Gain staging & automation.

[Jace Face]

Maybe someone can explain. Perhaps the #1 area of importance to a clean mix is the best source sound. And maybe gain falls into that, or maybe gain is #2.
Nevertheless, from what I understand, an audio file contains the very first control of gain/volume regardless of inserts and other faders.

I need to stage my question like this because I’m confused about the audio clip volume a little bit. For example, I get it averaging -18 VU but it’s still jumping around too much. So.. I need to do a volume automation (for smoother compression/easier dynamics control later). However, Reason 13 will only let me control the ends and overall volume of the audio clip. This is a problem. I need to automate within that clip. Am I missing something?

Yes I could razor into different parts, adjust volume accordingly, then cross fade but the process is extremely cumbersome as opposed to practically every other DAW which you simply draw the volume in the clip..

Please assist if it is actually possible in Reason 13. Otherwise, option 2: I must put the clip volume in -18 VU ballpark and then put a gain tool as insert, then automate that volume?

[PhillipOrdonez]

You can’t draw the volume in all other daws , for example in Ableton live you can’t. Just like in reason, you can automate the fader or insert a gain utility and automate that, such is what I would do if I wanted to manually ride the gain I would map a fader to the gain utility and hit record and automate away… but I don’t do that because… why? Are your recordings so dynamic compression doesn’t cut it? Do you have RX? There’s a leveller module there that will automatically level your recording (that’s what I use)

[jam-s]

For "gain staging" in a DAW only peak level really matters, thus using the channel VU meter can be quite misleading. Use the big meter in peak mode and the solo button instead to level your tracks to about -12dbFS.

[Benedict]

I think you are making this hard.

Start with the basics. These are all covered in this video incl what real Gain Staging is:

"Clean Mix" is #1 a result of a well-written piece and a good performance. Then it is a case of letting the Balance of the parts come together - you find Balance by feel (Not by applying Producerynezz as that warps things).

As to what your specific issue here is, it is impossible to say other than what I have and seeing it would help with more specific "I'd do that" suggestions.

[chimp_spanner]

Maybe someone can explain. Perhaps the #1 area of importance to a clean mix is the best source sound. And maybe gain falls into that, or maybe gain is #2.
Nevertheless, from what I understand, an audio file contains the very first control of gain/volume regardless of inserts and other faders.

I need to stage my question like this because I’m confused about the audio clip volume a little bit. For example, I get it averaging -18 VU but it’s still jumping around too much. So.. I need to do a volume automation (for smoother compression/easier dynamics control later). However, Reason 13 will only let me control the ends and overall volume of the audio clip. This is a problem. I need to automate within that clip. Am I missing something?

Yes I could razor into different parts, adjust volume accordingly, then cross fade but the process is extremely cumbersome as opposed to practically every other DAW which you simply draw the volume in the clip..

Please assist if it is actually possible in Reason 13. Otherwise, option 2: I must put the clip volume in -18 VU ballpark and then put a gain tool as insert, then automate that volume?

I think it depends on what exactly you're processing. Some signal sources naturally have a wide dynamic range and lots of variance (DI guitars, vocals, etc.). In these cases a compressor will get you much of the way.

If you're working with something more processed like a music loop and it requires this much automation then perhaps it's not a good source sound to begin with. You shouldn't really have to micro-manage the gain across the length of a clip to get it to play nice with a compressor.

But yeah if I was working with a vocal part and I wanted a word to pop out, I would just razor where I want the emphasis and increase the clip volume. Or alternatively you could just place a gain utility at the start of the chain (Khs do one, I think it might be free? I can't remember!). And then you can just line up automation points with the portions of the waveform you want to boost. Downside of course is that this won't be reflected in the waveform itself.

As for other DAWs letting you draw/edit the volume like this, I'm not sure they all can. It is a feature I've seen in one or two, where volume envelopes can have multiple points and segments within a clip, not just fade in/out. But it's definitely not a universal thing. Maybe it's gradually becoming more commonplace though.

[selig]

Maybe someone can explain. Perhaps the #1 area of importance to a clean mix is the best source sound. And maybe gain falls into that, or maybe gain is #2.
Nevertheless, from what I understand, an audio file contains the very first control of gain/volume regardless of inserts and other faders.

I need to stage my question like this because I’m confused about the audio clip volume a little bit. For example, I get it averaging -18 VU but it’s still jumping around too much. So.. I need to do a volume automation (for smoother compression/easier dynamics control later). However, Reason 13 will only let me control the ends and overall volume of the audio clip. This is a problem. I need to automate within that clip. Am I missing something?

Yes I could razor into different parts, adjust volume accordingly, then cross fade but the process is extremely cumbersome as opposed to practically every other DAW which you simply draw the volume in the clip..

Please assist if it is actually possible in Reason 13. Otherwise, option 2: I must put the clip volume in -18 VU ballpark and then put a gain tool as insert, then automate that volume?

This is all theoretical, so difficult to give precise answers.
To start, I don’t often deal with audio signals that have huge level swings. Rare cases where this could happen on my projects is a vocal that alternates between a very soft and very loud section, which is super easy to deal with using clip gain.
So if you are getting tracks that require a high level of gain ‘riding’ at the source, I can say this is not common because most folks address it at the source if it really is that extreme.

