@BachRules said:

No one's suggested it can. On the contrary, post after post after painful post, I point out that arrangement and mixing are totally irrelevant to this specialized topic.

I realize that you took the effort to point that out, but I'd argue that the sentiment is mistaken. Why would you ever be concerned about relations of volume if not in the context of musical arrangement, balance of orchestration and the mix. I mean you can do it if it brings you joy or whatever, but in the context of music making, it's idle.

@Another User said:

Why would you ever be concerned about relations of volume if not in the context of musical arrangement, balance of orchestration and the mix.... I mean you can do it if it brings you joy or whatever

I hadn't even known I needed your permission.

@BachRules said:

I hadn't even known I needed your permission.

Yeah, I was keeping that a secret too ... 😛

@Another User said:

As people continue to misunderstand this topic, I'll continue to clarify that it's unrelated to human perception. It's just about numbers in DAW's and pressure-waves moving through air, not about humans. If I could make that clear using fewer words, I would.

@Another User said:

Once you introduce art, taste, and creativity into a discussion, the discussion is bound to degenerate into nonsense.... But science -- e.g., audio engineering and computer science -- is different, and it has rules, and when I have a question about science and someone tries to change the subject to art, I will always attempt to redirect the discussion back to the topic: science. Thanks for understanding.

Thanks for understanding, everyone. If you cannot imagine Newtonian Mechanics functioning without humans to perceive it and artists to adjust it, you are incompetent for this topic. If you feel Newton was all about "hair-splitting and arguing semantics", please allow me to show you to the appropriate thread for you:

[img]http://i.imgur.com/ESLC6i5.png[/img]

What will be the next self-absorbed, vapid lout's excuse for prattling about his art in this thread.

@BachRules said:

If you set your volume faders all to the the same level, turn on Natural Volume, and use only velocity (and instrument-positioning in the MIR room) to control loudness, your will get combinations of timbres limited to the combinations which are possible in the real world.

No, that is not correct, because it ignores the differing dynamic range of instruments. There is probably one dynamic at which this is true, but as I have no idea as to how the Natural Volumes were calculated, I have no idea which dynamic it would be.

I'm not understanding your point. Not disagreeing with you, but I don't understand what you mean....

... the dynamic ranges of VI are not really accurate. For example the dynamic range of a Flute is far less than say a Trumpet, but in the VI player the difference is less severe than it should be. Maybe this is something that the VSL guys could comment on and possibly give figures as to how the normalisation (for want of a better word) has been achieved....

If my assertion in the inner quote is incorrect, there should be an example of specific settings where Natural Volume results in unnatural timbral combinations. If no such example surfaces, I'm going to chalk this up as superstition or unsubstantiated rumor.

BachRules,

let me put it this way: for obvious reasons, I find your use of terms like "egomaniac", "babbling", "self-absorbed" and "vapid" quite amusing. I'm sure you'll figure it out. Look at it as a chance for personal growth. It's not science, after all, isn't it?

@JimmyHellfire said:

BachRules,

let me put it this way: for obvious reasons, I find your use of terms like "egomaniac", "babbling", "self-absorbed" and "vapid" quite amusing. I'm sure you'll figure it out. Look at it as a chance for personal growth. It's not science, after all, isn't it?

That's deep, man, because I'm the one prattling about his art in a thread where it's irrelevant. Except it was you not I who really did that.

So you bought a DAW and learned how to move the faders, and from there BAM you inferred the entire natural world depends on your operation of the faders, and anything less is "hair-splitting and semantics".

You saw the earlier talk about worlds where there are no humans to perceive air-pressure waves, and that hypothesis was too much for your ego to bear, so you pretended it never happened and you charged in slack-jawed insisting how important you are to the functioning of audio.

And when someone asks about the operation of MIR's Natural Volume feature, there you are like a drooling Pavlovian dog preaching, "Hey it's not some magic do-your-mix-for-you button," and you preach this despite no one ever having suggested it was.

You're so distracted by your own mixing practices, you can't comprehend that there's such an objective phenomenon as Natural Volume that's independent of you and your mixing. And you'll remain ignorant because you'd rather attack than take the time to figure out wtf the words "Natural Volume" really mean (not what you assumed).

No one asked you about magic do-your-mix-for-you buttons. That's just some music-forum cliche you memorized before this thread started, and you'd see it's irrelevant if you read the thread.

In summary, you move your DAW faders until you like how it sounds, and you feel compelled to announce that mundanity to the world, even when you're disrespectfully hijacking unrelated discussions.

