Reverb is a beautiful thing. It creates a spatial context for sound and serves as a unifying force for tracks recorded in disparate spaces. While pop mixing allows for great variety and choice in reverb settings, orchestral musicians play together in a large concert hall and this can present some challenges when mimicking this musical setting with samples. When used well, the reverb acts as a unifier for getting all the samples to sit together in their own virtual space, even when using libraries created by different sample manufacturers.
Like many composers, I own a number of orchestral sample libraries.
I have my favorites for different sections of the orchestra and I find that some libraries are more suited for certain styles of music and orchestration. The libraries are most flexible when they are recorded with a minimal or adjustable amount of reverb. When I apply my favorite reverb setting, it can serve as a sonic “glue” for my virtual orchestra and make it sound as if all the instruments are playing in the same hall together.
The prototypical way of using reverb in a digital audio workstation is to create an auxiliary track, insert a reverb plugin and use post-fader sending to send level from the track that you want reverberated. Post-fader means that the signal gets sent to the bus after it hits the fader. If you lower the volume of the track, the level of reverb diminishes by the same amount. To adjust the amount of reverb you want on your track, you change the send level that is bussed to the reverb. Post-fader sending works great in musical situations where we want to hear a dry, unaffected signal mixed with the reverb. In most mix scenarios, we hear more of the dry than the reverberated signal.
Post-fader sending is how we hear reverb applied in most pop situations. The instruments are generally recorded with a microphone close to the source and reverb is added to give a sense of spatial depth using the technique described above. If one moves a microphone further from a source, more of the reverberation of the space that the source is recorded in will be captured on the recording. Close miking gives an engineer a lot of mixing options because we can add reverb to a sound to give it depth but we can’t remove it if it is found on the recording.
This approach of adding reverb to a dry recording works quite well in mixes where we are used to hearing a lot close miked instruments. Guitars, drums, keyboards and vocals are typically miked close to the source. But this approach doesn’t sound natural on orchestral samples. The reason is that, as listeners, we don’t hear an orchestral section from a close vantage point. We listen from a distance and, as a result, the sound of the room plays a big role in the sound of the sections. Simply applying more send level using post-fader sending doesn’t really work because we still hear too much dry, close miked sound.
A workaround for this is to use a type of sending that is generally reserved for headphone mixes. Pre-fader sending sends the signal down the bus to the reverb before it hits volume fader. This allows you to create a track where the reverb is more prevalent then the dry signal, because the fader now controls the level of the dry signal. As you lower the fader, you decrease the level of the dry signal but leave the reverberated signal intact since the bus level is unaffected by the change to the fader. This, in effect, reverses the function of the send and the fader from the post-fader scenario. The send now sets the level of reverberated signal and the fader sets the ratio of dry/wet signal.
This approach works well for orchestral instruments because if an orchestral sample is recorded with minimal room sound I can essentially put it in the hall by removing a greater amount of dry signal than would be possible with post-fader sending. If I want one section to sound further away from the listener than another, I just lower the fader for the more distant section. The result is that the tracks sound as though they are placed in the hall rather than the hall being added to the sound.
The downside to pre-fader sending is that since the fader no longer controls volume you cannot use the automation lane of the track to automate volume. With virtual instruments (VIs) there is a simple workaround – you simply use a MIDI continuous controller (either cc7 or cc11) to change the volume level of the VI. Audio tracks don’t respond to MIDI controllers, but you can to insert a plugin that can change the gain of a track as a substitute for your missing volume control. Most DAWs have a built in plugin designed for this purpose. Logic’s “Gain” and Pro Tools’ “Trim” are examples of plugins that perform this function.
I find that the best reverb plugins to use for orchestral samples are convolution reverbs. Convolution reverbs use impulse responses (IRs) to “sample” acoustic spaces and, as such, are more natural sounding than algorithmic reverbs. My favorite convolution reverb is Altiverb by Audioease, but there are convolution reverbs made by other manufacturers and the principle is the same. Many DAWs include their own convolution reverbs such as “Space Designer” which is found in Logic. These plugins come with IRs of different spaces and give you the ability to load impulse responses of your own.
An interesting result of using a convolution reverb with pre-fader sending is that it has a noticeable effect on the timbral quality of a sampled instrument. Pre-fader sending makes the choice of reverb instrumental to the sound of the VI itself. I’ve been amazed at how running string samples through a good sounding IR can warm up the sound and take off the harsh edges that I often hear in string libraries. It can improve the sound of the samples without the need for corrective EQ.
Many of the newer orchestral libraries allow one to adjust the level of recorded reverb through the control of different microphone positions. The library manufacturers provide samples recorded with microphones at different distances from the section. These are often named something like close, stage and room and you can mix in different amounts of the mic position within the sampler’s interface. If you have a library that is recorded in this manner (Spitfire Audio and Cinesamples are sample manufacturers that take this approach) you don’t need to apply additional reverb as the VI instrument has all the reverb control than one might need. However, each mic position requires additional samples so loading all mic positions for a preset can take a considerable amount of RAM. For this reason, I usually just load the stage mic setting and add additional reverb using pre or post-fader sending, depending on the sound I am trying to get.
If you are mixing libraries that have reverb in their recording with libraries that are recorded dry, it is important that the your reverb matches the one on the recording so that the instruments sound like they are in same space. Choosing a similar type room and matching the reverb decay works quite effectively in most cases. You can figure out the decay time by listening to a percussive sound and timing how long it takes for the reverb tail to die away.
Film music mixers often add some high end digital reverb when mixing orchestras for film. Lexicons or Bricastis might be used to add some additional depth and sheen to the sound. This can simulated with great effect on an orchestral mockup by adding a convolution reverb with an IR (IRs can be made from audio equipment like reverbs and delays in addition to acoustic spaces) of one these units to the master bus. IRs of high end reverbs can be found for purchase on the internet if your convolution reverb doesn’t have a preset for the unit you like. You can then add a little Lexicon shimmer to your mix (I find a 10% mix ratio is a good starting point) to get that film score sound.
Getting samples to sit together in a virtual acoustic space is a key to creating convincing sounding orchestral music electronically. Pre-fader sending can help to create more depth for the orchestral samples in your mix and integrate them convincingly into your reverb’s hall. Used in conjunction with a good convolution reverb, I have found this approach to yield more convincing results with my mockups than conventional post-fader sending.
Originally published in THE SCORE magazine, Volume XXX Number ONE.
Thank you for this very interesting article!