PDA

View Full Version : How to avoid heaviness, opaqueness



Guy Smiley
09-11-2006, 11:25 PM
I think this seems to be my biggest problem. Does anyone have any orchestration tips or rules on how to avoid heaviness? My softer sparse passages sound fine, so I'm mostly talking about loud tuttis. It seems like by the time I've thrown in all the parts, I have a muddy, nondescript mess.

Next time I'm just gonna try some of the obvious things (wider spacing, more doubling, more rests?), but I thought I'd ask to see if anyone had any clever tips or a similar problem.

Thanks
John

Skysaw
09-12-2006, 12:04 AM
If you want to post a few measures as an example, I can probably help assess what's going on. Just remember that orchestration is a fairly deep subject. If you haven't already gone through the Rimsky-Korsakov course, I suggest you do that when you can.

Hajo
09-12-2006, 02:51 AM
Hello,

in the first go I wouldnt consider the changing of the orchestration. Transparency etc. will be handled by you (the conductor).

It could be perhaps a problem of the "audio recording". So dont forget to adjust the overall volume of every single instrument and the position in the stereo image. You only have a fixed and not very big dynamic bandwidth (compared to reality) and that could cause some problems.

Hajo

Raymond62
09-12-2006, 03:30 AM
You only have a fixed and not very big dynamic bandwidth (compared to reality) and that could cause some problems.

Please, explain a bit further what you exactly mean by this.

Raymond

Skysaw
09-12-2006, 08:24 AM
in the first go I wouldnt consider the changing of the orchestration. Transparency etc. will be handled by you (the conductor).
I'm confused. Did John say he was approaching this as a conductor? It was a question about orchestration, so why would you suggest to him that he not change the orchestration if you haven't seen it? Heaviness can certainly lie in the orchestration itself, which is why I asked for an example.



You only have a fixed and not very big dynamic bandwidth (compared to reality)...
I'm curious about your use of this term as well. As far as I can tell, it has no meaning in this context.

Inkfingers
09-12-2006, 08:56 AM
I couldn't say without seeing your scores, but it could have to do with your chord voicings. Do a reduction of your score (or just the tutti passages) to two staves, writing all the notes in each chord with their doublings. Muddiness often comes from too many close intervals in the low registers. If you have that, you might consider spreading out those intervals more according to the overtone series. That is, octaves in the bass, fifths and fourths above that, then thirds, then progressively smaller intervals above that if needed. This technique is a good place to start, at least, considering all the masters used it to a certain extent.

Hope this helps!

Poolman
09-12-2006, 09:58 AM
It seems like by the time I've thrown in all the parts, I have a muddy, nondescript mess.

Well, Guy, first try not throwing in ALL the parts. A "tutti" is best not defined as one with all the players, but one in which all or most players take part, such that no solo lines are apparent. However, I will assume that you want them all in.

But (if your problem actually is an orchestration one) I guess that you have a) too much unison doubling, b) complete harmony in all three main departments. If you write a passage which sounds complete on strings, then add a woodwind group at the same pitch which would equally sound complete, then a complete-sounding brass section, assuming most doublings are at the unison, then you will get an "organ-like" sound which certainly sounds a bit nondescript in the concert hall, let alone on the computer.

Try a) making sure the woodwind are spaced so that most of them are well above the brass (except the bassoons which should be as low as convenient); b) getting the 1st violins up above the brass also; c) Important: prefer octave doubling to unison.

If that does not "lighten" the sound, then try giving some departments an independent part. For example, if you can invent a new countersubject, gice it to all the woodwind in octaves. If not, then maybe give the top part to most woodwind plus violins, the alto part to all four horns in octaves, and so on. Another idea is to give some of the instruments a decorated version of one of the lines, or conversely, a simplified version. The main thing is to ensure that each department has something different from the other departments. Work it how you may, but avoid the unison organ-like sound, except perhaps for very short passages.

Terry

DPDAN
09-12-2006, 12:11 PM
Well, Guy, first try not throwing in ALL the parts. A "tutti" is best not defined as one with all the players, but one in which all or most players take part, such that no solo lines are apparent. However, I will assume that you want them all in.

But (if your problem actually is an orchestration one) I guess that you have a) too much unison doubling, b) complete harmony in all three main departments. If you write a passage which sounds complete on strings, then add a woodwind group at the same pitch which would equally sound complete, then a complete-sounding brass section, assuming most doublings are at the unison, then you will get an "organ-like" sound which certainly sounds a bit nondescript in the concert hall, let alone on the computer.

Try a) making sure the woodwind are spaced so that most of them are well above the brass (except the bassoons which should be as low as convenient); b) getting the 1st violins up above the brass also; c) Important: prefer octave doubling to unison.

If that does not "lighten" the sound, then try giving some departments an independent part. For example, if you can invent a new countersubject, gice it to all the woodwind in octaves. If not, then maybe give the top part to most woodwind plus violins, the alto part to all four horns in octaves, and so on. Another idea is to give some of the instruments a decorated version of one of the lines, or conversely, a simplified version. The main thing is to ensure that each department has something different from the other departments. Work it how you may, but avoid the unison organ-like sound, except perhaps for very short passages.