Yeah, I bought a DAW. Smart move, I'd say. I also went to the university and had my joyful drills about SPL, sound travel, room acoustics and all the related sorcery. I might be the wrong person to push your sciolism pissing contest on to. Whatever. It's not important.

You can puff up and verbosely posture all you want - expecting software to do a job you're fearful of attacking yourself is exactly why you made this thread. Also, your questions have already been answered - by Dietz among others. My own post that hit your snooty nerve so much actually just echoed that. Take it or leave it. I'm gonna have a productive day today, perhaps you should try the same.

To come back to my snippet: it's an excerpt from Tchaikovsky's 6. symphony (first movement, just before the development section starts). You were right, the first snippet is my mock-up, the second one the recording. If you listen closely, you can hear that there's a big difference between the two when the bassoon takes over the line from the clarinet. In the mock-up, the transition is smooth and the bassoon plays it even a little more quiet than the clarinet (as it should be according to the score). In the recording however, the bassoon sticks out like a sore thumb. It's much louder than the clarinet instead of even quieter.

Now, there's the problem. You could actually say that the mockup sounds better in that aspect. But it's not 'natural'. A bassoon can not play as quietly in it's low register, whereas the clarinet can play very quietly over it's whole range. That's why this particular motiv is almost always played on a bass clarinet instead of a bassoon (eventhough Tchaikovsky asked for a bassoon).

Anyway, here comes a problem for the 'natural volume' feature. If you wanted to restrict your sample instruments to only what they can play in reality, VSL would have to delete the pp layer for the bassoon's lowest register. So for example, when playing in the bassoon's low register with a volume of, say, 22, there should either be now sound at all or one that is much louder than the requested. I can only imagine the complaints VSL would get if that was the case. And that's but one example. If VSL applied this logic rigorously through all instruments, their libraries would be very inconsistent and tedious to learn. You'd always have to ask yourself: 'oh, why can't I do that particular thing. Ah, yes, it is because this instrument can't play a trill on these two notes'.

And another thing why 'natural volume' can't account for each and every situation. In a real orchestra, the players listen to each other and balance their volume and pitch accordingly. When a horn section of 4 players has a ff marking, in one case they will play with all their force at 120 db (for instance when there are a lot of other instruments around, all playing very loudly too). In another case however, despite having the same marking of ff, they may only play at 90 db (when they are playing in a smaller setting, for example, or when another instrument has the main line and should be heard clearly over them). That's because dynamic markings in a score are no scientific instruction, but always depend on the context. And the 'natural volume' VSL provides can never respond to context. That's the users job, really. Despite your claims that 'natural volume' is a measurable, invariable quantity, it isn't. The balance between the instruments depends on the context. Like it or not, but you yourself will have to take it into account and adjust some things accordingly.

Software can only get you so far. There's no way around learning about the individual instruments you are using if you want a 'natural' sounding orchestration, and balance. So I'd advise to pick up a book about orchestration and learn the basics. That way VSL's 'natural volume' will still be a timesaver (because you don't have to set up everything from ground, but only tweak a thing or two depending on the context of your music), but you won't depend totally upon it. And you could use whatever sample library you'd want, not just the ones with 'natural volume'.

Hi everyone !

I had written a long and thorough explanation of my point of view, and then accidentally closed the window ^^.

In short : I understand Bachrules’ « quest », and I agree with him when he says there is a natural volume. It’s the world of physics, sound is like light, there are rules which are completely objective (at least in our macroscopic world ^^).

I too wanted that « ready to go » orchestra, and I was often mock or considered as a « do my mix » kind of guy.

But since, in the end, I completely nailed it and was utterly right, that’s ok ^^.

First, there’s one thing which really doesn’t exist : ABSOLUTE natural volume. You don’t know « where » is the silence on a computer. You only have a ceiling. You don’t control the listener’s volume.

However, there is a RELATIVE natural balance between instruments. But first things first, before even touching a volume fader, you have to check the dynamic range of all your instruments, so that something balanced in pp is still balanced in ff.

Since even a single instrument has different dynamic range depending on the note played, the best way to consider the dynamic range of a whole instrument is to get the difference between the softest sound and the strongest one, regardless of the played note (weakest and strongest register, respectively). So you have to search a little bit^^.

Once you have your different dynamic ranges set up, you can finally move your volume faders.

But for all this work, I needed data. That’s where internet comes in^^.

