Reverb Challenges in a Large Room
Episode 9 is the only one to date which takes place entirely in one "room." Namely, the fighting arena. Coming up with a large room reverb isn't terribly difficult: you want to give it a nice sense of space, but also a little extra echo, which can be accomplished by an elevated predelay setting. Getting the room to sound spacious is a piece of cake, really. The more challenging part is knowing when to use it and when not.
The danger one runs into is, it's very tempting (and seems very logical) to just route all your sounds to that arena reverb. After all, everyone's in the same room, and subject to the same ambience, right? Well, sort of.
Everyone is indeed occupying the same room, but the key is, reverb communicates not only a sense of surrounding space, but more importantly, a sense of distance. If a friend and I walk into St. Peter's
(and manage not to get summarily tossed out for heresy
) and my friend starts chatting to me while we stand a few feet apart, the primary sound I'm going to hear is what is coming directly from his mouth. If we then split up and go to opposite ends of the cathedral and resume our conversation, I will likely hear much more of the room. In fact, it's likely that the crispness of my friend's voice will be lost entirely in the natural reverb of the place.
It's over simplified, but this is the essense of the decision here in our arena: to reverb or not to reverb. As such, most of the dialogue is run through very little effect at all, whilst the punches and kicks - especially while hearing them from a distance - are resonating heavily.
It turns out that this helps us a great deal in clearly discerning what is being said... but that was not our purpose, merely a side effect of trying to be "realistic" which happens to work very well in our favor.Ready to Rumble
For the announcer voice, I decided to embrace the anachronism of making him sound amplified because it just worked so much better than without. Getting this effect is more than just applying an appropriate reverb. In order to accomplish that almost radio-like amplification sound, you need a heavily compressed signal (much as they do in radio). Here is a screenshot of the compressor/limiter settings I ended up settling on:
The Compressor (section 1) and the Limiter (section 3) are the only sections activated. What a compressor does, essentially, is squishes down the audio signal so it has a more limited dynamic range. This can make it easier to boost the overall volume of the signal without blaring / distorting. Normally, compression should be used with some caution, as it does distort the signal and can make it sound very unnatural. In this case, we are wanting a sound that is a little less natural and a little more cranked.
So, the threshhold is set to -36 decibels. This means that any part of the signal passing through this plugin which is louder than -36 dB will be affected, and anything quieter than that will not. A threshhold this low captures every utterance of the vocal, but leaves virtually silent parts alone. Next, we set the Ratio to nearly 5 to 1; this is the degree to which the signal above the threshhold will be compressed. 5:1 is commonly considered a "safe mark", after which compression starts to sound really drastic and not at all smooth. What 5:1 amounts to is, if the input signal goes up 5 decibels, the output signal will only go up 1 decibel. This flattens out the signal quite a bit, makes it easier to control.
Attack and release have to do with how quickly the compressor turns on and off when invoked by a signal crossing the threshhold. For our purposes, we want it to be lightning fast, so the attack is set to 10 milliseconds, and release is set to Auto (which will be very fast and forgiving).
Finally, we boost the MakeUp Gain by 21.5 decibels. This is volume added to the signal after compression. Since we sqiushed it down so much, we really needed to bring up that final compressed signal quite a bit, and this is the most effective place to accomplish that.
The Limiter, by the way, is just a safeguard to keep the signal from ever crossing the 0.0 dB mark. With the amount of compression we've applied, it's almost unnecessary, but I tend to leave it in place out of habit.
Those five controls are the essense of compression, and they can be very useful for boosting a signal in a controllable way, or "warming up" a vocal sound in a musical performance. You can find several freeware and/or affordable compressor plugins at KVR Audio
, and many audio editors (like Goldwave
) have some kind of compression effect built-in.Marrog, The Brute
How, you may ask, did I get Marrog to sound so... massive and beastly? What I exercised here is a principle that every sound engineer learns eventually: I did NOTHING*.
Know when to do nothing. Tom Urie
provided a mind-blowing voiceover track for this character, it required nothing.
* Okay, that's not entirely true. There was one scene in which I boosted his EQ a bit to give him even more mass... but that wasn't Tom's fault, it was the fact that he was competing with so much background noise and we just really wanted him to scare the crap out of you. :) So, barring that, nothing.