Don’t have my new mic all dialed in yet. I’m wondering if I’ve been baffled by bullshit. I was referred to these two articles to explain the importance of setting your recording levels correctly. It just doesn’t make sense to me. The idea is, if you set your recording levels low, you won’t record too much of your environment, translating to noise floor. Noise floor is super important when recording audiobooks, because the majority of listeners are using headphones where what’s in the silences is much more apparent than say, recording a vocal track that will be going into a mix.
But I’m not sure these issue apply so much when doing spoken word. Of course, we don’t want to clip, but I would have to work pretty hard to clip something. When I set up the mic and made some first recordings, I was dismayed by what I heard in the noise floor. I mentioned it to Justin when he called to see how everything was working for me. He suggested I try recording without the pre, so I did. It was really soft, so I cranked the gain to the roof, and was able to record something that peaked at about -12db. I couldn’t hear that noisy silence, but I knew I still had to boost the sound levels.
So I took the file and normalized it to -3db, then limited it to -6db. I imagine some sound engineer will get his panties in a twist over that particular move, only because I lack confidence. I think my logic is sound, though. Normalizing makes your loudest sounds only so loud (or you can work it the other way and make the loudest sounds softer) As it does this, it adjusts all the surrounding sounds an equal amount. Seems to me it’s just a way to control your amplification, but again, what do I know? Limiting takes any sound that is above your set level, in my case -6, and smashes that bit down to fit.
One thing you need to understand about sound is we are dealing with negative numbers, so -3 is louder than -6. So when I limit to six, it takes all my sound from -5.9 to the -3 peaks I normalized to, and adjusts their levels to fall into that -6 range. I like the results of doing things this way, since when I get expressive, I get louder and so my levels end up pretty uneven. I don’t know what the pros do, but I think Jared gets better results than I do, so I send my things for him to master. At least my big things. But for auditions and tests, it’s all me, and this is how I do it.
So I got into another discussion with Justin who cautioned me against having that gain cranked all the way up. He thought I should use the cloudlifter and not record so hot. He’s the one who referred me to the video and article, and I applaud him for taking the effort to educate me, but I believe my ears more than anything. So I recorded the way he suggested, and mastered it the same way as described above, and guess what? I don’t hear a difference! I’m not getting anywhere near clipping, either way I do it, and when I amplify the sounds to get the levels where they need to be, it seems like there is the same kind of amplification to the silences as there is to the sound.
It was suggested to me that recording with less gain, I wouldn’t pick up the low frequencies in the first place, so they wouldn’t be there when I amplified, but I don’t hear that happening. So I’m wondering, do I really need that cloudlifter? I could save myself $200 sending it and the extra xlr back, and I’m tempted. I know Justin will never tell me to do that because that is money out of his pocket, but I am just not following the reasoning here.
What do you think? Have a listen to my samples below. Put your headphones on and wait for the second sample. There is more mouth noise on the sample with the cloudlifter, but I think that’s because there way more mouth noise at that moment. I wasn’t planning on making these recordings public, so the read sucks, but I am really scratching my head on this one. I’d really like your input.