I did some digging, found some interesting sources and create my own interactive excel document to balance virtual orchestras. Since it’s still a bit buggy, I just posted pictures on this post, so you have the « theorical » setup.

https://dl.dropboxusercontent.com/u/4695665/Theorical%20dynamic%20range.pdf

https://dl.dropboxusercontent.com/u/4695665/Theorical%20setup.pdf

First, the Theorical dynamic range of the main orchestral instruments. One cool thing about dynamic range is that it doesn’t depend on the distance with the listener. It’s the « instrument distance with itself » so it’s an objective data. Sure, in real world, the background noise can « swallow » the softer notes, diminishing the effective dynamic range, but there is no background noise on a computer ^^ (unless you use Mir’s ^^).

Then, the Theorical volume. Instruments are considered « dry », before spatial placement of any kind. Their volume is based on the ppp of the clarinet, since it’s the softest sound of all the winds. Once you set up this correctly, you have to had the « distance correction » on the volume of the instruments (as for myself, since I use mir pro, i let it decrease the volume with the distance of the microphone, but I don’t use natural volume). As a rule of thumb, multiplying the distance of an instrument by 2 is equal to -6dB. (Give or take). For convenience, I usually consider that my « dry » instruments are 1 meter away. Works well enough.

PS : I only use vienna strings (dimension, mostly), with xfader on. In this configuration, the vsl instruments are NOT normalized, and have strong and weak register, like a normal instrument would (but it may only be that the xfader plays with the volume). For the rest, i’m mostly using sample modeling and modartt ‘s instruments.

PPS : of course, this is a theorical setup. The datas I found are not always very precise, and I even filled the gaps myself. I’m sure it can be improved. But I truly believe it’s a solid base that allows for direct artistic work. The only changes I make on my setup are for artistic purposes, never technical ones. 

Here are some short audio examples. I didn't mix them at all afterwards. The mix is completely "calculated" ^^.

To Dominique:

I respectfully disagree.

The best way to learn the balance of an orchestra is, on the contrary, to have limits. Not being able to do something is the best way to learn it's not possible. And to be able to learn that, you have to KNOW that it's not some technical issue, but a somewhat real orchestra behaviour. The thing you consider "freedom", I see it as a trap. 

And if for a solo line, it isn't much of a big deal, it becomes a catastrophy when you layer wrongly balanced instruments, ending with weird ensemble sounds, not reacting equally and properly throughout the different dynamic layers.

Why is it so important to have a setup based on obective datas and not subjective hearing ? Because it's the only way you can efficiently learn from all the great composers (even from their mistakes, like your Tchaïkovsky example). Learning is mostly mimicking. But you can't do that properly when your french horn is as strong as your trumpet in fff, which doesn't make any sense in any orchestra.

Freedom is good when you're already familiar with the way it normally works. And even then, it's often a trap.

IMHO, of course^^.

Interesting data Plougot. Thanks for taking the time to gather that together. But isn't it exactly what VSL's 'natural volume' does, i.e. give you starting values for the relative volume of the orchestras instruments in proportion to each other? In any case such measurements are a very good starting point. It doesn't free you from knowing orchestration though. From your table we can gather that a bassoon has a dynamic range of 30 db. That means that it's loudest possible sound is 30 db louder that it's quietest. It can be very different for each register though. The low register will have a smaller dynamic range than that, (probably around 15 to 20 db) simply because a bassoon can not play as quiet there. The opposite goes for it's high register: the dynamic range may be very similar to the low register (15 to 20 db maybe), but this time because the high notes can not be as loud as in the deep register. With that in mind, you would have to set an individual dynamic range for each register of each instrument. I think that's not currently possible in the VI Pro Software. And while it may be nice to have, I think that it isn't terribly important either. If you use VSL's 'natural volume' feature (and Dietz has explained how to set it up properly) you should be fine in most cases. When making a mockup there are far more noticeable deviations from a real recording than the instruments balance. Interesting as this discussion is, I think it focuses on a tiny detail that most people won't even notice.

If you have Sibelius, Wallander's Note Performer is a good tip. It uses algorithms (derived from real performances) to shape the playback of your score to be similar to a real performance. It won't sound as realistic as a mockup, but it's a great tool to check such things as the balance between the instruments etc. I guess that may come somewhat close to what you are wishing for. And in my opinion it is better suited to 'teach you how it normally works' than a set of samples, which also has to cater to the ones who need the freedom for their daily work (there is a rather large community of people using samples for media compositions. It's actually much bigger than the ones who do 'classical' music with samples, I guess. And for these people 'reality' isn't the first priority, or even the second. But rather consistency throughout the instruments, flexibility, freedom...).