But if you imagine what you’re asking, which if I’m understanding it is you want to reduce the most extreme level changes on an audio recording, there are other approaches. Going against the grain, I would first use a gentle compressor like an LA-2a style to get things more in the ball park - this is essentially what you would do with clip gain or (if your DAW supported it) clip automation. Others may say to start with a harder ‘limiter’ to catch the peaks, but that’s not what you’d be doing with clip gain etc. I find that approach too aggressive if the dynamics are wide enough to be concerned with such things in the first place, because you’ll end up with words/lines that are totally squashed and much of the rest of the vocal left alone. The result is often audible if it is enough to level things out in most cases, at least to my ears!

My theory is that by getting the overall dynamic range gently reduced first (as would be done using clip gain), and THEN using a limiter approach later in the signal path will be less obvious. Remember that in mastering, the more gentle ‘glue’ compressor is before the more aggressive brick wall limiter and I simply take the same approach with individual channels. The hoped for result is occasional light limiting vs extreme limiting on some occasions.

The good news is that as you work with more seasoned artists, there will actually be LESS work for you to do on your end. Seems cruel that when starting out you work with others who are also starting out and all you really learn is how to polish turds. Luckily, in this day and age you can download professionally recored tracks to practice mixing, which allows you to work on MIXING and not FIXING (one of my audio mantras is “less time fixing means more time mixing”)

It is very possible my approach won’t work for you, but when you get feedback/advice from multiple sources it gives you options for things to try out and pick and choose what works for YOU. That’s how I learned, from watching various pro engineers work, each with a different approach - with the lesson being “there are many paths to the top of the mountain”, since all these engineers got equally fantastic results each using their own methods.

There is no “one real path” to a good mix, there is only the best path for YOU. IMO, as always…

[Jace Face]

Thanks guys. I wasn’t specific enough. This is my own music. It’s all me. And I’m referring to a few raw NASA audio files, my own vocal recordings, DI guitar & bass.. stuff like that. I’m wanting to do an initial volume adjustment to “soften” the audio before any processing is done at all. I’ve read not to normalize this stuff. Or is normalizing ok?

I also watched your peak consistency video Selig. Very good information. And I have the Leveler. But I was looking to even the volume fairly well before I start with anything else. Great ideas everyone. I’ll give it a go.

[chimp_spanner]

Thanks guys. I wasn’t specific enough. This is my own music. It’s all me. And I’m referring to a few raw NASA audio files, my own vocal recordings, DI guitar & bass.. stuff like that. I’m wanting to do an initial volume adjustment to “soften” the audio before any processing is done at all. I’ve read not to normalize this stuff. Or is normalizing ok?

I also watched your peak consistency video Selig. Very good information. And I have the Leveler. But I was looking to even the volume fairly well before I start with anything else. Great ideas everyone. I’ll give it a go.

I mean with regards to DI guitar and bass, IMO, the dynamics of that audio are like...you/your playing/your sound. So just let your guitar effects react to it how they will and if you don't like it, try another take, change up your playing style, whatever. If you're getting reeeally wild spikes in volume or things are unpredictable, maybe look at things like pickup height (I've had them set too high before and sometimes notes will just pop out and overdrive the input), check batteries if you have active pickups too as that can cause the output level to dip or fluctuate.

More than anything though I'd try not to overthink it too much. Just trust your ears. Whatever dB you're hitting going into a plugin, if it sounds good to you, it's good. I mean obviously the input level should be healthy, the noise floor low, the quality generally good. But ultimately it's the feel and vibe that the mix and the song evokes. I never aim for specific dB or frequency values. Sometimes I just close my eyes when I tweak a control because I don't want anything to influence my decision making like "that's too loud, that's the wrong frequency". How it sounds is all that matters

[selig]

As for long form recording such as NASA stuff that was super dynamic over time, I’d just do a fader ride and bounce in place. One trick there is to record the automation as a ‘response’ to what you just heard, then slide the automation back in time accordingly. What I mean is you hear something get too loud and you pull it down in response rather than trying to anticipate every change and nail it perfectly as it happens (which is next to impossible). The automation can be edited after the fact anyway, no need to try to get it right the first time. Then bounce and use the newly leveled track as the ‘new’ source track for further processing. Make sense?

For the instruments, spend as much if not more time in learning how to get better recordings as you do in learning how to fix poor recordings! It was mind blowing to me the first time I put a decent mic on a decent player playing a decent instrument in a decent room. Before that, all had been ‘crap’ from the mic to the room. But once I had experienced a ‘level up’ on all fronts, it blew my mind how EASY the rest of the process was.

Fixing poorly recorded tracks is soul sucking IMO. And you know that even after you spend hours on the track it will at best be ‘passable’. And by that point I’ve drained all my creative energy and am ready to call it a day. Contrast that to STARTING the same process with “passable” recordings/performances, and spending a few hours making it actually SHINE, which is far more satisfying and leads to better results all around in my experience.

Yes, there are hacks and solutions to dealing with tracks like these, and every engineer should have some experience with basic repair work for those rare occasions you have to try to save a ‘dying patent’! But as always, “garbage in/garbage out” is a cliché for a reason…