming that is by "exactly" you mean "even the best of trained ears can't tell the diff." We just aren't there yet.

Yes, that is what I mean when I say "exactly." A cheap violin library will fool 99% of the general public, and the best violin libraries will fool the other 1% as well as many musicians, but no violin library will fool a highly-experienced violinist. The ears of experienced musicians are honed over time to hear their instruments in more detail and intimacy than anyone else. The same can be said for people who are closely associated with those instruments, such as conductors and composers. That's why some kid just getting into composing music for video games might go gaga over a new brass library whereas a classical horn player could find the same library rife with so many problems, he'd deem it unusable.
 
As it stands now, no doubt. Going forward? Who knows? But there's no basis to call it "impossible."
 
My statement about it being impossible wasn't concerning future development of virtual instruments. I'm sure as technology progresses, libraries are only going to get more realistic.
 
Oh OK thanks for clarification - so you meant people using it won't ever be able to use it in a way that is realistic enough that the trained ear couldn't spot it?
 
so you meant people using it won't ever be able to use it in a way that is realistic enough that the trained ear couldn't spot it?

I didn't say "won't ever". I said "can't".

As far as the future goes, I think it's safe to say that virtual instruments will become more realistic, and they will fool more and more musicians along the way. But will exposed, solo performances of, for example, virtual violins and saxophones ever fool virtuoso musicians who have been playing the real instruments their whole lives? My guess is no, or at least not for a long time yet. The technology to virtually recreate all of the infinite nuances of acoustic instrumental performances doesn't yet exist, but who knows? It may someday.
 
Last edited:
That's quite a back-peddle from "to make a virtual violin (or any solo instrument) that sounds exactly like a real performance by a real player is impossible."

;) Just messing with you. And again I hope your initial statement is correct. I just fear it isn't, given current trends. With luck it will be after people like me are gone though.
 
Well, I have to admit I've been a little confused by your replies over the past couple of days because I still can't tell if maybe you were misinterpreting my posts and we're actually on the same side on this issue. I've been maintaining the same point of view through the entire thread, which is that a good VI composer could compose something with a virtual violin, but a virtuoso will know it's not real. It may fool a sax player, or a piano player, or a drummer, but not an experienced violinist. It just can't be done. Not by today's technology. A violin performance is comprised of infinite sonic nuances that no library can mimic. Most libraries fool the public. Some get close enough to fool other musicians. But none can fool an experienced violinist whose ears are trained to hear those nuances.

I'm sure modeling technology will develop over the years, and it will make for some amazingly heightened realism beyond what we have now. But I won't speculate much as to who will be fooled by virtual instruments in the future. We'll just have to wait 'til we get there. :)
 
Yeah, I think I took your initial "impossible" post to mean "it can't ever be done" vs "it can't be done today." The coming year with tech in music will be interesting, maybe in ways both good and bad.
 
Because getting a violin to sound right/great simply requires a lot of work (and that includes real violins too).


At bottom it's because the "user interface" of a violin or tenor sax or solo electric guitar is just vastly more rich and complex than the user interface of a keyboard and drawing lines in midi.
 
I can’t seem to find the episode where they describe how to sample solo violinists, but this one covers (skip to 0:47) how to play the flute at a level reminiscent of the current state of the art in solo strings sampling:

 
I think If you're willing to bleed for the programming you can trick somebody well enough. The cost/benefit of the time needed is another matter. I think since a violin performance has no many moving parts it breaks down at the input level, more than the samples.
 
Whoops pressed send to soon ok my earlier post, deleted that one.

As a violin player myself (except I'm scared to admit how long it's been...) I have some thoughts here.

The key problem is that continuous parameters are a fundamental part of the sound. I think it's the same reason that voice is so hard too. On the opposite extreme, this is why percussive instruments are easier to sample. Most of the sound that most people hear about it comes from a one shot hit, which beautifully matches midi note-on and note velocity paradigm.

For strings, continuous parameters are things like bow speed, bow pressure, vibrato rate and depth, exact intonation, etc. It's not just a "set it and forget it" thing to figure out when sampling... The way they *change* is the part that people recognize as a violin sound. So maybe a sample out of context can sound great. But in a musical line, it matters a lot the timing of when vibrato kicks in, how the player should move the bow - what would sound continuous and natural is probably not what separate samples would sound like sequenced together. This problem still exists even if a library goes into deeply sampled legato and progressive vibrato and phase free velocity layers, even if they model up and down bowings and even if they go as far into modeling like AudioModeling. humans are just accustomed to hearing all the parameters being modulated in just the right way in a real performance - both physically realistic and musically appropriate. Without some next gen scripting or AI, there's no deep sampled instrument that could fill in those parameter decisions to fit the music and physical constraints yet.
 
Top Bottom