Page 1 of 1
sample modeling? a gem I just found
Posted: Sun Jul 24, 2011 6:28 pm
by kensuguro
These guys have some sort of sample based modeling tech and the results are amazing.
http://www.samplemodeling.com/en/index.php
Check out their demos. They seriously have the coolest sounding demos of all instruments demos I've ever heard. Sounds like you can remix every single one of them.
But anyway, I think my Garritan big band is going to be replaced soon, or at least augmented with these guys.
http://www.youtube.com/watch?v=gtvmcRKs ... r_embedded
This guys amazing, and the sax sound is amazing too. I'm going to get a EWI usb too, and an aloha shirt, and maybe start wearing glasses too.
The brass from Wallander I think still has better brightness, seems sample modeling's sax wins hands down tho. The thing with sample modeling's stuff is kind of like why synful's strings is cool.. it seems to do things that real people do in playing, whereas wallander's just plays the notes you tell it to. But overall, Wallander's stuff has a more polished sound to it. Sample modeling's stuff is definitely focused on jazz.
Re: sample modeling? a gem I just found
Posted: Sun Jul 24, 2011 7:10 pm
by astroman
I almost don't dare to disturb your enthusiasm (and had some expectations after reading the modelling site...)
but the funky shirt is the best about the demo...

it's a perfect argument WHY you need a sax player if it's about a sax player
calling that performance 'articulation' is plain kidding - and it's NOT because of the shabby YT sound.
the names of the Italian fellows read somewhat familiar, my guess is that convolution fx processor...
introduced with equally big words, similiar disappointment (imho)
cheers, Tom
Re: sample modeling? a gem I just found
Posted: Sun Jul 24, 2011 8:24 pm
by kensuguro
wow, you are hard to impress! lol. Well, if the argument is anything vs having the real thing, I guess you really can't beat having the real thing. But, that's not really the argument. Again, for certain situations a real player is not an option, so my quest is to find an alternative which can create an ensemble that is able to sustain an illusion for a short period of time. The difference between having a semi-okay sax/horn vsti (though never as good as a real player), versus having nothing is huge. (as in, I'd have to compose around an intro sequence of a game to not use any brass)
You've got to admit though, there's nothing that sounds quite like what they've created in the vsti realm. I think the argument's similar to VSL or virtual orchestras. Sure, they're not as good as the real thing, but they're good enough in many situations. And in many cases, have become standard for production work because of their convenience. And they're still not as good as the real thing.
Re: sample modeling? a gem I just found
Posted: Mon Jul 25, 2011 12:08 am
by dante
I quite liked the demo that the keyboard player played of the Sax brothers. But you'd have to be just about as skilled on keyboard as the sax player is on sax to get that good a demo.
Re: sample modeling? a gem I just found
Posted: Mon Jul 25, 2011 1:57 am
by astroman
Ken, I've had those points of availability and 'better than nothing at all...' in mind.
I consider you much more educated in that domain than myself (only humble listening experience), but it's very obvious that the guy with the EWI is TRYING to drive the horn to a point where it simply cannot get.
It speaks for itself that the sax is one of few instruments that still resist a faithful modelling.
Imho opinion (because that's a live demo) the 'virtual' instrument does not perform as announced by the maker.
It's certainly tweakable via keyboard-controller stuff (possibly) for a better than average result, but (as the player shows) not out of the box. I lack the experience with libraries to tell how big the difference really is.
cheers, Tom
Re: sample modeling? a gem I just found
Posted: Mon Jul 25, 2011 8:07 am
by ChrisWerner
*Play* few milliseconds later, oh it is a sax, cool..
Nice playing isn't it?
Yes for sure, damn my beer is empty.
Oh listen, there is a trumpet supporting the sax.
yeah, did you see the girl over there?
Yes, I know her, she played the sax on this track.
Oh cool, she knows how to use her mouth.
A beer please!
The music is over, nice track isn't it?
Ah my beer... Yes the mouth, ah, sax is nice.
Serve what they want, it brings butter on your bread.
Re: sample modeling? a gem I just found
Posted: Mon Jul 25, 2011 9:56 am
by braincell
I use the EWI USB. I used the use a Yamaha WX5 breath controller but it broke. The EWI models have no moving parts. A disadvantage of the EWI is that for some reason, it has been designed so that you have to blow really hard into it. Everyone hates this. It is not something you can fix by turning a screw. It is a design flaw and because wind controller demand isn't high, R&D is very slow if at all for new models. I would go with the Eigenlabs Pico if I were you.
You might consider buying a real saxophone on craigslist. I imagine you can get a nice Yamaha for the price of this modeling software. I have been playing saxophone for 3 years and now, cello, trumpet and guitar and I'm not spending any more than I was on virtual instruments.
Re: sample modeling? a gem I just found
Posted: Mon Jul 25, 2011 2:23 pm
by kensuguro
hmm, brain, how would you compare the EWI to your sax? (what sax and what number reed?)
astro:
I do agree the sax is a bitch to model. Impossible to sample, really.. Same for horns, especially in the jazz idiom since the play style is much more organic and detailed than clean symphonic playing, not to mention smaller ensembles. (more detail required for each player)
Personally, I think what many sample libs are missing are the assisting technologies that Sample Modeling and Synful (VSL, etc) have taken to, that help interpret notes as a player would, modeling a player's interpretation as part of the modeling chain. It's very vital since it fills in gaps in the performance that otherwise would take hours and days to tweak, or otherwise impossible to achieve with discrete samples. The interpretation is of course not right on, but it substantially livens up the track. Even simple things like auto switching between staccato and legato using note proximity helps a lot.
Re: sample modeling? a gem I just found
Posted: Mon Jul 25, 2011 3:43 pm
by astroman
yeah, it's likely faster to program than a standard library or tweak stuff with Melodyne like tools.
but I have to withdraw my criticism completely...
... at that price I might even get the Alto and Bariton myself

just for fooling around with the Casio Digital Horn (about which a fellow 'real sax' player said he preferred it over any other wind controller...

