There’s A Place For Us

By • February 1, 2011

Veteran scoring mixer John Rodd and Tony Shepperd, the creator of Tony Shepperd’s Master Series recording and mixing DVDs are both well known, very experienced and highly regarded engineers in Los Angeles. Both work extensively with “real” players and with virtual instruments. Here they weigh in with responses to some questions that I as a composer learning more and more about mixing posed to them. As you might expect, there are similarities and differences to their approach, which I found very thought provoking. Hopefully you will to.

First up is Tony Shepperd.


Hi Tony, it’s nice to talk with you. One of the challenges for many composers when mixing their own sampled orchestral instrument based compositions is achieving a similar sense of placement in the stereo field and relative depth that makes the sound like a concert hall or scoring stage, which of course can be set up quite differently, correct?

That’s is correct and it’s great to speak to you as well. One of the biggest problems composers face is that they believe just because a reverb comes with their string library, that that is what they should use. And use it often. However I’ve found that when you don’t use the reverb from the string libraries and you setup a few really good reverbs from different software and hardware manufacturers, it’s much easier to let the strings live in a unique space.

Complicating matters is that some libraries, recorded the instruments in their placed positions, so they are essentially pre-panned, while other libraries are not. Firstly, how do you deal with that issue when mixing various libraries?

Some of the smaller or lower end libraries have an interesting way of dealing with stereo. Even though they are trying to simulate real strings, their violins are VERY stereo to the point where they are encroaching on the viola’s sonic space. If you’re not careful it ends up sounding very monophonic. It’s okay to blend various libraries but the problem is the vibratos from various libraries tend to clash. I prefer using one library and changing the parameters from other things such as pan, multiple reverbs and so on.

Some software reverbs have positioning abilities built in to the GUI while others do not, Do you see this as an important component in achieving your panning goals.

Not really. Many times I will use multiple reverbs and just pan the returns harder rather than letting the reverb dictate how it deals with stereo images. Let’s take violins and violas, I would set up two different but similar reverbs for each section. In doing so the reverbs would sound similar but the RT may be longer for the violins, or I may put a touch of Plate or Hall on the violas that the violins don’t have.

Would you say that reverbs, particularly early reflections, are your primary tool for achieving depth perspective? In other words, if you are emulating an orchestra on a concert hall stage and you want the instrument to sound like it is coming from the back of the orchestra.

As I eluded to earlier I thing to do is multiple reverbs. on a typical mix I may use a Room a Hall and a Plate reverb, just to achieve my desired result. It really depends on the sonic space that best fits the song or cue I’m mixing.

There are great sounding hardware reverbs out there but for most up and coming composers or those with low budgets they are too expensive so they use some plug-in reverbs, which have fortunately been getting better and better. A lot of guys use 2 instances of a reverb per orchestral section, one for early reflections and one for tails. Do you favor that approach?

Yes. See above section.

Can you give us some general tips for the different sections’ reverb settings to achieve that sense of depth?

Have a variety of Reverbs from Rooms to Plates to Halls. Occasionally have your pre delay set really high around 80ms. This will allow the sound to “breathe” before Reverb envelopes the sound.

Do compressors, modulation plug-ins, etc. play much of a role in this for you or are they only sonic choices?

They sure do. Every plug-in counts. Anytime you instantiate a plug-in there’s a good chance you’ve changed the sound. Sometimes for better, but be careful that you don’t add so many that you that you’ve changed it for the worse.

Any other tips you care to share?

When it comes to mixing remember there are three dimensions to a mix.
1. The front to back of a mix: For instance, how far up front or how back that vocal sits in the mix.
2. The left to right of a mix: While you are mixing, do you feels as if you are standing in front of a large orchestra or a small quartet? Does your panning not give you the depth, breathed and feeling of an orchestra? Are you panning things so far in that the violins and violas feel as if they are sitting on top of each other.
3. The sonic top to bottom of a mix: Do you feel as if the kick has depth and balls way down at the bottom and the strings feel as if they are lighter than air above your head? Use the full spectrum of a mix to deliver the power and impact the song needs to have. Have the reverbs, delays and harmonizer complement all for the sonics.

Thanks Tony.


And now, John Rodd.


Hi John, it’s nice to talk with you again. One of the challenges for many composers when mixing their own sampled orchestral instrument based compositions is achieving a similar sense of placement in the stereo field and relative depth that makes the sound like a concert hall or scoring stage, which of course can be set up quite differently, correct?

That is correct – there are many possible layouts for an orchestra. Some of the factors that can influence the positioning of the musicians include composer’s preference, conductor’s preference, the acoustics of the venue, and the recording engineer’s preference.

Achieving a good sense of depth in a virtual orchestral mix is difficult for many reasons. One reason is that with a virtual orchestra, the complete sound of the orchestra isn’t being acoustically blended together in a big room (as it is with a live orchestra) it is separate samples being mixed electronically. You don’t have the benefit of a great sounding acoustic space to help things blend together naturally. For example – with a live orchestra, the main room mics capture the overall sound of the orchestra, and the spot mics are used to fill in detail. Virtual orchestral sample libraries don’t work this way.

Complicating matters is that some libraries, like East West’s QL Symphonic Orchestra, recorded the instruments in their placed positions, so they are essentially pre-panned, while other libraries are not. Firstly, how do you deal with that issue?