As an aside: I don't think Tchaikovsky made a mistake there, far from it. I think he noted an ideal that a performance should aim for, eventhough it can't be executed literally (a bassoon really can't play pppppp there). The problem in the recording lies with the clarinet, which is too quiet already so that the bassoonist has no chance for a smooth transition. He should help him out there and play a bit louder than that, while still keeping the overall impression of a morendo.

My bad if you were exclusively talking about VSL's natural volume. It's true that I kindof expanded the subject without warning ^^.

To Dominique:

About the dynamic range, I explained that my data is regardless of the note, meaning each time the softest equals the ppp of the softer register, and the strongest is the strongest of the stronger register. The result is a data not depending on the register you play, but it doesn't mean that your instrument will be able to reach that particular volume with every notes.

As I said, using the xfader at least, the vsl instruments I use are not completely "normalized". They have strong and weak registers already, and with a little tweaking on the dynamic range scaler, I can have some "realistic" behaviours on the whole dynamic range without having to think about it anymore. Of course, i'm sure it's not perfect, but it's more than close enough.

And for my winds and brass, which are all sample modeling's, once you set up correctly the dynamic range, since they are semi synthesis, they behave naturally already, with credible changes in the dynamic range depending on the register. The main purpose of my whole method is to set a "global dynamic range" on the instrument, to be able to balance it with the others. If internally, the instrument is unbalanced with itself, it won't be fixed. But in my experience, it's not the case.

One thing I didn't mention : I think this whole method became necessary because I combined several different libraries, which I had to balance intelligently. I guess VSL products are more or less balanced within themselves, so with a "Only vsl setup", it might not be as useful. Don't know for sure, though.

on the "Tchaikovsky" aside^^: I'm not convinced. Noting something so far from reality that it is not fisible (if i'm correct, many if not all orchestras switch that bassoon for a bass clarinet) is not "ideal", it's a mistake^^. Maybe Tchaïkovsky bet on future enhancements on the bassoon so that it would be able to get softer, i don't know. But on a score, if something is not practical, it's not "ideal", it's "not good" ^^. Please keep in mind that I know who we are talking about, one of the greatest composer of all time, whose work is purely astonishing. He's personaly one of my favourite composer (if not THE one), so i'm not at all making a point "against" him.

@Another User said:

I also went to the university....

But you ended up a sycophant attacking people for making feature-requests on computer-program forums, and that's a foolish way to behave, so apparently the university took your money and ran with it.

I think you two could almost continue your arguement in pm, now, since it's not interesting for anybody, and not even relevant to your own post 😛. At that point, who insulted who first isn't relevant any more.

I posted some messages, why don't you answer stuffs about that instead ^^ ?

@Plougot said:

I think you two could almost continue your arguement in pm, now, since it's not interesting for anybody, and not even relevant to your own post 😛. At that point, who insulted who first isn't relevant any more.

I posted some messages, why don't you answer stuffs about that instead ^^ ?

I skimmed your post and it appeared to be squarely on-topic, so I intend to read it carefully when I have the time it warrants, later. In the meantime that housekeeping I performed took only a few seconds.

JimmyHellfire, please use PM for your future personal attacks against me.

@Dominique said:

Anyway, here comes a problem for the 'natural volume' feature. If you wanted to restrict your sample instruments to only what they can play in reality, VSL would have to delete the pp layer for the bassoon's lowest register. So for example, when playing in the bassoon's low register with a volume of, say, 22, there should either be now sound at all or one that is much louder than the requested. I can only imagine the complaints VSL would get if that was the case. And that's but one example. If VSL applied this logic rigorously through all instruments, their libraries would be very inconsistent and tedious to learn. You'd always have to ask yourself: 'oh, why can't I do that particular thing. Ah, yes, it is because this instrument can't play a trill on these two notes'.

One last thing, Dominique, I think you may misunderstand what "volume" is, inside Vienna instruments pro (maybe you just spoke too fast though). But your volume "22" is purely a midi data, traditionnaly on cc7, and in no way indicative of a "real" volume. It just says to the virtual instrument "you have a data of 22 on a maximum of 127". It is written nowhere that the same cc7 data should result in the same dB output on every instrument (no reason whatsoever to have a mute bassoon at cc7 22, even if this bassoon has a pianissimo 10dB louder than the clarinet also playing at cc7 22, for example.) In that case, a logical way to behave would be to have the bassoon way louder, because a bassoon at 22/127 of its "power" is way more powerful than a clarinet at 22/127.

Eh, it's just BachRules and his usual diatribe, well documented on these forums and elsewhere. At least he's consistent.