Proper Tracking Levels for Noobs

[The Audio Cave]

Rather than post this in response to a generic question about levels (like I did in the mixing forum and then deleted) I'll do it here... there is a HUGE misconception about tracking levels when recording in 24-bit daws.

Instead of going into detail of why not I'll just say trust me on this. If you're running a mic into a preamp and tracking to daw peaking at -3 or -4, and you're using most any "budget" preamp you might be doing the wrong thing. Lowering the tracking levels can possibly make an audible difference.

Edit/Clarification: Actually, keeping any signal (keyboard, DI bass, whatever) down a little more is a good idea for reasons which will be clear later and have nothing to do with the "budget" preamp reason. The cheap preamp reason is a tracking reason while the others are mixing related. The preamp thing even includes more expensive pres but for a mixing reason. The focus was initially on budget pres because that's what we start off with.

I'll explain later if you want (or maybe someone else will) but record using -10 on the digital scale as a peak reference. Try to make most of your incoming signals peak around -10. There's a really good reason for this which will be partly evident when you try it a few times.

In the interest of keeping this post short... trust me.

 

[Nick The Man]

whats the really good reason for this

more headroom
no clipping
no distortion

 

[The Audio Cave]

more headroom
no clipping
no distortion

Correct on all 3 counts. Now tell the noobs the reasons why they don't need to, and often shouldn't use, most of the rest of that headroom on individual tracks. Put it in perspective for them.

As a new member here I'm trying to offer something useful to the forum by starting a helpful discussion without coming off as a know-it-all jerk. Which seems to happen all by itself sometimes.

Thanks Nick.

 

[Nick The Man]

lol yeah i by no means am a know it all im like one step above a noob

actually my problem is: i know alot of stuff but then when i try to apply it to my techniques and such i dont get desired results.

But anyways welcome to the board

you should check out fordvan's "secrets on mixing" thread some really good material in there and worth reading

 

[RAMI]

There are a couple of recent threads about this. But it can't be repeated enough.

 

[travelin travis]

lol yeah i by no means am a know it all im like one step above a noob

actually my problem is: i know alot of stuff but then when i try to apply it to my techniques and such i dont get desired results.

But anyways welcome to the board

you should check out fordvan's "secrets on mixing" thread some really good material in there and worth reading

if you read it you would know that he read it too.

btw, i'm in the the same boat. too much reading and not enough doing.

 

[The Audio Cave]

There are a couple of recent threads about this. But it can't be repeated enough.

In that spirit, and staying brief as possible, I'll address the first reason.

You track through your analog pre to daw and you push the pre to peak at, say, +4 on it's VU meter. Let's say you do this so...

1. You get a better signal to noise ratio on the recorded signal.
2. Because maybe the pre takes on a different flavor at +4 than at 0, giving up some subtle pleasing distortion (tube compression?) that you really like on this particular track.

Reason #1: You've stayed (hopefully) within the designed operating range of your analog mic pre and the signal going to your converters is a really good one.

Now you can turn up the output gain on the pre (post preamp circuit) so that it peaks around -10, if there is an output gain adjustment on the pre. If not you'll probably fall around -14 or so in the daw.

Reason #2 (which I'll leave to others) would be why you probably shouldn't use the output gain control (if there is one) to bring the recorded signal peak up to -3 or so, even though it wouldn't distort the analog stage of the converter or digitally clip.

 

[travelin travis]

Reason #2 (which I'll leave to others) would be why you probably shouldn't use the output gain control (if there is one) to bring the recorded signal peak up to -3 or so, even though it wouldn't distort the analog stage of the converter or digitally clip.

I would like to know the answer to that.

 

[The Audio Cave]

actually my problem is: i know alot of stuff but then when i try to apply it to my techniques and such i dont get desired results.

You have to realize that "knowing" something and "understanding" something are two different things entirely. Reading something that tells you to "do A, B and C" to get X (even when you don't even understand it) can be very effective in practice for technical things such as analog gain staging or tuning your daw for audio. It works.

For more creative things like mixing tips that often isn't true. You can be told "Do A, B and C" to sound like X and do it exactly as told and not get the result you expect for many different reasons.

On the technical side A, B and C are practical and measurable. On the creative side they're often subjective and the "great mixing tip" can't accurately describe the entire thing, only the basic practical (technical) steps leading into it. You have to do learn to do the rest on your own.

Which is why it often doesn't work as well. It works much better in person when you can see it AND hear it.

