This is excellent - at least as good as many of the user demos.

"Shades of quiet" - yes. That is why I love Ravel and Debussy. I am so tired of heavy, loud, aggressive, conflict-filled music. I used to write it and can't anymore. So I am extremely attracted to the impressionists who were above all the conflict, depicting states of pure being and emotion. Also - what a melody! If you can write a melody like that you've got it made.

On this performance I have these criticisms which are only small points. The reason I bring them up is because your recording is so good:

maybe there needs to be some EQ in the violins, because they are a little bright. The reverb could also be a darker hall in keeping with the mellowness of the music.

I think you are making the solo instruments a bit too loud, because you are so concerned (as you should be) with making each one audible. I worry about doing this myself, and making everything too clearly audible. But related to this is the dynamic range. I had the reaction that the dynamic range on the solo instruments was too large, compared to the ensemble strings which sounded too thin in the quieter passages. In other words, the ensembles should retain their size, but be hushed, as opposed to thinner in the pp. The solos sound "huge" in comparison to what you might expect, as if they are each close miked in a live orchestral recording, which is not as natural obviously.

One thing that could be tried - and I wonder if Dietz or anyone agrees with one I'm saying here - is to use some compression on the solos, so that they do not become overly prominent. The orchestration by Ravel is so masterful that most of the lines stand out without being emphasized.

I have one question - the dynamic shading both beginning and ending of lines is extremely good and natural sounding all throughout the piece. This is one of the major pitfalls of sample performances - having notes just cut on and off at the same volume like an organ. You avoid this, and I'm wondering if you used dynamic change samples, cross fade dynamic layers, filter mod, controller 7 changes, velocity envelopes in your mix, or what to accomplish it. Or maybe a combination of all of them?

This performance is not only shades of quiet, but shades of legato - an excellent demonstration of the fluidity and smoothness of the legato instruments. Congratulations on a fine job.

Re: Compresssion - agreed! :-]

/Dietz - Vienna Symphonic Library

William—

Thank you again for your insightful comments. I agree with them all.

Taking the “edge” off strings without losing their “aliveness” and “texture” remains an elusive goal for me. I actually did EQ them with a gentle low pass filter, but I was afraid to attenuate too many of the top frequencies, so maybe I didn’t go far enough. Everyone always says “use your ears” and this is good advice, but I’d also love to hear about some specific solutions and EQ settings people are using.

The reverb here is the Lexicon Pantheon Medium Hall that comes with Sonar 3. I tried to simulate depth by making the horns “wetter” and the strings somewhat “dryer.” But how does one create a “darker” reverb sound? Which reverb settings do I fiddle with?

I was indeed obsessed with making the solos stand out and probably went overboard. And because I wasn’t totally happy with the timbre of the strings I probably subconsciously tried to “hide” them by bringing them down too much in the mix and thus further exposing the solo woodwinds and horns. Also maybe I should have placed the instruments in a larger hall, but Ravel states this piece is “pour petit orchestre” so I thought that maybe I should strive for a somewhat smaller and more intimate sound. But unfortunately the Pantheon Medium Hall reverb that I did use is a little “metallic” for my taste.

I spent gobs of time on the dynamic shading, so I’m glad you noticed it. I’m using a combination of modwheel crossfading between dynamic layers (including the custom MONO modwheel crossfade performance patches I posted in the Mother Goose thread) and further shaping of the sound (and volume) with expression controller #11. Every phrase (and almost every note) is shaped in some way. I have found that for most samples unless they are moving dynamically in some direction, they almost immediately start sounding extremely artificial. This is doubly true for ensemble strings, especially at the beginning and ending of phrases and during all longer sustains. When I get lazy and don’t do this, my wife inevitably says the samples start sounding like a “squeezebox.” Or she’s says it’s as if she can “hear my fingers pressing down the keys and triggering the samples.”

If anyone’s interested I can post the midi (or Cakewalk) file so they can see exactly what’s going on.

Thanks again for taking the time to give your feedback.

