@JohnnyMarks said:

Felix and others, "Intensities of Orchestral Instrument Scales Played at Prescribed Dynamic Markings" is a 1964 MIT paper, principal author William Strawn, available from AES online. I've uploaded a couple snippets to: http://homepage.mac.com/johnnymarks Useful data and interesting observations in the paper. Worth the $20 download. Cheers.

Nice find, thanks.

Before I make myself the stress to explain any further digital audio production specifications, I post a few links for my distinguished friends here.

Below some links to papers which deal with the subject of mastering, signal overload, transcoding etc.:

[B]Overload in Signal Conversion[/B]

A paper about the headroom needed in processing, routing and reproduction of digital signals to prevent distortion and listening fatigue from happening. It includes a discussion of AD, DA, sample rate conversion and data reduction systems.

by Soren Nielsen & Thomas Lund

[url]http://www.tcelectronic.com/media/nielsen_lund_2003_overload.pdf[/url]

[B]Mastering and Formats - Stop Counting Samples[/B]

Report from the frontline of the CD Loudness war, 2006: It's not over yet. Read about distortion developing in CD players, data reduction systems etc. The Appendix suggests improved procedures to follow in production and mastering.

by Thomas Lund

[url]http://www.tcelectronic.com/media/lund_2006_stop_counting_samples_aes121.pdf[/url]

[B]0 dBFS+ Levels in Digital Mastering[/B]

This paper examines the sonic consequences when 0dBFS+ signals are reproduced in typical consumer equipment. The performance of a variety of domestic CD players exposed to such signals are presented and evaluated.

by Soren Nielsen & Thomas Lund

[url]http://www.tcelectronic.com/media/nielsen_lund_2000_0dbfs_le.pdf[/url]

[B]Level Control in Digital Mastering[/B]

Music mastering is becoming a battle for maximum level rather than a quest for audio quality, because counting consecutive samples at OdBFS is not an adequate restriction of level.

by Soren Nielsen & Thomas Lund

[url]http://www.tcelectronic.com/media/nielsen_lund_1999_level_co.pdf[/url]

[B]The Secret of the Mastering Engineer[/B]

Mastering is an art and a science. In this acclaimed booklet, Bob Katz shares good advice about monitoring, metering and processing. About listening to the music and supporting it as the road to Nirvana - from one of the true yogis of our industry.

by Bob Katz

[url]http://www.tcelectronic.com/media/katz_1999_secret_mastering.pdf[/url]

cont.

Level Practices

Part I: The 20th Century Dealing With Peaks

http://www.digido.com/bob-katz/level-practices-part-1.html

Part II: How To Make Better Recordings in the 21st Century - An Integrated Approach to Metering, Monitoring, and Leveling Practices

http://www.digido.com/bob-katz/level-practices-part-2-includes-the-k-system.html

.

Thanks very much, Angelo. I look forward to reading through those papers.

Colin Thomson 

@ColinThomson said:

Thanks very much, Angelo. I look forward to reading through those papers.

Colin Thomson 

You are welcome.

The three K-System meter scales are named K-20, K-14 and K-12.

Wide Range Music (K-20)

The K-20 meter is intended for wide dynamic range material, e.g., large theatre mixes, audiophile music, classical (symphonic) music, "audiophile" pop music mixed in 5.1 surround, and so on.

Pop Music (K14)

The K-14 meter is for the vast majority of moderately-compressed high-fidelity productions intended for home listening, e.g. pop, folk, and rock music.

Broadcast (K-12)

The K-12 meter is for productions to be dedicated for broadcast.

[invalid link removed - /Dietz]

Production Techniques with the K-System

First choose one of the three meters based on the intended application. Wide dynamic range material probably requires K-20 and medium range material K-14. Then, calibrate the monitor gain where 0dB on the meter yields 83 dB SPL (per channel, C-Weighted, slow speed). 0dB always represents the same calibrated SPL on all three scales, unifying production practices worldwide. The K-system is not just a meter scale, it is an integrated system tied to monitoring gain.

I have a simple question -

Why is this thread a sticky?  It is completely inaccurate information. 

[edited - /Dietz]

This thread is sticky because the _discussion_ of this topic is interesting. To allow for a healthy debate doesn't mean that everybody (or _anybody_) shares the opinions presented.

...  there' s a message from me at the very beginning of this thread pointing towards the critiques, making very clear that there is a lot to be said against the proposed system.

Kind regards,

 Thank you Dietz.

I will attempt to explain then what my specific objections are, simply as discussion -

It has to do with the nature of both conducting/performing/recording orchestral music and sampling. 