Terry

that is how it's done!
Poolman knows his stuff. :)
Dan

Hajo
09-12-2006, 01:42 PM
Hello,

sorry to have caused some troubles: I didnt realize that the focus of the question was on orchestration. :(

I was refering to a problem which I had lately: The balance of voices when working with Finale and sound intensity (loudness) when working with a lot of instruments. I had to minimize the input of all instruments and I switched to a daw to controll the mixing. The Finale studio-view was for my eyes and ears somewhat limited.

So thats the point where the conductor came in who controls also on the audio level the transparency of the sound. And I still dont understand how to "translate" the dynamic range from a real orchestra to that what is possible on a cd. :confused:

Hajo

Guy Smiley
09-12-2006, 08:12 PM
Wow Poolman, I think you probably hit the nail on the head. Here are two of the problem pages. It's untransposed (except basses which sound 8vb like usual). It's also a draft, so it's missing details like slurs, and chords over the timpani are just for my own reference:

http://www.mediamax.com/guysmiley/Hosted/A%20Minor%20New.pdf

Poolman, I have exactly what you describe! I guess all sections sound complete by themselves. Then I throw them all together, and out comes the organ sound. It's particularly thick in the middle register. I think the top and bottom come through ok.

Skysaw, I've definitely been paying attention to the RK course! Haven't touched much on tuttis yet though. I'll appreciate any input you have!

I double a lot at unison, which I will do differently from now on. Actually the brass melody is doubled at unison exclusively! :\

What's funny is that the orchestration books I have go into great detail into how to build chords out of the different sections. So then I just kinda assumed to do a tutti, you build each section, and then add them all up! I guess not...

So is it a good idea to just stick with SATB voicing then? B doubled 8vb, S and A doubled 8va? I think Kennan in his book talks about that strategy...

Skysaw
09-12-2006, 10:24 PM
Thanks for posting some examples, smiley.

Honestly, they are not that far off the mark, but there are some problem areas. Here are a couple of quick suggestions.

In traditional harmony, everything will sound much clearer if you follow traditional 4-part voice-leading where it makes sense. You would do well to study counterpoint to clear some of this up. But in a nutshell, you can improve it quite a bit by observing the following:

1. Try to work out the harmony in four voices where it makes sense. Let the instruments play these voices in different octaves but do not stray from their original outline unless you have a specific reason to do so. Ex. measures 45-46. Compare ob II, Clar I, and Tpt I. They all seem to want to play the same line, but notes occasionally pop into a different voice. This loses the sense of continuity between the individual lines.

2. Don't double the bass line an octave up if it crosses another voice. Ex. measures 45-46 the fourth horn should not double the bass as it is higher than other voices in the trombone. Also Ex. measures 49-50 fl II, ob I are playing the bass line.

3. Where the bass sounds confused, consider simplifying. I believe some of what you perceive as "heavy" comes from all the energy in the lower register, some of which is obscuring the bass line. Measures 49-50 have the contrabasses playing a figure that adds harmonic content against the cellos and violas that does not seem to further your cause. Add in the trombone and bassoons on the further simplified bass line, and it gets denser still. The figure in the cellos and violas is not a bad one, but you do have to consider that notes in these arpeggios are in a low register, and should be considered as taking up the full harmonic space it encompasses over the course of the measure. Between the CB, Vlc, and Bssn, you are basically scoring a very densely packed d minor chord in the orchestra's lower range.

When you wish to keep everyone busy, it is not necessary or desireable to come up with a completely different idea for each instrument. Everything will be much cleaner sounding if you have a clear idea of what each voice (melody, counter-melody, bass...) does, and then let the instruments choose which of the voices it is representing.

I hope this helps somewhat. There is so very much to consider, but I think you are on your way.

Edit - one more rule of thumb. In a triad or seventh chord, if the bass is playing the third of the chord (first inversion), no other voice should double that note. The most common practice here is to double whatever note is in the soprano.

Skysaw
09-13-2006, 01:33 AM
I thought I'd give this further attention by posting a concrete example of one solution for your work. I hope you don't mind my doing so.

First, regard the following four-part harmony example:
http://www.allhands.com/gpo/smiley.jpg

This is your work! Play it on the piano, and you will immediately recognize it. It is the section from bar 45 through 49 reduced to its essential information. Your example had clear soprano and bass parts, but the alto and tenor were impossible to see (or hear), as the instruments jumped between parts. Note how here we have four distinct voices, each with its own logical line (the soprano is simplified), and each harmonizing with the others to produce the chord progression. This is not the only four-part solution given your soprano and bass, but it is probably the simplest.

The next step is to allow some doubling at the octave for the different voices. Do not allow the bass line to cross another voice. This means the tenor should not be doubled 8va below, and the bass should not be doubled 8va above. The only other recommended restriction is that the soprano voice be the highest. This means that while you can double the alto up an 8va, doing so means you should also double the soprano up an 8va.

Let's see what this might look like:
http://www.allhands.com/gpo/smiley2.jpg

Stay with me here... the theory is just about over, and the fun part begins!

Here is a pdf file (http://www.allhands.com/gpo/smiley.pdf) of the orchestra playing this section. It is only one of a million ways it could have been handled, but I believe I have kept the flavor of your original. It would be worth it for you to assemble this in your sequencer and actually hear how it sounds, even if you aren't crazy about some of the liberties I took with it. (continued...)

Skysaw
09-13-2006, 02:11 AM
Go ahead and print out the pdf file (http://www.allhands.com/gpo/smiley.pdf), and take a good look at it. See what has been removed, and what has been added. Try to get a feel for how the above four-part harmony remains in this version, and in what ways I took artistic liberties by straying from the rules I layed out. Yes, of course you are allowed to break the rules! But it's always best to know what you're breaking, and why.

Let's take a look at this instrument by instrument:

Flutes - Here is the soprano voice played in octaves. This simplified version of the melody works well with the complete version played by the violins, but they could of course have played together with them. The first flute is in a nice clear register, and bolstered at an octave below, the two will be heard nicely.

Oboes and clarinets - Alto and tenor parts an octave above the original. I break out of our 4-part mold at the end of the second measure for the sake of a new moving part. Why? I just thought a little motion leading up to the big chord at ms. 47 might be nice, though it did require me to break my voice-leading. This new motive will be heard even through the rest of the orchestra because, with the violas, it is the only moving part.

Bassoons - Bass line

Horns - Alto and tenor voices at the original octave. Horns III & IV are joined by I & II at the full tutti bar, and the notes are played as repeated eighths. Even though the rhythm has been altered, they still follow the voice-leading we have set out for alto and tenor.

Trumpets - I have given them the first two bars out, so that their entrance is more dramatic. I changed from the original thirds to a nice clarion octave figure. A bit of a cliché perhaps, but why not?

Trombones & tuba - Again, joining only for the full tutti on the A major chord. They play alto, tenor, and doubled bass lines.

Timpani - Typically, the timpani does not need to follow any voice-leading rules, as it has a limited number of notes available.

Violins I & II - Soprano voice in octaves. I changed your melody just a tiny bit because I wanted to hear it in this form. Nothing wrong with the original!

Violas - Here is where it gets a little fun. I wanted to keep some of the driving rhythmic element from the original, so I gave them arpeggiated 16ths. The effect is very similar to what I believe you had intended, in that the first two 16ths of each set of six will be heard most clearly. The reason is that it is the only moving part there.

But what of the notes they are playing? They are in fact playing three voices simultaneously: The alto and tenor at the original octave, and the alto one octave below! Even here we follow the voice-leading we wrote, as you can see if you look at how each note in the arpeggio changes from beat to beat. However, they actually are playing the version of these voices that I gave to oboes and clarinets, by arpeggiating the moving figure I gave to those instruments at the end of bar two. Then I begin to break our voice leading up through the end, to provide a driving figure to our resolution

Cellos - Begin as the bass line, but break after the C# in the third measure. This is a good spot for the change as we have the heavy brass adding to the bass here, so it's not as necessary for the cellos to cover it. The other reason is so that it can double our new viola line an octave below, otherwise that figure would begin to be covered up, and I wanted to make sure it was heard.

Bass - Boring, perhaps, but I let them play the bass line! :)

I hope this has helped you. Please feel free to ask any questions about what I've presented you with. I'm more than happy to help.

Poolman
09-13-2006, 03:29 AM
Smiley, that is how it is done. Skysaw knows his stuff!

Terry

Guy Smiley
09-13-2006, 04:14 PM
Haha wow! Skysaw, I must say, I'm speechless! And pretty stoked that you put so much thought into a response. Very cool, I really appreciate it. I'll definitely have a look at everything in more detail when I get home, and then likely have more questions :).

Guy Smiley
09-13-2006, 09:29 PM
Ok this has been very educational. Thanks to both Skysaw and Poolman, and Inkfingers too. I think my next endeavor will be written out entirely in 4 voices before I go to the big page. But when I first got GPO, I had no patience for that! :)

One question I have is, look at the example in RK's course, lesson 11, right above Belkin's note (6). Any tips about when is it or isn't safe to do something like that? Those kinda look like several voices walking all over each other, or should that be considered as texture #1 "doubling" texture #2?

Thanks,
John

Skysaw
09-13-2006, 10:21 PM
One question I have is, look at the example in RK's course, lesson 11, right above Belkin's note (6). Any tips about when is it or isn't safe to do something like that? Those kinda look like several voices walking all over each other, or should that be considered as texture #1 "doubling" texture #2?
There is a question of interpretation of what RK was saying here. Belkin's comment was "especially when the timbres are varied," but I believe that is what RK meant by "different texture." That is to say it will not sound as muddled if the different staves are each handled by a different instrument group, for example.

It's also usually safe to combine these sort of figures if one of the elements is static, such as the long chords in the bottom staves of the example. For the most part, this is done to thicken texture, which is what you were trying to avoid in your composition. It would definitely lead to confusion if the whole orchestra had multiple variations on these ideas all superimposed.

By the way, thanks for pointing me to that section. I was happily reminded that those were my jpgs on that page. I guess I hadn't seen that chapter online till now. :)