)
sorry, I had expected that stuff to be way more expensive
cheers, Tom
Re: sample modeling? a gem I just found
Posted: Mon Jul 25, 2011 4:15 pm
by braincell
I use the EWI for legato flute and legato oboe. In the VSL Special Edition they have a section for mapping the expression curve, also I use Cubase controllers to map breath to volume. Haven't used it in a while but it worked well. They say the standard fingering is easier than the sax fingering even if you know sax fingering already. Mostly what I do is based on improvisation and the nice thing about midi is you can edit and fix that easily.
I worked my way up to a number 3.5 Alexander classique reed. Most people start with a 2. The new synthetic reeds are easier but most agree don't sound as good. Saxontheweb.net has all the info you need. You can get so many sounds out of a sax, a sample library or physical modeling will always sound generic. The way your mouth works on a saxophone is so complicated, it's not just how hard you are blowing. Same with the bow on a string etc.. Try borrowing or renting an instrument. You might like it.
Just checked, I was afraid I lost my chops because I haven't played in a couple weeks but I'm no worse, maybe better for it.
The cello I started because I was paying a cellist to make samples in a good studio, that got expensive. Now I'm a cellist (rent to own), not as good as him but able to put my personality into it. They make some nice cellos in China now if you have the bridge and pegs replaced.
Re: sample modeling? a gem I just found
Posted: Mon Jul 25, 2011 5:40 pm
by kensuguro
sry my question wasn't clear. How would you compare the amount of blowing needed for the EWI versus the 3 1/2 reed on your sax? (alto? tenor?) I assume embrasure pressure doesn't really affect tonality in the EWI.
Re: sample modeling? a gem I just found
Posted: Mon Jul 25, 2011 5:57 pm
by braincell
kensuguro wrote:sry my question wasn't clear. How would you compare the amount of blowing needed for the EWI versus the 3 1/2 reed on your sax? (alto? tenor?) I assume embrasure pressure doesn't really affect tonality in the EWI.
With reeds we are talking about embouchure muscle and with the EWI it is lungs only but generally speaking, the EWI takes much more effort than a light reed to blow. The Yamaha breath controllers are easy blowing.
Re: sample modeling? a gem I just found
Posted: Mon Jul 25, 2011 9:50 pm
by kensuguro
Fxk it, i just ordered the ewi. Lol
Ive played all sorts of woodwind instruments so im hoping its an easy switch. Longest was on clarinet, but strangely clarinet fingering isnt an option for ewi. On the ewi i can pick Flute, oboe, sax or ewi fingering it says. I guess clarinet has too many hinges and crap to translate to ewi.
Last i saw my clarinet was when i sent it to yamaha for refurbish, and i left it there for 5 years. Now i feel bad about it. Lol
I also got the sax brothers. Quite intensive to load a aattb full section but the sound is amazing even using just mod wheel. I have never heard any sax patch that can slur notes like the sax bro. It can also do strong tonging, which many sample guys just dont understand the importance of. Itll do funk and bigband just fine. Supercedes garritan jazz lib and too many ways.
Re: sample modeling? a gem I just found
Posted: Tue Jul 26, 2011 3:53 am
by braincell
The importance of tonguing depends on your style. Unusually these days, it's just used to play faster which requires a lot of practice to make your tongue move like that. I don't think you really hear it in the tone unless you are talking about 1950's rock sax. I hate that style.
Re: sample modeling? a gem I just found
Posted: Tue Jul 26, 2011 6:15 am
by kensuguro
I was thinking more in terms of funk arrangements for chord stabs, esp when there's a powerful trumpet section. I like a strong tongued attack so the hit overall (1/2 sax, 1/2 horns) has bursting energy. With Garritan, the sax section in particular is very smooth (great tone for relaxed stuff) so it's easy to be over powered by 2 or even 1 trumpet on a hit.
You talking about 1950's over the top growling tenors? Ya, it's excessive, and once the entire solo / play style becomes constantly over the top, the dynamics is gone. Can't ever be that "limelight grabbing" in an ensemble setting anyway.
Re: sample modeling? a gem I just found
Posted: Tue Jul 26, 2011 6:21 am
by ChrisWerner
How about Arturias Brass 2.0 ? I dunno?
Re: sample modeling? a gem I just found
Posted: Tue Jul 26, 2011 6:47 am
by kensuguro
just checked the demos... It seems more versatile than wallander brass in terms of controllability, but it still has that distinct physmod sound to it. The attack on most physmod stuff is very similar, it doesn't have that ripp and explode attack great for ensembles. Apart from the attack though, it sounds like it's one of the more versatile physmod horns.