Angelo, what would be a "Standard work every engineer knows" in english be? Do you have recommendations of a book to buy? Thanks.

Colin Thomson 

@Angelo Clematide said:

Okay, let’s assume I would write down this level values for each instrument, for example from Hans Werner Henze’s Sinfonie Nr. 1. At what points should I measure the level, and how would you apply the collected data to your music?

 

@Felix Bartelt said:

Quite easy: Imagine, you are Hans Werner Henze, standing in front of the orchestra, ready to start playing. Then U say: "Stop! I want every single instrument or section (such as the HO-4) play its loudest note allone!" Result: If the VI-16_mV play the loudest possible G3 at - 13 db, the VA-10_mV reach a peak of - 16 db, the single horn reaches - 20 db and so on. If U transfer this situation to your setup: U have everything set to start composing (all volumes, reverbs, pans, etc.). Even if U plan to automate your volume faders afterwards, it is still a setup, U consider to be a good mix to start with, no matter what velocity the single instruments might play later. So as soon as you are confident with your mix, stop it at any time of your work to check, what maximum peak every single instrument is able to play. Then all people like me would have something very good to start with, because I would just have to adjust the volume levels. In my situation, not knowing the relative levels, I might set the HO-4 volume fader too loud and later on balance that in a false way by letting them play mf instead of ff. Do you see my point?

A-HA - We are not talking about the same matter.

I see now what you mean, but that is what we call absolute dynamic (Felix sez: "play its loudest note"). But there is a difference between absolute and relative dynamic.

Relative dynamic of a musical instrument is the total loudness range. For example measured over the full range, and each step played as soft as possible plus as loud as possible. This aquired dynamic data can be presented in envelope graph (for example as in Jürgen Meie book "Akustik und musikalische Aufführungspraxis"). This data helps the engineer to place the microphone at the proper spot in various recording situations.

There are several reasons why relative and absolute dynamic data would be unsatisfying in build a dynamic orchestra default setting. One of the main reasons is what is called "distant cue in closed rooms" - this is the spatial information stored in a stereophonic recoding who tells us how far away the source is.

I agree with what Felix is asking for. I don't think Angelo has this information though. If he did he would have simply posted it by now. The debates about WHY this information wouldn't be accurate or helpful are drifting further into obscurity. If anyone does have this information though please share it with us. It would be enormously helpful as many have pointed out.

Thanks!

Rob

@CommanderFunk_28576 said:

I agree with what Felix is asking for. I don't think Angelo has this information though. If he did he would have simply posted it by now. The debates about WHY this information wouldn't be accurate or helpful are drifting further into obscurity. If anyone does have this information though please share it with us. It would be enormously helpful as many have pointed out.

Thanks!

Rob

No, I don't have this information. I think nobody ever made this measurements.

I guess what you want is, that some one writes down the values from the software meter for each orchestra instrument playing the loudest possible sound when the recording setup is one stereo microphone in a concert hall. The measurement in the mentioned book "Akustik und musikalische Aufführungspraxis" is not made in that way, but each instrument is measured from same distance, similar as the VSL samples are made.

.

Hey Angelo,

Those measurements may be useful though as well. VSL was normalized so your data would be very helpful. You could calculate the diff. and let the panning/narrowing reverb etc take care of the rest. It would require more tweaking but what you describes sounds like it would be great!

Thanks again!

Rob

"The debates about WHY this information wouldn't be accurate or helpful are drifting further into obscurity."

Oh, No!  We can't have that!!

Ahem, to illuminate:

Two reasons why a standardized environment is moot:

1.  Every mix is different.  There is not one environment.  There is not one method of recording an orchestra.  There is not one method of mic'ing an orchestra.  There is not one method of playing each instrument.  There are as many dynamic ranges of an instument or ensemble as there are players and ensembles.

2.  Sample mockups with VSL are especially variable (and customizable=this is GOOD) because of the normalized, silent stage process.  Why limit yourself by Standardizing (bastardizing) variables, thus limiting the potential?