I request that the person who is bouncing the audio out of the composers DAW to create the multitrack undo any panning that has been applied, so I get each sample library in it’s most basic stereo format. I also have any reverb or EQ or compression removed as well. Once I have all the stereo tracks in front of me, I listen to each track individually, then as instrument sections, then as the entire ensemble – and I adjust panning & reverb & EQ to make it all work together. When I’m mixing a virtual orchestra, or a hybrid of live & virtual orchestra this process can take some time to get it right. As a recording, mixing and mastering engineer, one advantage that I have is that I have been a part of a huge number of live orchestral sessions, in all genres of orchestral recording, (film score / classical / chamber music / pop / rock / experimental / etc) so I really know what the real thing sounds like.

Would you say that reverb is the primary tool for achieving a depth perspective?

Reverb is extremely important to me, but the overall ‘mix’ of the music is equally important. A really good mix will have depth, width, and a good balance of tonalities. To get to this place I have to make literally hundreds of small adjustments to panning, levels, EQ, reverb, compression, delays, and every other aspect of the mix. On their own, each one of these small changes are pretty inconsequential, but it is the total of all of these changes that hopefully add up to a really great mix.

I would say that mixing is a bit like being a fine art sculptor. I have some raw materials given to me, and I need to cut away, accent, shape, add texture, and generally massage the raw materials into something that has depth, drama, dimension, excitement, and beauty. An inexperienced mixer has similar tools to mine, (reverb, EQ, panning, levels, automation, compression, delays, etc) but the trick is to know what to cut away, what to highlight, and what to shape in special ways to make the mix work well as a whole. A great mix that has depth isn’t achieved by simply using one technique or fancy reverb…. but by all the many, many small mix decisions that get you there in the end.

I’ll add that when I get to record the music it is a bit easier for me to get the mix where I want it to be, as I selected the mics and preamps and mic positions to all work well together. With virtual instruments I obviously don’t have that luxury!

I know you are primarily a hardware guy but that you also use some plug-in reverbs. A lot of guys use 2 instances per section, one for early reflections and one for tails. Do you favor that approach?

Over the years I have experimented with that approach a few times but I’ve never stuck with it. I didn’t find that it offered any significant advantages for me.

I’m very lucky to have a plethora of fantastic hardware reverb boxes at my disposal, and with some tweaking they always give me exactly what I’m after.

I do frequently use multiple types of different sounding reverb in a mix, but they are all doing different things sonically, and they are not dedicated to early reflections vs. tails.

Can you give us some general tips for the different sections’ reverb settings to achieve that sense of depth?

I simply evaluate each stereo audio track on the multitrack individually (and as part of the section, and the whole) and I adjust all the elements of the mix to taste. Every sample library and cue and project is unique. This is why I love what I do………. no two days are ever the same!

Do compressors, modulation plug-ins, etc. play much of a role in this for you?

For creating a sense of depth in an orchestral mix – be it live, hybrid or virtual – I’m generally not reaching for compressors or modulation plug-ins. However I’ll use every tool in my toolbox when mixing, generally speaking.

Any other guidelines you can share?

My best advice to the aspiring orchestral composer who is using virtual instruments would be to educate yourself as much as possible. Go listen to live orchestras frequently. You’ll learn many things, and the performers will be happy to see you in the audience.

If you can get into an orchestral recording session (or even a string quartet recording) – sit in the control room, then go sit in the live room (and don’t sneeze). Listen to how the sound blends in the room, and then evaluate how well the recording engineer is capturing and representing the live sound.

Study orchestration as much as possible.

Listen to live orchestral film scores on CD, and then study the actual printed score if you have access to it.

Learn to discern the different production styles that different film scores have used, and decide for yourself what you think sounds good and bad.

Lastly, I would encourage virtual orchestral composers to use as many different sample libraries as possible as they all have their own sonic strengths and weaknesses.

Thanks Jay for including me in this article! It is always a pleasure to be interviewed by you.

Comments

By Paul Henry Smith on February 1st, 2011 at 10:48 am

Interesting and valuable information. Thanks, Jay, for asking these questions! As an orchestral musician, conductor, and now digital orchestra performer, I am simply astounded at the expressive possibilities we can get with reverb, and to a lesser extent EQ and compression.

I often put a different reverb on violins, violas cellos and basses (and then use another overall reverb for the strings or the whole orchestra). These are convolution reverbs usually, and they’re pretty good. The other day, however, I had the pleasure of using a really nice Lexicon hardware unit. The difference in cleanliness and detail (not to mention tweakability) was astounding. Makes it hard to go back to plugins!

Still, it seems that, no matter how painstaking I am, and how high-end the tools, I don’t get the ineffable blend of real sound in the same room. I don’t know what it is, or how it happens, but as anyone can hear by going to live orchestra concerts or sessions, there is something in that sound we can’t get with virtual orchestras. Yet.

I actually tried a very simple solution that worked pretty well: Using dry samples (VSL) and 5 BeoLab 5 speakers, I played the digital orchestra in actual concert spaces. Nothing like “architectural” reverb to beat hardware and software both!

By Richard Ford on February 2nd, 2011 at 9:03 am

Really good article, Jay. Although it doesn’t solve anything for me, only gives more to experiment with. :D I love the recommendation to go to hear live orchestras and recording sessions and score study. I learn something new every time I crack a score open. It sounds like both add reverb and use other tools to “take things away”. I don’t really see how to create depth that way using compression or modulation. I would love more discussion on that.

Leave a Comment