Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Forum Jump  
Legato Issues
Last post Thu, May 29 2008 by DG, 13 replies.
Options
Go to last post
Posted on Wed, May 28 2008 06:22
by rtaube02
Joined on Mon, Aug 20 2007, Sydney, Australia, Posts 7
Hello,

I having some issues with Appassionata Strings sounding legato. I know I'm probably not doing something correct. So any help would be appreciated. I looking for a really smooth fluid like legato to create really nice melodies. I have uploaded a clip I made using the violas. I have them playing a scale and then an arpeggio at the same velocity level. All notes are overlapped. But I'm still getting a slight bouncing feel to the legato. It doesn't seem to flow from one note to the other. I'm not necessarily looking for portamento, but just a little more of a slide or smearing sound, just so the transitions sound a little more seamless. If I can figure this out, it would help me create more lyrical phrases. Is there a way I can control the amount of slide between notes, so I can make the legato a little slower and not so choppy? Thanks
Download Ryan Taubert - Vienna Viola Scale
Posted on Wed, May 28 2008 08:49
by PaulR
Joined on Mon, Dec 22 2003, England, Posts 2370
Increase the release time on the edit page to around 60 to 75 and experiment with that.


Also, your examples are completely dry. I would imagine release time increase and a good reverb would make a big difference.
Posted on Wed, May 28 2008 14:44
by rtaube02
Joined on Mon, Aug 20 2007, Sydney, Australia, Posts 7
Thanks for the advice, but I've actually tried all of these things. I uploaded a completely raw track so you could clearly hear the attack I am talking about. I've done may compositions using longer release times and good reverbs. But it seems like nothing is really hiding the quick transition sounds. Even some portamento transitions seem somewhat violent and not sweet sounding. Is there any programming advice for getting more realism? I find that I will spend hours and hours on manipulating and editing the modulation to duck slightly at the start of each note, and to swell on certain notes to create a more lyrical sound. But it seems that I can't fix all of the transition. I feel like I'm spending way too much time on editing the modulations and not enough time on writing. Also I find that I have to change the way I write and some melodies to accommodate the samples and the transitions. If any of you can give me programming advice that would be great. I've even tried layering sustain samples with portamento samples. I lowered the volume of the portamento samples with the hopes of creating that slight transition noise underneath the sustain samples. Not sure if that made any sense, but I know that there has got to be something I'm missing. Overall I'm extremely pleased with the sound quality of the Appassionata Strings.....couldn't get any better. Thanks
Posted on Wed, May 28 2008 15:42
by jasensmith
Joined on Tue, Jan 15 2008, Arizona, Posts 1505

I'm not trying to insult your intelligence but I'll ask since you didn't mention it.  You did adjust the attacks and the release in the Perform View and "edit cell" in the Patch Assign?  Maybe I'm wrong but it seems to me that edits made in the "edit cell" are more coarse and profound then edits made in the Perform View's attack and release.

Also, did you try to use just the sustain patch with nothing else? 


"If at first you don't succeed, try, try again. Then quit. There's no point in being a damn fool about it."
- W.C. Fields
Posted on Wed, May 28 2008 16:50
by rtaube02
Joined on Mon, Aug 20 2007, Sydney, Australia, Posts 7
Yes I actually have tried that. I've tried just about everything possible. If I start reducing the attack time, the transitions seem to have a wah wah effect, which makes it even more separated sounding. Changing the release doesn't change the attack sound either, however it may somewhat mask the sound but it's still there....at least to my ear. I think maybe the issue is the fact that all of the strings at once are transitioning together almost perfectly, where as in a real live situation, during a passage, the transitions are slightly scattered among the strings creating an almost seamless transition. Maybe the answer is to layer appassionata with chamber or solo strings.....slightly off setting one of them. Here is a link to a piece that I wrote. Listen in particularly to "Reverie of Hope" when the strings come in at 00:31. Please excuse the mp3 sound quality. Anyways that is the best I've been able to get them to sound. But I spent so many hours on just the strings working with the modulations to get them to sound that way.

