Chapter 16: Mixing With Reverb
AUDIO FILES (To download all WAV examples at once: 329MB ZIP)
- Choosing Reverb Presets For Blend/Size: I've used the following sample file (Ex16.01:WAV/MP3 ) to feed a variety of different reverb presets in order to illustrate some of the issues to consider when choosing between them. In all these cases you'll hear only the effect, without any of the dry sound mixed in. The first four presets are ones which sound as if they would suit blend/size-enhancement roles, because they have some sense of acoustic realism about them and exhibit few unmusical metallic overtones. Preset 1 (Ex16.02:WAV/MP3 ) is a 'Soft Chamber' with a rather muted sound and truncated tail, which would suit blend situations, although I'd probably shorten it a little further for this and roll off some high end. Preset 2 (Ex16.03:WAV/MP3 ) is a 'Small Chamber' with a longer tail and a pronounced attack 'bump' (which rather compromises the realism for me). For blend-enhancement, I'd shorten this, while for size enhancement I'd look to smooth or tempo-sync the attack bump, and probably remove some high end again. There's a suspicion of pitched resonance at around 495Hz too, which might need to be notched in some applications. Preset 3 (Ex16.04:WAV/MP3 ) is one of my favourite Lexicon 960L size-enhancement presets: 'Acoustic Fill'. I like the smooth and neutral decay, as well as the nice sense of space, although I usually end up reining in the frequency extremes a little with EQ for chart styles. Preset 4 (Ex16.05:WAV/MP3 ) is a nice dense and fairly believable room sound which could be shortened into a blend reverb if you wanted. However, there's a pitched resonance at around 360Hz which I'd probably want to zap with an EQ notch in most cases.
- Choosing Reverb Presets For Tone/Sustain: The next four presets are all the kind of thing I'd associate with tone/sustain-enhancement effects. Preset 5 (Ex16.06:WAV/MP3 ) is a 'Medium Warm Hall' preset which is trying to be realistic and failing badly. However, this patch has a decent stereo image and a characteristic glassiness to the tone, without too much in the way of attack 'bump', so would potentially work well for mid/high-frequency sustain enhancement. Preset 6 (Ex16.07:WAV/MP3 ) has an interesing metallic tone, but has dreadful mono compatibility, so I'd be wary of using this one unless I was planning to narrow its stereo width a good deal. Preset 7 (Ex16.08:WAV/MP3 ) is a 'Concrete Room' patch which is exactly the kind of thing that I find tends to work well for tonal modifications, because of its rather unnatural timbre. However, I'd probably plan to shorten it a little for most mix situations. Preset 8 (Ex16.09:WAV/MP3 ) has a very pronounced resonance at around 445Hz, which could work well if in tune with your arrangement. There's a lot of high-frequency information here, though, and I'd expect to tone this down somewhat with EQ in the majority of cases.
- Setting Reverb Time: Here's a demo mix completely without reverb (Ex16.10:WAV/MP3 ). Adding in a reverb which is too short leaves the mix a bit lacking in warmth (Ex16.11:WAV/MP3 ), while a reverb which is too long swamps the mix details and gives itself away as an unnatural effect (Ex16.12:WAV/MP3 ). Between these two extremes lies a more appropriate setting (Ex16.13:WAV/MP3 ).
- Equalising The Reverb Return: Here's a demo mix which uses a single reverb, complete with EQ in the return channel to focus its effect into the most suitable frequency regions (Ex16.14:WAV/MP3 ). Now I'll bypass each of the EQ bands in turn to demonstrate their impact on the effect balance: high-pass filter bypassed (Ex16.15:WAV/MP3 ); low-pass filter bypassed (Ex16.16:WAV/MP3 ); and midrange peaking-filter cut bypassed (Ex16.17:WAV/MP3 ).
- Adding Predelay: This demo mix again uses a single reverb, but in this version the reverb uses no predelay (Ex16.18:WAV/MP3 ), whereas in this version (Ex16.19:WAV/MP3 ) there is a 35ms predelay. In this scenario the difference is quite subtle -- it can help to focus on how 'close' and 'clear' the lead vocal sounds in each case.
- Blend Reverb: Blend reverb is frequently useful for blending a selection of drum close-mics together into a more cohesive-sounding instrument, as in this example (without reverb Ex16.20:WAV/MP3 ; with reverb Ex16.21:WAV/MP3 ). Here's an example of blend reverb in action over a full mix: the first file is completely dry (Ex16.22:WAV/MP3 ), whereas the second has a touch of blend reverb applied (Ex16.23:WAV/MP3 ).
- Size Reverb: Following on from the blend reverb example above, here's the same dry demo mix (Ex16.22:WAV/MP3 ) with a dusting of size reverb instead (Ex16.24:WAV/MP3 ), and here it is again with both treatments simultaneously (Ex16.25:WAV/MP3 ). Here's an example of subtle size reverb helping to imply a slightly larger space around drums (without reverb Ex16.26:WAV/MP3 ; with reverb Ex16.27:WAV/MP3 ). Another illustration is the feather-light size reverb here, this time applied to all the instruments in an acoustic jazz trio (without reverb Ex16.28:WAV/MP3 ; with reverb Ex16.29:WAV/MP3 ).
- Tonal Reverb: A lot of the reverb treatments I've used in Mix Rescue have tonal elements, but the best example of pure tone-enhancement is from this Mix Rescue, where I used it on this dry Cello DI signal (Ex16.30:WAV/MP3 ) to lend it some life and resonance (Ex16.31:WAV/MP3 ).