CAVEAT:

This does not mean that you can't have a "starting point."  For me, this means setting up general levels at -15 to -20 (in Logic), general pannng, general reverb type and level, etc.

The only other exception is if you want to emulate a particular recording.  To which I say, Listen, Listen, Listen.  Take the time to get the same reverb, e.q., DYNAMIC RANGE, and everything else.  Then make a template from that that says "John Williams Jaws Temp" or whatever.

Liberty and Freedom for all!

Clark

Think about it: Dynamic Range is a non-issue.  Every recording ever made has been subjected to (tape or channel) compression, mixing of near and far mics, soloist gets their mic raised for eight bars then ducked down again.  Newer recordings are 20-100 tracks.  Probably all automated and compressed with extra reverb, e.q., EVERYTHING.

If you don't have the ears or arranging sensabilities to understand how instruments blend or behave together you need to go back to school.

Clark

To be honest, this discussion about having the startup levels being useful or not, let´s me think, that there are other reasons for not telling the levels. Don´t get me wrong! I think it´s absolutely o.k. to keep some mixing secrets and most of you guys gave more hints, I expected to get.

But: Please don´t come up with the same arguments like, every mix is different or listen to orchestrations yourself!

1. Not all instruments of VSL have the right level and the right level relation to eachother out of the box. (Example: Solo violin compared to VI-16). So every instrument may be changed in master output level, may it be, by using the volume fader, by changing the position in altiverb, by EQ, whatever. Fact is, that in your master output not all of your VSL tracks will have the same relation in peak to eachother as they had directly after being loaded out of the box into your audio instrument track.

2. The relation of the level peaks may be different from song to song. I don´t care, cause I don´t ask for every level setting of every type of orchestral setup. I´d just like to have something to start with. So just take a normal orchestral song like the one Angelo took to show the dynamics as our example setup song. Then check the level peaks, so we beginners know, that these peaks in that kind of music would be a good start. Or post any song and tell the peaks of that song.

3. No matter if a template is useful or not: I don´t want to have the perfect template. I just need something to start with. Once again: I can get two instruments in balance. But I might be doing it wrong cause I chose the wrong start volume levels. I can make a pp VI-16 sound as loud as a full ff chord of the whole rest of the orchestra, if I chose a very high volume level for the VI-16 and very low levels for all other instruments.

And I hope U agree that I would get a wrong timbre, because of chosing wrong level peak relations.

So how can one doubt, that this information would help!?!

4. It´s not right, that no one hast this information. Actually in my actual orchestal song, it would take me about two minutes to get it:

I´d just have to take a mix, that I like, click thru the single instruments, play the loudest note and write down the volume peak of every instrument at my master output. (Once again: It would not matter to me, how you got to this peak and if instrument x has a lower peak than others because instruments have different reverb settings.It´s just one thing, helping a lot.)

I hope, this time I explaind better, what I would like to get hints about, since Angelo and I seemed to be talking about different things.

Thanks for your help anyway.

@ColinThomson said:

Angelo, what would be a "Standard work every engineer knows" in english be? Do you have recommendations of a book to buy? Thanks.

Colin Thomson 

There was a book we called the "Studio Bible," published in the seventies, I lend my copy I bought at the Opamp Technical Bookstore in Hollywood to some r&b producer and never got it back, the titel of the book was:

"The Recording Studio Handbook"

by John Woram

Hardcover, 1976

Plus

"Microphone Handbook"

by John Eargle

Elar Publishing, Plainview NY

"The Cameo Dictionary of Creative Audio Terms"

Published by Framingham. MA

Paperback, 1979

...of course all lend out to some r&b producers and bnever got them back.

 Clark, interesting that you say you use lots of volume automation. I always thought this would take away from the overall realisticness of the VI samples. They are sampled with crossfades from PP to FF (or so). What is the use of automation? If you bring the volume up on a mp sample to bring it to around F, won't it sound strange to have something playing at that volume in the mix, yet with the sonic quality of what would normally be quieter? What is the use of volume automation (with VSL in mocking up a real orchestra), and how and why do you use it?

Angelo, thanks for those book recommendations. I will take a look at them.

Colin Thomson