If you are present at a concert, you will notice many examples of dynamic ranges being altered by both the conductor and the player.  That statement will immediately raise eyebrows, as people will immediately think "How could a person change the dynamic range of his instrument?"  It is done every day in the orchestra.  Because a certain pp oboe note will be played completely differently in a Mozart symphony as compared to a Bruckner -  perhaps at mf, though articulated and phrased at pp.  They are two different entities both acoustically and musically. This cannot be gauged according to any numerical system, but is entirely subjective.  And yet it is done all the time. And that is only one minor alteration.  

As soon as you start talking about recording a live orchestra, everything is thrown out concerning any natural or standard dynamic range, because individual instruments are always altered - sometimes to the point of having absolutely no relationship to their real (or normal) dynamic range.  A solo flute will be brought way up, tympani will be brought way down,  overall levels will be compressed, etc. etc.  In the 1950s there was a purist approach in classical recordings sometimes used,  with a single perfectly placed microphone and nothing altered.  This is of course not at all the case now.  Anyone does, and can do, anything to make the final recording sound good.  And this includes totally distorting all semblance of standardized or naturalistic dynamic ranges.

And the third element which seals the doom of any standardized approach is SAMPLING.  Even more profound distortion of any relationship in dynamic ranges of orchestral instruments is introduced by sampling.  A tiny piccolo has the same range as a huge set of orchestral percussion in a sampled recording.  A huge violin section is indisitnguisable in level from a solo oboe.  This is of course necessary in recording, mainly to avoid increased noise levels.  But when these samples are translated into a performance of actual music, the performer must make enormous changes constantly in order to equalize these huge distortions of level and range.  This cannot be done by any theoretical system, because of the complexity of interaction between these different  aspects of sound.  It can only be done by listening in detail to each part, getting them right individually, and then comparing them to each other as they are added into the mix, and then adjusting overall sections in relation to each other, and then finally adding compression - usually in woodwinds, perhaps violas and basses - and further tweaking the entire mix.   In other words, there is no substitute for simply knowing how the instruments sound individually, and in relation to each other.  You must be able to FEEL the difference between a trumpet playing an ff solo over a section of strings that are also playing loudly but the soloist is making sure he is still heard loudly even when the other instruments threaten to obliterate him.  All of this is immensely and absolutely subjective and can NEVER be turned into a paint-by-numbers system.  Added onto this series of variations which reduces to meaninglessness any numerical system is the fact that different articulations in a matrix have equal dynamic ranges that may be accurate within themselves but inaccurate in relation to each other - for example, you will notice that a smooth sustain multi-velocity sample set will be rather loud in relation to a double sforzando, and so these must be adjusted relative to one another IN THE MATRIX - further complicating an already bewildering process. When faced with this sort of complexity, it become absolutely essential to apply artistic and subjective judgement - as the basic procedure, not an afterthought - in order to make sense at an intuitive level of something that is essentially an artistic process to begin with: music

I particularly object to the repeated statements here that every studio in the world uses this system and has done so for a long time as a normal professional standard. That is absolutely false.  No studio in the world has done sample performance of complex orchestral music that equals what is RIGHT NOW being done for the first time with VSL.  The demos you hear on this site are unprecedented in their complexity, expressiveness and realism, and in fact are setting a standard that other performers and studios are trying to figure out how to emulate.

Yes, what you are saying is true Tanuj.    I would add one other thing, besides emphasizing that no one is using this system universally because no one has had these samples universally until the last few years  !!!   

you must FEEL dynamic ranges.  You cannot program them.  You must know in your soul how to create the correct balance.  It must become an intuitive process, by becoming more and more familiar with the sounds you have available.  You must DREAM of the correct dynamic range and change your mix in the morning.

I have been using VSL for some time, now, but the awe-inspiring potential of this arsenal of sounds has not even been scratched yet.  And in order to properly use it, you must incorporate into your intuition as well as your conscious, linear, simplistic numerical thoughts the way each sampled instrument sounds in relation to the other, based on what you have learned of the musical instrument known as the Vienna Symphonic Library.  For example, I am only now beginning to realize the significance of the difference between strong, medium and light dynamic articulations in how they can be used for artistic rather than merely realistic purposes.  The light and the pfp are especially significant because they are not just for crescendo/diminuendo but are extremely espressivo.  I remember Herb noting how he used on the great "Parachutes" demo - which originally sold many, many copies of VSL -   the dynamic articulations on brass because the notes on those particular samples sounded like "they were going somewhere." This is a very important point, but one that can never be put into a numerical system that uses a mechanized or formulaic basis for how to control the single most expressively crucial element in musical performance - dynamics and dynamic relationships between instruments. 

