A Guide to Producing an Epic Orchestral Track

Although I’m not a huge fan of epic productions, they are definitely the thing that producers/directors/game developers search for nowadays. Now, as a whole, epic orchestral tracks can be heard on movies, TV, commercials, video games, trailers. They always support tension, battle, chase – all of these extremely dynamic moments that always catch the attention of the customer (whether a gamer, listener or spectator, doesn’t matter).

So, this tutorial is a guide that I do hope will help you create your own epic orchestral track. Let’s get started!

Step 1: Composition

A good composition is a very, very important part of the music. It can help you very much in arranging, orchestrating and MIDI-mockuping.

First of all, I’d very much like to point something out – epic orchestral tracks are really similar. Hollywood is a huge machine that manufactures epic tracks all the time, and most of them sound 70-80% the same.

So, in order to get you and your clients happy, you should make something that sounds familiar. But you definitely should try to add a piece of yourself to it. Tired of those I-VI chords? Add something new, tweak the harmony, the melody, add some counterpoints.

There are five things that you should take into consideration when speaking about the composition:

A. Melodic Material

Melody is one of the most important things in music. The ‘leitmotif’, developed by Wagner, is a common technique that is used in modern film/game music – a melody that is connected with a character or event. Remember “The Imperial March” from Star Wars. Or if you’ve watched “The 13^th Warrior” (music by Jerry Goldsmith) you’ll definitely notice how he uses a special motif for the deadly mist (a symbol of the upcoming evil).

Now let’s think about that melodic material, that motif. You should be very careful when repeating the melody. That means that you should not introduce the melody every 8 bars. If you do so, it will eventually sound boring and dull. Begin your track with the first half of the melody, or just a particular interval. For example, the music for Terry Gilliam’s “Brazil” (by Michael Kamen) has a specific melody that begins with a major sixth. Often he uses just this specific interval to point out things that are important.

Also, the melody should be easy to remember (just think about John Williams) and original as well. Yet, the main problem (and paradox) would be that most of your clients (assuming you’re “in the business”) would like that melody to sound popular. (Worst case scenario – to sound “just like “Batman Begins” or anything that is well-known today.)

Now, let’s sum up:

• You should try to make a melody that is easy to be remembered (and sung).
  
• You should try to introduce variations of the melody, so that it doesn’t get boring.
  
• You should try to use specific fragments of the melody at some places.
  
• You should try to make an original and interesting melody.
  
• And you should try to make a melody that is well-known.

Sounds nasty, right?

In my case – I have used specific intervals (I-V-III-VII) as the beginning of the theme. The theme itself is square – four bars, divided into two small motifs of two bars. The beginning of the theme is actually the first two bars of the piece. Listen to just the beginning of the theme and the theme itself afterwards:

Motif, played by french horns:

The melody, played by strings:

B. Harmony

Most of the time melody and harmony go on hand by hand. The same things that are applied to melody should be applied to harmony as well. Try using some chords that are not so popular – diminished or augmented, try different inversions. Using only the first inversions of the chords can be very boring, and you should try expanding your ‘arsenal’ of harmonic language. Mike Elliot has some great articles, called “Adding Flavor to Chords”. If you haven’t checked it out, you should totally do that!

Again, in the track for this tutorial, I have used mostly a very common chord progression – I-III. In the beginning this is F# minor → A major. The thing is that first I use a different inversion – the III is a six-four chord, meaning that the E is in the bass. Now this does not entirely please me, so I decided to make the III an altered chord – augmented. So A-C#-E becomes A-C#-E#, and when using the six-four inversion it becomes E#-A-C#. For more explanation, check the tutorial for the altered III by Ryan Leach.

Hear how it sounds, and how it sounds with the melody applied.

Unusual harmony in brass section:

Harmony plus melody:

Another fragment – harmony in staccato (and ostinato line in low strings):

Finally, consider that tonality can get boring as well. Do you think that using only C major for a whole movie would be good and interesting? Not at all! So you should try to modulate through different keys.

In my track, which by the way is called “Roof Chase”, I wanted to represent a real chase, very tense and dynamic. My imaginary character is insecure. He tries to run for his life. Everything is at stake… So I modulate through different keys – I try not to stop for a long time on a specific key. My character is being chased on rooves, he isn’t likely to slow down and drink a cup of tea, right?

C. Form

Now, my colleague Ryan Leach has a great article on SCOREcast Online regarding form. Nowadays forms like rondo, sonata and etc. are rarely used because most of the commercial music is applied to picture (excluding popular music). So, as Ryan tells us: “Picture is king.”