Best,
Jay

Hi Dietz—

Could you elaborate a bit on the proper use of compression in this context (or any context—I’m quite ignorant about this)? It’s been my very limited experience that compression can sometimes make certain instruments actually sound even more prominent (louder) in a mix. What sort of settings would you recommend I use to “tame” the solo parts so they don’t sound too large in comparison to the strings? If there’s already a thread on this subject could you post a link?

Many thanks,
Jay

Jay, this is simply very beautiful. GREAT JOB!

Your dynamic perfomance is outstanding. It would be great to have a look in your midi file.

What made you choose GOS for sordino strings?

I understand your point about hiding the violins. EQ-wise try to dampen the trebles a bit (beyond say 8 kHz) and push a bit around 400 Hz, maybe this helps a bit. What articulations did you use for violins? Sometimes I thought a more kind of sfz attack could help. Perfomance articulations are not always the best choice, I found.

The Panteon is really flexible reverb and there are several possibilities to get a darker sound, most common of course the locut filter (or hidamp). This simply cuts of trebles. But the interesting part of this reverb begins when you fiddle around with density and spread parameters and predelays. Did you read the manual? I found it quite informative. Especially that metallic character you should get rid of with adjusting these parameters.
Instead of just send less violins in the reverb try different reverbs for the different instrument groups, so don´t be shy with Panteon instances. I sometimes have altogether seven or eight reverbs and ambience generators running.

Again, thank you very much for posting it, it was a real pleasure listening to it!
Bests,
- Mathis

@JBacal said:

Thanks for taking the time to listen. I hope you enjoy it. Best,
Jay

Very enjoyable to listen to. Must take you ages to do something like this Jay! Thank you for sharing this. Well done! [:)]

@JBacal said:

Hi Dietz?

Could you elaborate a bit on the proper use of compression in this context (or any context?I?m quite ignorant about this)? It?s been my very limited experience that compression can sometimes make certain instruments actually sound even more prominent (louder) in a mix. What sort of settings would you recommend I use to ?tame? the solo parts so they don?t sound too large in comparison to the strings? If there?s already a thread on this subject could you post a link?

Many thanks,
Jay

Jay, our Mixing- And PostPro-Forum should contain quite a few threads on this (and related) topics.

HTH,

/Dietz - Vienna Symphonic Library

Mathis—

Thanks for your tips on EQ and reverb. I will read the Pantheon manual and try fooling around with the controls you suggest.

Originally, I wanted to use only VSL products, but the con sordino strings in Opus 1 have a lot of “detail” and the violas have a lot of high-pitched overtones that I didn’t want. There is probably a way to EQ them to create a smoother and softer sound, but unfortunately I don’t really have the mixing skills yet. The GOS con sordino strings were closer to what I wanted out of the box, so I went with them.

I agree that I should have tried using more articulations for the strings. This was partially out of laziness and partially because con sordino strings in the libraries I own have a limited number of articulations. Also I only have 1 computer and 768 MB of ram so I generally try to limit the number of articulations I use so that I can hear as many of the instruments simultaneously as possible. This makes balancing the instruments against each other a lot easier. I created some pseudo sfx articulations with CC #11, but I know they’re a poor substitute for the real thing. Next time I’ll go the extra mile and use more articulations.

I’ll try to post the midi file soon.

Best,
Jay

@JBacal said:

If anyone’s interested I can post the midi (or Cakewalk) file so they can see exactly what’s going on.

I, for one, would love to see the MIDI file, as I am still in the early stages of trying to learn how to use VSL.

Best DG

Here's the midi file.

http://hometown.aol.com/jbacal/music/Pavane.mid">http://hometown.aol.com/jbacal/music/Pavane.mid

If anyone has questions about it, don't hesitate.


--Jay

@JBacal said:

Here's the midi file.

http://hometown.aol.com/jbacal/music/Pavane.mid">http://hometown.aol.com/jbacal/music/Pavane.mid

If anyone has questions about it, don't hesitate.