The subject of this thread is about headroom, calibration of the studio monitoring, reference levels, the utilization of the available data width on the digital recording medium, and how to achieve the optimal dynamics for the particular medium. In this sense there preexist several standards in recording technology, e.g maximum level for films, total dynamic ranges for the different media etc..

This thread is not about composing, and not about how the composer handles dynamics and loudness relations between instruments, or groups of instruments within his composition.

The recording of a real orchestra is another art form. However, I think it is possible to transfer parameters from the reality of recording a real orchestra into the virtual reality of computer generated orchestra music, for example room and distance parameters, but then everything has to be programmed.

"how to achieve the optimal dynamics for the particular medium - Angelo Clematide

How can you do this in samples without addressing elements that arise in composition, orchestration and performance?  Try it.  You will discover it is impossible.

Also, your discussion begins with adjustments of levels of  different instruments, and that is very definitely of musical, compositional signifcance.

Besides, it is totally unnecessary to use this system for studio equipment calibration. If one owns or works with equipment, one should have already learned long ago how to use it properly unless one is an idiot.  Though your system may be of use to you, and I am not trying to deny it's validity for your purposes if you want to work that way.  But this thread implies that this is the only way anyone professional should work, and that is flatly wrong.

@Another User said:

Besides, it is totally unnecessary to use this system for studio equipment calibration. If one owns or works with equipment, one should have already learned long ago how to use it properly unless one is an idiot.  Though your system may be of use to you, and I am not trying to deny it's validity for your purposes if you want to work that way.  But this thread implies that this is the only way anyone professional should work, and that is flatly wrong.

Every recording studio is calibrated, as well every certified cinema theater is also calibrated. When the engineer and mixer would work in a non-calibrated environment, he would not know how loud the music is.

"A composer knows in advance where the loudest part of his composition is, and from there the available headroom, respectively the maximum loudness is set."

This shows a complete ignorance of performing and recording MIDI music.  The loudest or softest notated or sequenced scoring might be far lower or higher in overall level depending on which samples are used and how they are processed.

"Level is recording technique terminology. For musical loudness we use the usual Italian terminology such a ppp or fff"

You still do not understand. A written ppp may sound higher in level than an fff, depending on how the sample is recorded and mixed.  It is naive to assume writing a pp will have the slightest significance - except with timbral considerations - in the final mix. And yet recording and mixing have a huge influence on the musical expression.

Finally - if this is not about composing and orchestration then why did you put it on the composing and orchestration forum?

@William said:

"A composer knows in advance where the loudest part of his composition is, and from there the available headroom, respectively the maximum loudness is set."

This shows a complete ignorance of performing and recording MIDI music.  The loudest or softest notated or sequenced scoring might be far lower or higher in overall level depending on which samples are used and how they are processed.

"Level is recording technique terminology. For musical loudness we use the usual Italian terminology such a ppp or fff"

You still do not understand. A written ppp may sound higher in level than an fff, depending on how the sample is recorded and mixed.  It is naive to assume writing a pp will have the slightest significance - except with timbral considerations - in the final mix. And yet recording and mixing have a huge influence on the musical expression.

The dynamic range MIDI Velocity can trigger is 47 dB. We all have the same dynamic range available. How you use this range in composition is completly up to you.

That is completely false.

Another example of misinformation on this thread.

I can use MIDI to destroy a system.  I can trigger, via continous controlllers as well as velocity, a dynamic range that spans as far as any of the instruments I am using are capable of playing.  This includes everything from no signal at all to the highest signal my system is set to - including beyond clipping and into the destruct range.

You on the other hand, are capable of only a paltry few decibels. I don't know what you're smoking but it is giving you delusions that you should not be posting in public. Or perhaps you need to continue with those meds you offered me on another thread.   Your brain is showing some serious signs of clipping.

@Angelo Clematide said:

[...] The dynamic range MIDI Velocity can trigger is 47 dB. [...]

@Angelo Clematide said:

[...] The dynamic range MIDI Velocity can trigger is 47 dB. [...]

What makes you tink so? If I want, I can program the synthesizer or sample-player of my choice to play any chosen volume at any MIDI-velocity sent to it.

That is the MIDI Standard.

1. This is a VSL vl-14 patch with velocity 1-127 playing the the single note A3 (220Hz). Total amplitude (dynamic range) = 35 dB:

2. This is a 0dB sine wave with velocity 1-127 playing the single note A3 (220Hz). Total amplitude (dynamic range) = 47 dB:

@Angelo Clematide said:

[...] That is the MIDI Standard.