- Sustain Reverb: For this Mix Rescue project I used two different sustain reverbs, both using plate-reverb emulations. Here are the reverb returns in isolation (Ex16.32:WAV/MP3 ; Ex16.33:WAV/MP3 ) and also within the context of the completed remix (Ex16.34:WAV/MP3 ). Notice how each effect has a very different tone, and are each applied to a different balance of the instruments.
- Spread Reverb: Here's an example of a reverb being used primarily for stereo spread, although with some inevitable tone/sustain side-effects (without reverb Ex16.35:WAV/MP3 ; with reverb Ex16.36:WAV/MP3 ).
- Blending By Other Means: Although I've already demonstrated that you could blend this dry track (Ex16.37:WAV/MP3 ) using reverb, there are a number of alternative treatments which can also blend tracks together more surreptitiously, as in the following demonstrations (each of which appears first in isolation so that you can hear its characteristics clearly, and is then toggled in/out of the mix every bar so that you can more easily hear the subtle changes): simple tempo-sync'ed stereo delay (solo Ex16.38:WAV/MP3 ; in the mix Ex16.39:WAV/MP3 ), synth pads (solo Ex16.40:WAV/MP3 ; in the mix Ex16.41:WAV/MP3 ), and Foley 'room tone' (solo Ex16.42:WAV/MP3 ; in the mix Ex16.43:WAV/MP3 ). Another couple of blending tricks can be heard on the lead vocal in these examples: subtle double-tracking (solo Ex16.43:WAV/MP3 ) and gentle stereo widening (solo Ex16.45:WAV/MP3 ; in the mix Ex16.46:WAV/MP3 ). Although these effects are all fairly subtle on their own, they can be surprisingly powerful in combination, as in this example where I've added all these alternative blending treatments simultaneously, once again toggling the effect every bar for better audibility (Ex16.47:WAV/MP3 ). For a simple real-world example, here's a Mix Rescue project where I used simple background noise (Ex16.48:WAV/MP3 ) to create additional blend (with noise Ex16.49:WAV/MP3 ; without noise Ex16.50:WAV/MP3 ). For a slightly more involved example, check out this Mix Rescue, where I used reverb, electrical background noise, and low-level synth drones together. This is what those added elements sounded like in isolation (Ex16.51:WAV/MP3 ) and the following pair of files show the effects in context (with additions Ex16.52:WAV/MP3 ; without additions Ex16.53:WAV/MP3 ). There's another more extended example in this Mix Rescue, where I added various sampled sound effects and ambient noises, as you can hear isolated in this file (Ex16.54:WAV/MP3 ). Now compare the completed remix (Ex16.55:WAV/MP3 ) to these alternate versions without the samples (Ex16.56:WAV/MP3 ) and without any reverb/delay send effects (Ex16.57:WAV/MP3 ) so you can get an idea of the impact of each on the production sound.
- Soloing Reverb Returns In Steinberg Cubase: I've had several readers remind me of something that's annoyed me for ages: when using Steinberg Cubase for mixing, by default it's not impossible to solo an FX Return channel without also soloing its feeder tracks. This makes it difficult to audition the 'wet' reverb-only signal from a plug-in while setting it up, as suggested in the book. My usual workaround for this is to switch the reverb send to pre-fade, solo the return, and then fade down the dry signal. (Alternatively, you can just set the dry channel's output assignment to 'No Output'.) However, a more elegant solution is to do your monitoring through Cubase's onboard 'Control Room' facility, because this gives you an additional Listen button on every channel which does enable you to audition the wet-only reverb signal from FX Returns.
- Affordable Algorithmic Reverb Plug-ins: Here's a large selection of freeware algorithmic reverbs -- they don't all sound very hi-fi, but then (as I discuss in the book), that's often not the point: Smartelectronix Ambience32-bit; Voxengo OldSkoolVerb; PSP PianoVerb; Kjaerhus Audio Classic Reverb32-bit; Variety Of Sound EpicVerb32-bit; Silverspike RoomMachine 84432-bit (and also their affordable Reverb.it); DaSample GlaceVerb32-bit (the developer's site is no longer available, but you can still download the plug-in here); Togu Audio Line Tal-Reverb; the reverbs in the Antress Modern plug-in collection32-bit; GSI TimeVerb32-bit (and also their affordable Spring Reverb Type432-bit); Martin Eastwood MVerb32-bit (the developer's site is no longer available, but you can still download the plug-in here); Greg Schlaepfer Mutant Reverb32-bit.
- Affordable Convolution Reverb: To use convolution reverb, you need a convolution plug-in and a set of impulse-response files. The best-known freeware convolution engine is Christian Knufinke's SIR32-bit. (His reasonably priced cross-platform SIR2 is one of my all-time favourite plug-ins, which I use it for most of my mixes.) However, an alternative is Liquid Sonics's freeware Reverberate LE32-bit or their very affordable cross-platform Reverberate). There are free impulse responses all over the web, but a good source is the Echochamber (German-language) site. For a good general-purpose set of impulse responses, try the Lexicon 960L impulse responses from Echo Chamber -- here are direct links, in case the idea of reading a German website gives you the willies: ambiences (2MB ZIP), chambers (5MB ZIP), halls (12MB ZIP), plates (5MB ZIP), rooms (3MB ZIP). Samplicity have a great set of donationware impulses from the Bricasti M7 (which isn't on the Echochamber site) for download here.
- Reverb Effects In Practice: There are reverb tips to be had in most of my Mix Rescue articles, but there are a few where I discuss the subject of mix reverb in more depth: (article 1, article 2, article 3, article 4). It's also worth mentioning this article simply to demonstrate that there are some situations where reverb isn't really called for. If you'd like a little more information on classic reverbs and their use, check out this article and this article.
- More About Convolution: Convolution processing is capable of more than just reverb. Here are a couple of articles which discuss some of the further potential of the technique: article 1, article 2.