I need sleep... I'll visit with you guys again.

 

[SouthSIDE Glen]

I'll just say trust me on this.

AC,

What you say is absolutely correct, and is IME one of the three biggest problems rookies have with the whole audio production process. But asking the forum to "trust you " - regardless of who is doing the asking - is an exericse in futility.

This is the Internet, where any clown with a keyboard and a demarc thinks that all opinions are created equal. While the limited number of early posts I've seen from you so far have been pretty spot on, you'll have to build up a track record (read as: *real* reputation, not fake rep points) before you find people "trusting" you, and even then it'll unfortunately be an uphill battle.

There is a real Rev. Jim Jones mentality on these boards, where logic is meaningless and name is everything. If you came on here and said you were Rick Rubin, people would flock to you and believe you and quote you as gospel if you told them that mixes sound better if you wear a tinfoil hat and jump up and down on one foot while mixing on Tuesdays. But as an unknown entity, you can come on here and give perfect mathematical, legal, and sensible proof, complete with Powerpoint presentations, eyewitnesses and smoking gun evidence that the hat, the jumping and the day are absoluely irrelevant, and you will only be believed by those few who already know that answer to be correct, and labeled a blastphemer by the rest.

Just giving you fair warning . You're off to a good start from what I've seen, but it's waaaay to early to expect anything resembling "trust" from the Great Unwashed. Welcome and good luck

G.

 

[RAMI]

AC,

What you say is absolutely correct, and is IME one of the three biggest problems rookies have with the whole audio production process. But asking the forum to "trust you " - regardless of who is doing the asking - is an exericse in futility.

This is the Internet, where any clown with a keyboard and a demarc thinks that all opinions are created equal. While the limited number of early posts I've seen from you so far have been pretty spot on, you'll have to build up a track record (read as: *real* reputation, not fake rep points) before you find people "trusting" you, and even then it'll unfortunately be an uphill battle.

There is a real Rev. Jim Jones mentality on these boards, where logic is meaningless and name is everything. If you came on here and said you were Rick Rubin, people would flock to you and believe you and quote you as gospel if you told them that mixes sound better if you wear a tinfoil hat and jump up and down on one foot while mixing on Tuesdays. But as an unknown entity, you can come on here and give perfect mathematical, legal, and sensible proof, complete with Powerpoint presentations, eyewitnesses and smoking gun evidence that the hat, the jumping and the day are absoluely irrelevant, and you will only be believed by those few who already know that answer to be correct, and labeled a blastphemer by the rest.

Just giving you fair warning . You're off to a good start from what I've seen, but it's waaaay to early to expect anything resembling "trust" from the Great Unwashed. Welcome and good luck

G.

So lemme guess....You write children's books for a living???

 

[SouthSIDE Glen]

So lemme guess....You write children's books for a living???

You have NO idea how many years it took me to come up with "See Spot run". Turned out to be a classic!

G.

 

[RAMI]

You have NO idea how many years it took me to come up with "See Spot run". Turned out to be a classic!

G.

There's a local band here called "See Spot Run", I always though that was a great name. Easy to remember for anyone who graduated Grade 1.

 

[iqi616]

Correct on all 3 counts. Now tell the noobs the reasons why they don't need to, and often shouldn't use, most of the rest of that headroom on individual tracks. Put it in perspective for them.

As a new member here I'm trying to offer something useful to the forum by starting a helpful discussion without coming off as a know-it-all jerk. Which seems to happen all by itself sometimes.

Thanks Nick.

Other benefits...

You spend less time watching for overs and more time paying attention to what the performer is doing and helping them give their best performance. If you are the performer you can ignore the recorder from the moment you hit record until the moment you hit stop and simply trust the machine to do its job while you concentrate on yours.

Faders will be closer to the top of their travel when you mix thereby giving you more control.

If you aim for a similar target level when you mix you are far less likely to run into the situation where your mix runs out of headroom and you have to pull down all the faders a bit and in the process lose that nice tight sound you spent the last four hours achieving. Hot tracks are achieved at the mastering stage. In 24-bit you are not losing any discernable quality unless your signals are peaking at something like -36 dBFS.

 

[The Audio Cave]

Other benefits...

Faders will be closer to the top of their travel when you mix thereby giving you more control.

If you aim for a similar target level when you mix you are far less likely to run into the situation where your mix runs out of headroom and you have to pull down all the faders a bit and in the process lose that nice tight sound you spent the last four hours achieving. Hot tracks are achieved at the mastering stage. In 24-bit you are not losing any discernable quality unless your signals are peaking at something like -36 dBFS.