Nevertheless, you should think of a structure that you can apply on your track. In my case I have an ABA form, meaning – one part, after that you have another, different in style and thematic material part, and finally you have the A again as a reprise (with a dynamic reprise, meaning that I didn’t copy A directly and furthermore developed ideas from A).

D. Ostinato

Ryan Leach has another great article, called “How to Create a Compelling Ostinato”. Today’s music is 80% created out of ostinatos – whether it will be the bassline or those repetitive cello notes, introduced by Hans Zimmer in “The Dark Knight”.

Ostinatos are great, but again – you should try to add something new. Here in the beginning violas and second violins are playing the same rhythm and the same note (B natural), but after that there are these 16^th notes that make a specific mode (minor key with II flat, in our case – B natural – C natural – D natural – E natural – F#).

One thing with an ostinato rhythm is that if you want to have a dynamic musical piece that has the listener in its clutches from the beginning till the end, you should not abandon the pulse. When you start an ostinato you should keep it till the end. (Of course, there are always exceptions – you might want to add a slow, romantic part that will contrast the aggression.)

When it comes to orchestration, you can transfer the pulsation through different sections of the orchestra – strings / woodwinds / brass / percussion. That tip is very useful because ostinato as well can get boring – and by changing timbres you can add variation to our repetitive pattern, creating a stable, yet interesting, foundation of the track.

A nice example is Schostakovich’s 8th Symphony, Part 3. You can see how he uses a repetitive pattern in the violas (at the beginning) and after that in the strings and just when you think that the strings get kinda repetitive, he places the same ostinato in the brass section (trombones).

Ostinato strings (violas and second violins, together with subtle harmonics in first violins):

Another part of the piece, violas and second violins again with bass/celli stabs:

E. Counterpoint

This is a term, that covers the relationship of two or more voices. Now, these epic orchestral pieces are very far away from the masterpieces that for example Bach created during the Baroque. Yet, you can use techniques as imitation, adding different voices (punctus – contrapunctus, meaning that you have one melodic line against another one), mirrored imitations, canons and other.

So, I have a certain moment that I start an imitation between different orchestral sections (French horns – trumpets – violins). It’s not a literal imitation, though it has the same intervals, same rhythm as the main melody. The changes come from the harmony. Hear for yourself:

Only soloing instruments:

Tutti:

Step 2: Orchestration and MIDI

The art of orchestration is a very complex and interesting one. I think that today it can be divided into “classical orchestration” and “MIDI orchestration”. The latter is created with the help of software and samples – check my previous tutorials and quick tips.

Now I think that one rule can be applied to orchestration – when done properly in a classical way, MIDI orchestration becomes piece of cake. Lots of problems will be solved when a nice classical orchestration has been made.

For example if you need high frequencies to your mix, instead of adding some EQ plugins and playing around with the sound – just add a piccollo! This will help add the needed high frequencies and moreover – it will colour up the mix. So a few things that I’ve used in my track:

A. Combining Brass Harmonies with String Melodies

Have you listened to a live orchestra? If so, then you probably have noticed that a single trumpet can blow the entire string section. (James Newton Howard’s score for “The Last Airbender” was recorded with 83 string players!). Now, what about two, three or even four trumpets and twelve french horns? I guess you can picture the situation.

When you place the harmony in the brass section, you won’t need the strings to play the harmony as well (apart from double-basses or cellos that could play the bassline). In most of the scores I’ve looked at, when there is a massive fortissimo harmony in the brass section, first violins, second violins, violas and even some of the woodwinds play in… unison! Therefore they create a thick melodic line (great example is “Dance of the Knights” by Prokofiev’s Romeo and Juliet – there is an amazing video on YouTube, conducted by Valery Gergiev).

There is a certain part in the middle of my piece that has a similar orchestration – there is an ostinato in the whole string section (the ostinato is changed now, to add something new for the listener), the harmony is in the brass section and woodwinds play the harmony in the high register and make some effects (arpeggios in triplets).

Hear the brass section only:

Now hear the string section:

… and the woodwinds:

Finally – the whole orchestra!

B. Blend Different Patches

My next orchestration hint will be to blend different patches together, therefore creating a thicker sound. I mostly used this in my strings section to get my strings to sound as real as possible.

Now let’s talk about violins: I’ve used a preset from Hollywood Strings that is called “NV NV VB MV” (“1^st Violins – Long – (Sus) 3 NV NV VB MV RR”) – meaning that this patch is controlled by modulation wheel and goes from pianissimo and nonvibrato (NV) to con tutta la forza and molto vibrato (MV). Moreover it’s round-robin (RR), meaning that each note triggers a different sample – in our case this is the change of bows (upbow/downbow).