--Jay

Thanks; I'll have a look at it over the weekend, when everyone else is enjoying the sun [:(]

DG

NOTE: Almost all patches in Pavane mock-up are multilayer multi-dynamic modwheel crossfade legato performance instruments (except for harp, pizz strings and string harmonics). Please see Mother Goose thread for details about creating many of these multi-dynamic modwheel performance patches.


--Jay

Very good indeed Jay, very well put together. You beat me at it ! [:@] [:D]
I've started a version of that beautiful piece myself a while back as a pet project to learn the VSL and get inside Ravel's orchestration and writing. I've only done the first 20 bars so far...

A few minor comments in terms of accuracy with the score. I find the solo horn in the opening a bit too present. it should be more delicate. It's marked pp and in your version it sounds more like mf. It's not only a mix thing it's also the timbre itself. Not sure if you have a pp solo horn on the Opus.
The very first harp gliss seems a bit late to me. Overall your tempi seem a tad too slow. Don't take me wrong your rendition is very musical and sounds fine but these are a few things I noticed that may not be faithful to Ravel's score.

You are now in the club of the distinguished VSL masters!

=m=a=r=c=

Hi Marc—

You’re correct about the French horn. It’s a custom modwheel combination of pand mf legato performance layers. I felt that adding the mf layer made the passage more expressive and more “French horn-like.” I probably went too far.

The tempo is indeed in large part slower than the score indicates. But William encouraged me to take more liberties with tempi, and so I did. I’ve actually heard recordings that are even much slower than this. It’s hard to know what’s “right.”

I’m not too knowledgeable about the realization of harp glissandi. I’ll have to listen more closely to my actual recordings and hear how a real harpist does this.

Most importantly, please finish your version and post it! I for one would be very interested to hear someone else’s solutions to the same technical problems as well as someone else’s musical interpretation of this gorgeous music. I think this would be very interesting and educational. And besides, you can never hear too many versions of this beautiful piece!

Best,
Jay

I chose that piece because it was short! It still feels like a gigantic tasks to do a midi mockup of it with the VSL. I' ll see if I can do a few bars every day, it will stll take me a while. How much time do you think you spent on it?
In any case I really doubt I can do as good as you did. [:)]

=m=a=r=c=

Marc B,

I worked on the piece off and on over a period of about 10 days. Hopefully, future pieces will go faster for me with more experience. And if I win the lottery, a second or third computer will also help. For example, I would balance the woodwinds and then "lock" them in by creating an audio track. But then after hearing them in context with the strings I would want to change something about the woodwinds or change the tempo and so I would have to redo the ALL the woodwinds again and then create a new audio track. This was frustrating and time consuming. If I could keep all the instruments and articulations in midi until the final mix, this would save a lot of time and yield a better end result as well.


--Jay

Marc,

I just relistened to the first harp gliss and you're right, it is more than late, it's in the wrong place! The gliss must have gotten accidentally shifted a couple of beats in my sequencer. Oops! I'll have to fix that. Thanks for noticing.

--Jay

No worries Jay, it did not sound at the wrong place because you would have heard it. I felt bad pointing those minor points when the whole was so masterfully put together. Glad you did welcome my feedback!

=m=a=r=c=

Wow! This demo is beautiful! I told alot of my freinds to visit this site to listen to this demo and the 'La Mer' demo. Just amazing what the VSL can do, isn't it! My freinds listened to these demos and some became discouraged that they were going to lose their jobs as musicians to this technology. I don't think that will happen though,
-Julie

Jay-
Sorry to trouble you, but I am trying to learn from your demo. I opened it in Cubase and see a bunch of instruments that I don't know. What is 'solo double', and 'ambient'? Will you give a list of the 'score order' of your mockup? I would like to know exactly what you are doing here so there are no secrets. [:)] It is really so good, and would help me a great deal. But, I also understand that you are busy and may not be able to do this. Either way, thanks very much for the demo and midi file.
Julie