http://www.myspace.com/ryantaubert

Posted on Wed, May 28 2008 17:56
by garylionelli
Joined on Wed, Jul 09 2003, Los Angeles, Posts 376
IMO, VSL has worked wonders with this legato technology and it's helped my work immensely. However, we all know that if a real section played this passage, with a conductor, there would most likely be some emotion and expression which can't help but be missing from samples, even with this true legato. The legato here is unvarying, and while it is a huge step forward, it's still devoid of most of the feeling you hear with live players. Because of this, we tend to scrutinize the VSL performance, hoping to find some way to tweak it and make it approximate the sound we have in our heads, of real players. I don't think there's too much wrong with VSL's legato here, but rather all this other stuff coming into play.
Pro Tools Ultimate, Avid MTRX, Mac Pro 3.5 Ghz 6-Core (2014) 64 GB RAM, Mac OS Mojave, Samsung T5 SSDs
Posted on Wed, May 28 2008 18:57
by rtaube02
Joined on Mon, Aug 20 2007, Sydney, Australia, Posts 7
I completely agree....Vienna's legato is revolutionary in samples. Maybe I'm just trying too hard to get samples to do something which they just can't do at this point in time. I thought maybe there was some programming trick to simulate or mimic the aforementioned affect. I do realize when working with samples you have to do some sound design to get more realism. There is never anything like using real musicians. However, there really isn't anything wrong with VSL's legato other than the fact that it's still SAMPLES....BUT I am so glad that a company like VSL is always looking to be on the cutting edge of sampling technology. It will only get better and better....
Posted on Thu, May 29 2008 09:22
by PaulR
Joined on Mon, Dec 22 2003, England, Posts 2370
rtaube02 wrote:
I completely agree....Vienna's legato is revolutionary in samples. Maybe I'm just trying too hard to get samples to do something which they just can't do at this point in time.




I thought when I listened to your example mp3 that the only thing missing was maybe what I previously suggested - and perhaps some added dynamics. I know a lot of people like to 'draw' in expression etc.


One of the problems with forums is - it's sometimes difficult to assess what is actually being meant as opposed to being said. At the moment WE WILL NEVER GET ANYWHERE NEAR A REAL LEGATO PERFORMANCE FROM A REAL ORCHESTRA. If that is what you actually meant, then apologies because all the trickery in the world won't do that at the moment.


This is not really the purpose of sample libraries as far as I'm concerned. Samples are a series of little snapshots in time THAT WILL NEVER CHANGE - unlike a real performance - that will change every nanno second a phrase is being played - IN A REAL HALL WITH REAL PLAYERS. With samples if you stop to think about it - this will never happen. So you will always get a sample performance regardless of how good it sounds. And how GOOD it sounds is what it's all about. Samples are like making a picture out of loads of photographs all joined together. They look like the real scene - until you look out the window and see the shadows, sky, people etc actually all moving about. Photographs never change - playing in real time does.


It's not about legato transitions and all that sort of discussion.

Posted on Thu, May 29 2008 09:35
by DG
Joined on Wed, May 12 2004, Posts 8608

PaulR is quite right about real vs samples. However, there are ways to get much closer, but you have to understand a huge amount about string playing, such as:

  1. Bowing
  2. Different styles of playing within  the section (due to different teaching methods, instruments, bows etc.)
  3. Position changing and portamento
  4. Lack of tight ensemble
  5. Tuning issues of individual players
  6. Vibrato changing speed and amplitude during notes, and at different times for all players.

Now some of this is not possible with current sampling techniques, and some of it just takes too long, but if you think about this as you are working on your sequence, and do as many layerings as you are able to, mix those layers differently with each other (so that the sound is not always the same) you will get a lot closer.


