This is an extremely short version of a MASSIVE thread on one of my favourite sites, Gearslutz. The original post is by Skip Burrows and soon evolves into an in depth discussion with Paul Frindle – ex SSL designer & general digital audio boffin. Since reading this thread and putting its various ideas into practice, the sound of my mixes has noticeably improved, so I thought it might be worth sharing with anyone who’s also looking to improve their sound. Of course, if you’re looking for the best results, then you’ll need to hire ME! 🙂
If you’ve been recording as long as I have (nope, you’ll have to guess!), you’ll have learned that the best way of recording onto tape with the best signal to noise ratio was by recording as hot as possible (as close to 0VU as possible). Even the early days of digital recording in 16 bit favoured this methodology, as the argument was that the hotter you recorded, the more of the original quality of the signal would be preserved in the conversion.
Nowadays, that’s all changed with 24 bit recording at 96 and 192khz with the increased dynamic range, plus DAWs are not calibrated the same way as most mixing desks. Long and short, recording at much lower levels is more ideal in the current scheme of things and is one step of many that will add clarity and definition to your mixes. On top of this, the 3rd party plug-ins we all love to use don’t operate at their best with a hot signal, most manufacturers assume that the track won’t be slamming the VUs as it passes through, so making sure that your signal goes in clean will get the best sound out of the plug-in.
One way that engineers are working this with hot audio tracks sent to them for mixing is to insert a trim plug-in first in the chain with a dB cut. There are divided opinions on how much gain to trim, but the thread mentions anywhere from -8 to -20dB. Quite a difference huh! Well, as with everything, I just used my ears and worked out over time what works best for me and eventually opted for anything from -8 to -12dB cuts, which often get made on the input stage of the UAD plug-ins. You’ll make up some gain anyway as tracks are processed and more plug-ins are added, but bottom line, careful gain-staging from start to finish is going to make an enormous difference to the overall sound and will make your mixes less squashed and narrow sounding.
“But what if I WANT it to sound distorted?” I hear someone ask. Good question. Distortion was originally obtained in the analog world by pushing 0dB, however, digital distortion over 0db is really nasty & crunchy – not a pleasant sound at all, what’s more it can adversely affect the CD pressing or even glitch on air, so is to be avoided at all costs**. The pleasing ‘analog’ distortion we’ve all come to know and love is more accurately referred to as ‘harmonic distortion’, being that it IS the adding of harmonic frequencies to the original signal. This is the domain of a whole new generation of harmonic distortion and saturation plug-ins, such as the Soundtoys Decapitator, iZotope Trash, UAD Studer a800, Waves NLS etc. This, then is how to add acceptable distortion (however nasty and trashy you want it) in the context of a carefully gain-staged mix.
Lastly, I tend to create groups, in the same way I used to on my mixing desk, so a drum bus, vocals, guitars, keys and maybe bass separately, depending on the style, complexity of my sound set etc. Having stereo subsections of my instruments gives me more flexibility on the final stage and let’s me pull things down a little if they get too hot. Alternatively if everything needs pushing a little to get the mix pumping, it’s still a much easier way to do it than going through hundreds of tracks adjusting them.
If you’ve enjoyed this article or found it helpful, please click on the Facebook links and like our page. Feel free to leave a comment if you have something to add or ask, it’s always great to hear from people with fresh ideas.
Don’t forget to check out the original post if you’re hungry for more technical info.
Thanks for reading!
** Any mastering engineer worth her salt will tell you there’s a further consideration when converting to mp3, as anything mastered too close to 0VU can take it over the top during the conversion process. This is why most mastering houses will set the output of their brick wall limiters anything from -0.1 to -0.3dB.