Studio Monitoring – Tips On Creating Theater Sound Cue Projects In Advance
Posted by scott kane - ceo at 6:27 pm
One of the primary purposes behind using a computer for theater sound cue playback is automation and the ability to prepare that automation in advance – at home, at the office or at the studio, or home studio for those of us fortunate enough to own one.
However, as much as it solves many problems it can in fact introduce many new ones. One of these problems is the whole issue of reproduction in terms of the acoustic differences between the location we create a project and the place it’s ultimately performed – live theater.
For this multi part article we’ll be looking very briefly at the issue of studio monitors and reproduction live and the differences. The assumptions we make in our “project creation environment”, henceforth referred to as “studio” in this article are often responsible for some sonically ugly sounds. These can be minimized though with the right tools and the right techniques. We’ll be discussing two issues together, in these articles, because they are specifically related. They are: Sound Levels and Studio Monitoring.
1. Sound Levels
Sound levels, in a software product like MixAction, afford us with something that, until computer software technology, wasn’t easily visualized. So learning about them and recognizing the good, the bad and the insanely ugly is a lot easier than, say back in the early 1980’s when I did my first gig in a pro recording studio. These days most of us are familiar with the “waveform” display.
What many are not so familiar with is what that display can tell us, beyond the obvious “start, end, quiet, loud” peaks and falls that appear to tell the story on first glance. I won’t be going into any of the theory of physics and sound in this article, we’ll keep it simple. So, consider the screenshot below (taken from inside one of MixAction’s editors BTW):
You can click on the image to zoom in a bit closer if you like…
Now. Why is this “ugly”?
Take a look at the largest sections – the “loudest” sections – represented by the gray and black areas in the screenshot. Those “loud” sections are certainly “loud” but they are also “clipped”. Clipping in digital audio is a little different to analogue clipping one might expect to hear from our speakers as “distortion”. Not all digital clipping can be perceived by the average person as “distortion”. Yet – distortion it is. The sacrifice here is that the signal, while very “loud” is lacking in clarity in the audio spectrum. Detail and true frequency prescience is lost.
The track illustrated was created on purpose by me to give you a starting term of reference. If the tops of the waveform are squared off or flat like this you are digitally clipping and the better the reproduction system (speakers) the worse it’s going to sound.
Take a look at this next screenshot, again you can click the image to zoom in if you wish, that shows what the original sound looked like as “recorded” in our studios:
See how there is a wide variation of dips and peaks, the signal to noise ration is good (no noticeable spikes in quiet sections indicating noise) and there is significant variation in the levels. You’ll also notice the waveforms are not chopped off. This is a natural waveform in the sense of a recording, though the content is actually a combination of analogue Moog and DX7 synthesizers and a collection of modern digital and samples synth sources.
The two screenshots – above and this one to the right – are of the same audio signal, the same recording, but the one above has been mistreated.
Now – who would do this to an audio signal? You’d be surprised! In fact in the quest for the “loudest sound” it’s very common to see this both amateur and pro recordings.
In fact take a look at the track “I Kissed A Girl” performed by Katy Perry shown here:
Now I’ve not done anything to the signal except read it up into the analyser in the MixAction editor straight from the MP3 released by studios and sold online.
While there is very little “cut off the top” going on here the signal is extremely dense and there is very little variation. I can’t give you as sample to play of the Katy Perry track here, but I can for the first two images above. So in a moment we’ll use our “ears” to show use what our “eyes can hear”.
But before I do – one more screenshot – the same track as the first two above, not the Katy Perry track:
You’ll quickly notice that while louder than the second screenshot above, it’s not as pronounced as the first example or indeed as pronounced as the Katy Perry track pictured directly above.
Yes, there are some “cut off” points here. Best to avoid this, true, if we can, but there are specific times when this kind of waveform can actually be our “friend” and not our “enemy” and I’m going to explain that too in a moment.
Before I go into it, click the player button below to sample the audio. This is the “untreated” natural waveform, second screenshot in this article:
Notice that it sound kind of natural, but, perhaps, just a little low or even hard to hear on some, or your, computer speakers? We’ll get to why that is so presently…
Now sample the same audio file from the screenshot directly above:
Hear how it’s just that little bit louder?
OK. Here’s the last sample, which is the first screenshot of the waveform we showed you at the start of the article. The resampling to MP3 in this instance has introduced audible distortion too:
See how much louder it sounds compared to the “natural waveform”? I would argue for theater cue usage it’s boosted to far. Remember, normally we’ll be performing through the relatively powerful and usually suspended speaker bins of the PA system. A more natural waveform is going to give us, usually, more desirable results. The place to use the last sample, the one you just heard above, is where we need to broadcast a signal over, for example, the Internet. In fact I use this signal in the video, below, for advertising MixAction’s forthcoming release.
The reason is that computer speakers are designed to play games primarily. They are not designed for accurate reproduction. So to get the signal to a level that it “sounds impressive” you have to cheat a bit. So the signal is boosted. This is also why modern pop songs are boosted, BTW, to levels similar to the “I Kissed A Girl” sample waveform above. These days much of this material is played on iPods and portable players. Boosting to this extreme level makes the song sound “louder” and “louder” is perceived as being “better” by the average listener. As engineers we know this is not actually true, but since when did fashion and trends necessarily stand in the way of the truth – and sending yourself stone deaf? 
In the next article we’ll zero in on speakers in the studio, compared to our PA bins, why we need to use reference monitors, the kinds available and how to fit out a little studio on a budget in the article after that.
Oh…
The video with the boosted signal. It’s here. Yes, it’s corny, yes we know, we know and the interface isn’t the final one, but that of a few months ago in respect of MixAction. But switching in between a recording studio and a software compiler all day every week brings out the corniest in us and we had to get Sony Vegas working properly on 64 bit Vista, which has been a challenge until now. But – I love the Moog synthesiser and it’s an example of where an unaturally boosted signal should be used and that place is not in our theater sound cues and productions…
//
Scott Kane
CEO and Primary Developer MixAction Theater Sound Software
