Okay, so you have an idea of how your reverb works, you know where everything goes and you’re eager to show the little virtual players inside your computer to their seats in your custom made software concert hall. Great! So let’s move on to making this happen within your DAW of choice.
Single reverb buss
The easiest way of setting up an orchestral reverb is using a single reverb plugin as a send effect. This will let you put all sections of the orchestra in a single coherent space, and you can adjust their front-to-back placement by simply changing the send levels. The more reverb you add to a section, the farther back it will appear to be, and vice versa. Just like in the example with you and your friend in the big room. Even if this method isn’t 100% optimal, for reasons discussed below, I really think you should try it first. Don’t attempt anything more clever before you’ve explored the possibilities of using a single send reverb.
Why? Because it’s a straightforward solution that will get you a long way. It’s easy to manage, its resource footprint is as small as it gets, and it can sound great provided you have a good reverb and manage to set it up so that it works with all instruments. That’s the main part of the work right there: dialling in a reverb that sounds good on all instruments, both melodic and percussive. After that, all you have to do is apply it in suitable amounts on the different sections and boom, done.
I say “dialling in” because I’m not an advocate of using presets for anything more than as starting points. Most plugins don’t have presets that are intended to be used in real-life scenarios anyway, they’re just there to showcase the features and capabilities of the plugin. So do yourself a favor and make it a habit of always exploring reverbs (and other plugins!) beyond what the presets offer.
If you give this a try, you will notice that when you add a reverb send to a track, it will get slightly louder. This is to be expected as the original dry signal is now doubled with a wet version of itself. Once you have found the right balance between wet and dry, just knock the channel volume back a couple of dB to counter for this if necessary.
The magic behind the Predelay parameter
Using a single reverb buss will always sound a bit flat (in terms of Z-depth that is, not pitch) because all instruments share the same predelay times and amount of early reflections.
Some people claim that instruments in the back should have a longer predelay time than those in front. At a glance this appears to make sense as it takes longer for the sound in the back to travel to the listener’s ears. But if you think about it, it should actually be the other way around. For example, the first seat violin is in the very front of the orchestra and you will hear its direct sound a split second before the reverberations from the surrounding hall kick in. In other words, a distinct predelay. On the other hand, the crack of a snare drum in the back will have become so diffused and mixed with the hall reverberations by the time it reaches your ears that you can’t actually hear any “dry” sound. In other words, no predelay.
Setting up varying predelay times for the different sections is easy if you’re using multiple reverb busses (see below). If you’re using just a single send reverb you might still be able to do it if your DAW lets you route signals freely. Consult the manual if you’re unsure whether it does. I can’t speak for other hosts, but here’s what I’d do in REAPER, which has very flexible routing.
First of all, make sure the reverb is set to zero predelay. Then create two new tracks. Insert a basic delay plugin on each track (just make sure that it’s a stereo delay). Set both delays to 100% wet, 0% dry, and set the first one to, say, 40ms delay time. Set the second to 20ms. Make sure feedback and other setting are disabled on the delays so you’re getting just the clean signal, only delayed slightly. Disable the master sends on the delay busses so the delays are not fed to the master output, then route both tracks to the reverb, 100% wet. Finally, send the front instruments to the first predelay track, the middle instruments to the second, and the back instruments directly to the reverb. Voila — one reverb, three different predelays.
Here’s the hitch though: this is all a compromise. Even with differing predelay times, all sections will still have the same balance between early/late reflections. You will either have to live with that or consider using…
Multiple reverb busses
Optimally you should you use several reverb busses using the same basic reverb sound but with subtly different settings to mimic the way sound waves behave coming from varying distances. I normally use three — front, mid and back — plus a room ambience for various percussion that I want separate from the hall reverb. Here’s a rough guide to how it’s set up and what’s routed where:
Settings: Long(ish) predelay, rich in early reflections, slow attack
Instruments: Strings, harp
Settings: Medium predelay, less pronounced early reflections, medium attack
Instruments: Woodwinds, horns
Settings: No predelay, no early reflections, fast attack, less pronounced highs
Instruments: Trumpets, trombones, tuba, percussion, choir
The above is of course only 100% applicable to my personal setup.
Mixing multiple sample libraries
Setting depths with reverb busses and levels usually works fine on samples that have a natural ambience, but less so on close-mic’d stuff. A dry trumpet with a hall reverb on it doesn’t really sound like a trumpet playing in a hall. It just sounds like… well, a dry trumpet with a hall reverb on it. It doesn’t have that characteristic roominess you hear when listening to a real instrument that is played some distance away. It just floats ambigiously somewhere out there, a dry signal mixed with a wet one.
If you are mixing samples from different libraries and you find that distances are all over the place, and you can’t seem to fix it with only different reverb busses and levels, there’s a very cool freeware plugin called Proximity that might come in handy. Proximity lets you adjust the perceived distance of a recording using various psychoacoustic models, and it can make a close signal sound further away and vice versa.
You can use Proximity as an insert effect if there’s only a few instruments that need tweaking. For adjusting the z-depth of entire sections or instrument families, it’s much easier creating a submix of the tracks you want and adding an instance of Proximity for each of them.
The more the mushier
Never ever use different reverbs as inserts on section channels! If your reverb has presets called things like “Big Strings” or “Percussion Room” or “Brass Hall”, it might be tempting to use these on the respective sections. This is a bad idea for many reasons. First of all, it is almost guaranteed to sound like crap. Those presets aren’t meant to work together and will clash, turning your composition into a mess with no sense of spatial coherence. Secondly, it’s not realistic at all. When was the last time you saw an orchestra performing a piece with the brass in one hall and the strings in another? Thirdly, good reverbs can be fairly processor-heavy and having more instances than you actually need is a complete waste of CPU cycles in case you’re on a more modest machine.
That said, there are of course situations when you might want to use a completely different reverb for one or a few instruments in a mix. Non-typical orchestral percussion, or small percussion loops, are usually best kept separate from the rest of the orchestra. A repeating shaker loop drenched in hall reverb will become a continuous wash of background noise that might drown out the nuances of other instruments and distract the listener. Therefore it’s usually best to put stuff like that in front of everything else, using a much smaller reverb. It’s not realistic, no, but listen to a few action movie soundtracks and you’ll find that it’s common enough not to raise any eyebrows.