Now I don’t think that this is enough for me, therefore I’ve used another patch from Hollywood Strings – “Sus 13” (from the folder “Long Powerful System”). These patches also control vibrato and volume, but the difference is that the sustain patches control both vibrato and volume independently. The vibrato is controlled with MOD wheel (CC#1), while the volume is controlled with expression (CC#11). This patch, being one designed for a “powerful system” has five levels of loudness for non-vibrato, five levels of loudness for vibrato, and three levels of loudness for molto vibrato. All that means that you can get a very detailed and real strings sound!

Hear our violin melody with just the first patch:

And now the second patch:

And finally with the patches combined:

C. How to Double String Sections

Note that my next orchestration hint will be an apocryphal one and one that you should not use in real orchestration. So, this hint is about doubling each string section… with another one.

Have you heard what ‘divisi’ means? This is when a string section is divided into pieces (two or more). Let’s see for example the celli – they can play double-stops, but double-stops usually sound tense and are not always possible for playing (especially in faster tempos). So, if we want the celli to play in an octave, we should write ‘divisi’ on top of them – our players will see this and will divide the notes between themselves.

Now, the main problem with divisi is that they affect negatively on thickness and volume. When a string section is divided, it sounds as half of the players and doesn’t have that depth in it. So, our 8 cellist for example will sound like 4 cellist + 4 cellist, not like an entire cello section.

So, what’s this hint that I am talking about? Making fake divisi. I’ll double each section with the previous one, starting with the celli, and therefore I’ll thicken the sound. Using that formula, I’ll double the celli line with the basses (without affecting the main bass line), I’ll double the viola line with celli, the second violin line with violas and etc.

The important thing with this technique is that each doubling should be with 40-50% of its volume. If we take our basses for instance, they will play their line in their most natural register and doubling the cellos can be too high for them, therefore the sound will be very unnatural. We only want to have the idea of doubling, just the small appearance of another timbre that will colour up the celli.

Remember – this should be like cooking. You don’t overdo the spices, right? Now listen to how this technique sounds:

Do you notice the subtle octave, that gets in after four bars? (I’ve muted the first 4 bars for the demo, as shown on the image above.) Listen carefully, for best results use headphones. Now hear the whole and “thick” string section:

And remember – do not do this with real orchestras!

Also, another thing that you can do, is to double strings with different timbres – woodwinds, percussion, piano. Hear some ideas:

Violins + xylophone:

Strings + piano (I personally love doubling piano with low string ostinatos!):

Violas + bassoon:

D. Use the Proper Voicing for Brass and Woodwind

Using proper voicing is crucial to great brass and woodwind sound. In my experience I’ve developed a way of placing chords for brass section.

A while ago I read this question in internet: “What are the trombones for?” Actually yes, this is a very interesting question, because mainly our ear is familiar with the epic-adventure French horn or the marching trumpets… but no trombones.

Well, the thing that I can tell you for sure is that trombones are really dynamic brass instruments with a very colourful timbre – meaning that trombones can play low and harsh (like a tuba), and they can play in the high register very softly and their sound can be closer to a that of a trumpet.

So here is the way I’m making my chord voicings (with exceptions of course!):

• Trumpets
• French Horns, Trombone, Tuba.

Therefore trombones are closer to the tuba (low, harsh timbre) and trumpets (majestic, powerful, high). Between them – French horns (noble timbre, thickens the whole “brass section” sound). I guarantee that if you do this voicing right, you’ll have a solid and powerful harmony in your piece. Check it out.

As for woodwinds – there are several types of positions, that mostly change the timbre and volume of the music. If you want to check most of them – try getting some orchestration books like the one from Samuel Adler or the “Principles of Orchestration” by Rimsky-Korsakow. My track, “Roof Chase” is a very dynamic and epic track, mostly – drums, brass and string unisons. This is very, very powerful. Therefore I’ve duplicated timbres – for example I’ll have a simple C major chord and I’ll divide the notes like this:

• 2 Flutes – E + G
• 2 Oboes – C + E
• 2 Clarinets – C + G

So, the whole woodwind section (apart from bassoons), which is equal to six players, will play C + E + G and each one of the notes will be doubled. This technique will help the chord to fight the power of the brass section and percussion and therefore stand on its own.