Dead on target and covering all but one of the mixing reasons. I'm too sleepy to stay with this tonight. I just had to take a final peek at this thread.

Pay attention to this man, he understands what his daw is doing and why.

Great post!

 

[TragikRemix]

ill 3rd or 4th the too much reading, not enough doing thing..

yup. dont track too hot is right. we'll get that volume up in the end one way or another

 

[ez_willis]

Don't get me started!

 

[snow lizard]

I kind of suspect something in the area of sonic decimation to be a factor.

On another recording forum, I've seen a member who defined the word "swoggle" as a combination of suck, swallow and gargle.

If we're going to print 20 tracks or so that will end up in a mix at levels of, say, -18 dBFS or whatever, when the mix happens, the levels sum. Things get louder. Okay, somebody already mentioned that. You end up having to yank the levels down to keep things from slamming the digital zone of perpetual endamnment. (<- is that actually a real word?)

I've always been curious to know if there's a direct relationship between adjusting volume levels in a DAW (or any extensive processing that will affect dynamic levels, for that matter) and swoggling.


Mr. Cave?


Anyone?


sl

 

[snow lizard]

Without actually knowing what I'm talking about, I'll try to explain in a different way.

Given how bit depth (eg. word length in PCM digital audio) works from most significant bit to least significant bit, in that the total range of 24 bit audio is something like 144 dB (sort of), and that each bit in the word acts on 6 dB of dynamic range, the more you would have to adjust the volume level of an individual track (or any digital audio source) downwards, you'd essentially lose much of the information that was recorded to describe the dynamic level originally printed. I suspect the process would introduce small artifacts that chump the realism of the track worse than if it were printed at the lower level to begin with.


Any basis for that?

Does it make it sound like this thing?

http://www.frostwave.com/sonicalienator/


Also, what the heck is "32 bit float" all about?



Confused,

sl

 

[The Audio Cave]

I'll add on to iqi616's great post by saying that the other reason to keep the tracking levels down that will help during mixing is for leaving headroom on individual tracks for insert plugins.

A track with a very high average level and a really high peak (-2?) might very possibly cause overs (inside the plugin process) when you insert a plug on that track. You're not leaving enough headroom for the plug to be most effective. Basically when I brought my levels down (for all the reasons we spoke of already) everything sounded better, including my plugins.

On the subject of "leaving the final level for mastering" ...

I agree with that 100%, but... typically tracking levels have nothing at all to do with mixing levels - that is the overall peak and average level of the master bus when your mix is done. Unless you track really really, really, really, low and you can't get the levels you need because the fader stops at the top of it's travel. Nobody does that.

At the same time you can take the "hotter" tracks and create a mix that peaks at -15. It all depends on where you set the individual faders.

Here's how I look at it:

Tracking up to -1 or tracking to -20... either way you can easily fill up the entire master bus (have it peaking to -1 with very high average levels which is also not a good idea but you already know that) to get it louder (before L2 or whatever) with those tracks. In either of those circumstances depending on where you set the kick to start (assuming you're mixing from the drums up) You can get the same loudness before overloading the master bus with either. The only difference will be in the mix where you tracked at -20 the faders will be higher up. That's all. Same loudness.

So the idea that by tracking reasonably lower you won't be able to achieve the same loudness in the final mix is just a myth. You just push the faders up more if you want to mix hotter into the master.

So tracking levels have little to do with final mix levels unless you track so low your faders stop before you can get the mix levels you need.

It's not the tracking level that determines the mix level, it's the amount of each those tracks that you decide to send into the master bus when you mix.

Typically when I mix (even after tracking to -10 or lower) there are hardly ever any tracks where the track fader is at or near unity gain (0). So almost all the individual tracks end up sending a lower level of themselves into the master bus than they were actually recorded at. Partly because I mix up to about -5 or lower.

Make sense?

So to kind of summarize why to keep levels lower...

1. Keeps you within the operating ranges of your analog gear.
2. Leaves some headroom for insert plugins.
3. Allows you to use the "sweeter" part of the fader's logarithmic travel, giving finer control during mixing (* see iq1616 post)
4. You don't have to worry about overs when tracking*
5. It's not going to affect final mix level.
6. You're not losing any quality or bits.
7. It's simply totally unnecessasry in 24-bit recording to record that hot and actually can be detrimental.

I might have missed a point or two there.