DG

Nuendo 6.03, 4.3
2 x Intel Xeon x5675 3.07GHz Hex Core
48GB RAM
Windows 7 (x64)Pro
RME Multiface II
Intensity
ATI HD5400 series graphics card
Posted on Thu, May 29 2008 10:42
by PaulR
Joined on Mon, Dec 22 2003, England, Posts 2370
DG wrote:

PaulR is quite right about real vs samples. However, there are ways to get much closer, but you have to understand a huge amount about string playing, such as:

  1. Bowing
  2. Different styles of playing within  the section (due to different teaching methods, instruments, bows etc.)
  3. Position changing and portamento
  4. Lack of tight ensemble
  5. Tuning issues of individual players
  6. Vibrato changing speed and amplitude during notes, and at different times for all players.

Now some of this is not possible with current sampling techniques, and some of it just takes too long, but if you think about this as you are working on your sequence, and do as many layerings as you are able to, mix those layers differently with each other (so that the sound is not always the same) you will get a lot closer.

 

DG




(1) Bowing - the bowing will never change with a sample library. It is always fixed as it was recorded.

(2) is of course right - even though there is obviously different styles being recored for the sample library. The different styles are caught in a single snapshot and will never change.

(3) Correct.

(4) Correct. Clicks when recording and wanting to see everything neat and tidy on a computer screen does not help. If possible, a midi file of a real orchestra recording would look a complete mess.

(5) Tuning issues are present in a sample library - and of course never change and are locked in time on the recording forever.

(6) Vibrato and volume changes during the sample recording - but once recorded never changes. It will always be the same every time you play it. As of course will a DVD of an orchestral recording - but there is a massive difference between that and sample recording.


Daryl is right about all of this. Use as many layers as you want to make it sound GOOD. Not REAL>
Posted on Thu, May 29 2008 10:54
by DG
Joined on Wed, May 12 2004, Posts 8608

Paul, when I say bowing, I mean that thought must be given to when the bow changes in a phrase would occur, how to achieve this effect with samples, what happens to the previous note before the bow change etc.


FWIW I'm working on a little project where I'm going to A/B and orchestral mix with a sample mix of the sample piece, just so I can see how far away we are. I can't wait. Sleep


DG

Nuendo 6.03, 4.3
2 x Intel Xeon x5675 3.07GHz Hex Core
48GB RAM
Windows 7 (x64)Pro
RME Multiface II
Intensity
ATI HD5400 series graphics card
Posted on Thu, May 29 2008 11:57
by PaulR
Joined on Mon, Dec 22 2003, England, Posts 2370


Well of course you realize that I'm not even a Grade 1 violin player at this stage of my violin playing career:)) Although my tutor went to the same college as I did coincidentally - and he's nowhere near as big a snob as I am! :()()()())))


How are you going to A/B that Daryl? Just by using your ears or is there something technically obvious I'm missing? When I think of A/B'ing orchestral mixes, I immediately think of those EQ devices/programs you get in Cubase/Logic that allows you to do this after a fairly big learning curve.


Yours


Mr Writing Block (blamed on the weather and the fact I can't get out to play golf).


Posted on Thu, May 29 2008 14:44
by DG
Joined on Wed, May 12 2004, Posts 8608

Paul, for me it's more a question of performance than sound. One of the big things that I don't like about sample performances is the cleanliness of it all. I'm really trying to find out how dirty I have to make it in order to breathe life into the thing.


I also think that those matching EQ plugs are mostly a waste of time. They can possibly get you part of the way quicker, but you still need to use your ears. For example, with the reverb I sometimes use there are a couple of notes that really stick out with the cellos, but this is not fixed by EQing the entire mix; just the cello part. There may be a master EQ, but there are many more subtle things that need to be done first.


A/Bing will be done by putting the mixes on parallel tracks and then alternately soloing one, then the other. It's not meant to be a scientific test, just a little experiment to see which bits hold up in a musical sense, and which don't.



DG
Nuendo 6.03, 4.3
2 x Intel Xeon x5675 3.07GHz Hex Core
48GB RAM
Windows 7 (x64)Pro
RME Multiface II
Intensity
ATI HD5400 series graphics card
You cannot post new threads in this forum.
You cannot reply to threads in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.