E. Use Hits

Remember that you should use the whole power of the orchestra, but not each bar. One way to do this is to use hits. I’ve written a quick tip regarding hits that may help you. The same techniques apply here, only that I’ve used instruments like piano, marimba, synths.

F. Add Interest with Orchestra Effects

Orchestra effects are a great way of making things sound interesting. The perfect instruments for this in such an epic production are the pitched percussion (marimba, xylophone, celesta, etc.) and woodwinds.

Listen to some ideas:

Xylophone making some transitions:

Woodwinds playing triplets (in unison, to make a thick sound that will endure the power of the other sections):

Woodwind runs:

Step 3: Recording Live Instruments

As I’ve written in my previous tutorial, “Making Your MIDI Strings Even More Realistic”, whenever possible you should record some live instruments that will give the production more timbre and colour. In my case, I didn’t have the budget to call for string or brass players, so I decided to record some drums in my home studio – so I’ve recorded some sticks, bendir (Turkish frame drum), bells, cajon and a strange African drum that a friend brought me from Uganda.

The important thing is that apart from sticks and bells (which serve more like a “decoration” to the main beat) all the other percussion should be in unison with the MIDI ones.

Hear for yourself.

Recorded drum solo:

MIDI drums only:

Drum mix:

Step 4: Adding Synths

Another technique is to add some synths to thicken the sound. Arpeggiators are a good way to add colour to your track – I’ve used one patch that I made myself with Cakewalk z3ta. You can add some modulation effects (tremolo, vibrato, flanger) and some distortion to make the sound more industrial. One great thing to read will by Ryan Leach’s tutorial about tremolo.

Hear how my synth blends with the strings at the beginning:

Another great use of synths will be using some subbass sinewaves to thicken the bassline or to add depth to hits. Hear how I used a single subbass note on some of the hits in the track (it’s delicate, though):

Step 5: Mixing

Now we come to the tough part – the mixing. There are few points in the mix process that you should take care about very precisely. It is not going to be easy – with my template all my tracks go to 80+ (without the audio channels), so it can get kinda messy there! There is a rule that the good mix makes mastering easy. Well, as I said before – good composition and orchestration makes a track easy to mix. When you compose with the mix in your head you can achieve great result on the sound.

There are lots of things that I could say about mixing. The main three things that you should care about are: depth, panning, frequency balance.

Frequency Balance

Overall, you always have to think about balance – when mixing in stereo you should think about having your left and right channels to be equal. Also, you should not overdo some of the frequencies – for example highs over lows and so on.

Some points regarding frequency problems:

1. Try not to put each instrument in the same register. Use different ranges, think about your chord voicings. Be careful about doubling different instruments.
2. As I said previously, if you need more frequencies, think about orchestration first, and after that think about equalizers. You need more low-end? Add tuba, bassoon, contrabassoon, timpani, gran casa. Need more high-end? Add flutes or piccolos, add celesta, glockenspiel, xylophone.
3. If you have lots of low-end problems (commonly with drums) add a nice high-pass filter. Same things that apply for popular genres can be applied here as well!

Depth

Now, the depth. One of the biggest problems is how to use reverb and, most commonly, how much reverb should be added. In orchestral tracks I’d suggest using send effects. See another great article by Mo Volans.

Why using sends? In my case, my template takes about 15 gigs of RAM and my CPU is used on 80%. Putting reverb on each of those 80+ channels would be a total sucide.

Another idea is that having one reverb setting for your orchestra will give you a much more coherent sound. As you’ve probably noticed from the demos above, Hollywood Strings and Brass (Gold) are pretty dry, so I’d like to add a nice reverb that could get the sound to a next level. So, I’ve used one SEND fx track (stereo) where I’ve put a simple impulse-response reverb (the Cubase-built Reverence).

One thing that you should take notice of is that all the signals go there with different strength. In a real orchestra hall the brass section is WAY behind the strings (= more reverb), while the strings are upfront. Hear the “before” and “after” demos:

Before:

After:

Parallel Compression

And my final tip for this guide should be the “parallel compression”. Check Joel Falconer’s tut “Give Drums Impact with Parallel Compression”. This will thicken up the sound of the drums and will make them sound very, very epic. Also, this will help you bring the drums more upfront in the mix, making them that stable basis of the track.

No parallel compression:

With parallel compression (listen how the drum section ‘thickens’!):

Conclusion

So, I’ve come to the end of this tutorial. Hope this has inspired you to compose, orchestrate and mix your own music. Don’t forget to share your tracks in the comments section!

Here’s how my track sounds after all those steps:

… and the final, mastered, version: