Tracking levels ...

[MarkRB]

Originally Posted by TheArchitect

Not if you have turned the input trim down and are recording a very low level because that part won't be loud in the mix for he sole purpose of all the faders lining up. Thats exactly what they are talking about.

Well the more gain used the more noise, so why add more noise if you don't have to?

[Dean Roddey]

So, having thought about it a bit while getting some chow, let me sum it up in a way that may help people who were seeing the thing the way I was...

If you take an evenly spaced grid of any number of divisions and lay it across the whole range from 0dB down to the noise floor, then all of the divisions represent an equal change in amplitude (I keep wanting type Amplitube every time I type that.) So, anywhere in that range the quanitization is the same, since all the divisions are equal. And, since you can interpolate the curve between any two points, you effectively have infinite resolution.

Therefore, all that really matters is the ratio between the sample amplitudes, not their absolute values. I.e. if sample A is at 10 and sample B is at 20, that's the same as if sample A was 20 and sample B was 40. Both will reproduce the same curve, one will just be louder than the other. They will both reproduce the same curve, because, in both cases, sample B is twice the magnitude as sample A, and so the same curve shape will be interpolated between them. Only an increase in sample RATE would change that.

So, effectively, all the bits are being used all the time, because the quantization error is the same whether loud or soft, and the ability to interpolate extends the effective resolution such that it's only limited by the sample rate.

[MHB850]

Originally Posted by Dean Roddey

It would take an pretty crazy amount of effort to do that, compared to just getting good levels (I go for max peaks around -9 to -6dB) and adjusting as required. EQ and compression will eat up some of that on those tracks that use any generally, and you get a good, healthy signal through the plugs so that they have plenty of bits to play with.

Not to mention that it requires knowing absolutely what hte song is going to sound like before you start, and never changing your mind later, even if a better idea comes to you during the mix. Many people are creating the song as the go and could hardly be expected to have an absolute understanding of every aspect of every instrument before they track, such that they would never change their mind about desired levels.

we didn't know what the song was going to sound like, we didn't have an absolute understanding of every aspect of every instrument before we tracked. we or the producer changed their minds alot during the recording process and the mix. it was a technique that took time to develop and it had great advantages as i've mentioned in the other post. But that technique was for recording to analog tape. I can't see how this technique would work in PT or any other digital format since you must record to get the best signal as you mentioned in your post. michael brauer

[TheArchitect]

Originally Posted by MarkRB

Well the more gain used the more noise, so why add more noise if you don't have to?

OMG.....


The S/N ration is the same, at worst, when recording at a hotter level. Certainly a bit a headroom to protect from clipping is desirable but the lower the level you record, the closer you get to the noise floor of your pre's and converter's etc. i.e the S/N ratio is getting worse. This is 101 stuff, wtf?

[MarkRB]

Originally Posted by TheArchitect

OMG.....


The S/N ration is the same, at worst, when recording at a hotter level. Certainly a bit a headroom to protect from clipping is desirable but the lower the level you record, the closer you get to the noise floor of your pre's and converter's etc. i.e the S/N ratio is getting worse. This is 101 stuff, wtf?

Dunno mate, seems pretty basic to me.

If you are adding 10dB gain at your pre just to reach zero and losing 10dB gain at your fader you are increasing the noise within your signal (actually dependant on the ratio) by 10dB more than necessary. The actual amount of noise will depend upon which has a better SNR, pre or channel.
If you are really starting to have problems with your recording mediums noise floor then you should decide if you are happier with additional pre noise or channel noise over Tape/Converter noise.

[steveschizoid]

Originally Posted by TheArchitect

OMG.....


The S/N ration is the same, at worst, when recording at a hotter level. This is 101 stuff, wtf?

Not so.

The s/n ratio's relationship to the the mic pre as the level is raised is entirely dependent upon the mic pre. Many pre's are designed to work best at 0 dbU - (-18 or so DBFS dependent upon the calibration of your converter), and as you exceed that you get more noise and distortion. That distortion may result in a sound you like, but the s/n ratio has decreased.

[TheArchitect]

Originally Posted by MarkRB

Dunno mate, seems pretty basic to me.

If you are adding 10dB gain at your pre just to reach zero and losing 10dB gain at your fader you are increasing the noise within your signal (actually dependant on the ratio) by 10dB more than necessary. The actual amount of noise will depend upon which has a better SNR, pre or channel.
If you are really starting to have problems with your recording mediums noise floor then you should decide if you are happier with additional pre noise or channel noise over Tape/Converter noise.

Your forgetting that when I pull that fader down, the noise drops as well as the audio does it not? The ratio stays the same. By recording at a proper level, you get a better ratio.

[TheArchitect]

Originally Posted by steveschizoid

Not so.

The s/n ratio's relationship to the the mic pre as the level is raised is entirely dependent upon the mic pre. Many pre's are designed to work best at 0 dbU - (-18 or so DBFS dependent upon the calibration of your converter), and as you exceed that you get more noise and distortion. That distortion may result in a sound you like, but the s/n ratio has decreased.

That an assumption and a specialized circumstance. They are talking about recording at a reduced level for the sole purpose of where they can leave the fader when they are mixing. Entirely different circumstances

Your also assuming that I am talking about pushing things to 0. I'm not. A peak of -15 to -10 is optimal on a digital platform imo.

[steveschizoid]

Originally Posted by MHB850

I can't see how this technique would work in PT or any other digital format since you must record to get the best signal as you mentioned in your post. michael brauer

This is what I said at length in a much earlier post. Would I be a total jerk if I said I told you so to all you real fader pushers out there? If so, please forgive me!

With plenty of atty's you could get this to work, but it would take extra time and effort while musicians are sitting around ready to play, and, with a really soft source, you might actually be flirting with the noise floor of the converter. I just don't see the point.

[nickynicknick]

Originally Posted by Jules

Hmmm.. for mic pre gain and getting the sweet spot sure but all you need is a line trim device to set the level to to converter (or tape recorder nicely - so a ATTY or a fader (if working on a console) in fact when pushing mic pre's its pretty hard to do it and not slam idiotic levels to a converter any way so you really do need some sort of level control unless you want to be round the back of your converter with a tiny screwdriver adjusting its input calibration all the time)

why the hmmmm Jules? you obviously don't understand. ...here's an example...sometimes I feel that hitting my converter (which has pretty good analog front end) with a bass peaking at -8 sounds better than peaking at -20...obviously at -8 it's probably not gonna sit right in a mix if the (digital) faders remained in a nice line...really buddy this is as old as the hills....people slam tape all the time to get a desired effect...does this mean they don't appreciate good gain structure?...I have tracked your way lots and appreciate it but it is definitely not necessary...it all depends if you are going for a sound or capturing a performance realistically...for the latter your way works well. I just find myself going for a sound more often than not.....and.... the reason I mentioned not digitally clipping is because I feel strongly about this and consider it relevant to those that may not think it's a big deal. It's relavant to "Tracking levels" and it is important.

Nick

[gregohb]

Originally Posted by recall

Are we not recording music where a true representation of a performane is the goal?

The drums will always be louder than live viola for instance. I don't see the point of recording the viola as hot as possible so that it sounds as loud or louder than the drums.

As I said before there is plenty of room in 24 bit to get adequate signal to noise while still recording an accurate balance of what it actually going on in the room.

The whole art of recording is to make things sound better than they really do. Just like photography is making people look better than they actually do.

If everything was naturally volumed, then how would any singer sing over the drums? For violas, there might be some subtle notes, some finger squeaks, etc, that you want to capture too. What if a singer, guitarist or viola player moved their body or direction while recording and some good bits were recorded off axis? As someone else said, it doesn't hurt and helps any possible noise floor when you later turn that fader down.

[steveschizoid]

Originally Posted by TheArchitect

Your also assuming that I am talking about pushing things to 0.

I don't mean to be bellicose, but I didn't assume anything; I quoted what you said and rebutted it. You basically said the hotter the better; you didn't say "only up to certain point" in the original post.

[nickynicknick]

Originally Posted by steveschizoid

I don't mean to be bellicose, but I didn't assume anything; I quoted what you said and rebutted it. You basically said the hotter the better; you didn't say "only up to certain point" in the original post.

I don't usually agree with the Architect...but he is bang on here

[steveschizoid]

Well of course he is, and, after he clarified what he meant, I have no issue with it. But when you run some pre's past 0dbu, they get noisier. He originally implied that louder was better with no s/n ratio consequence. That whole bit about -15 to -10 DbFS is an important distinction that he didn't make. Read the whole thread!

[TheArchitect]

Originally Posted by steveschizoid

I don't mean to be bellicose, but I didn't assume anything; I quoted what you said and rebutted it. You basically said the hotter the better; you didn't say "only up to certain point" in the original post.

:)

If you think that's bellicose you don't venture to far into the internet. lol.

Quite right, I should have specified it. In my world that's just a given. I didn't even think to specify it.

[steveschizoid]

Originally Posted by TheArchitect

:)
If you think that's bellicose you don't venture to far into the internet. lol..

My wife and I once had a violent arguement; it ended up with her throwing a pillow at me!

[peeder]

Gain staging IMO is 50% of audio engineering. It's far too subtle an issue to make broad sweeping statements about what is right and wrong. Each device often has numerous gain stages within it, each an opportunity for glory and tragedy, and the whole chain has to be considered as a system.

After you've learned everything else, you'll still be learning about gain staging.

[steveschizoid]

Originally Posted by peeder

Gain staging IMO is 50% of audio engineering. It's far too subtle an issue to make broad sweeping statements about what is right and wrong. Each device often has numerous gain stages within it, each an opportunity for glory and tragedy, and the whole chain has to be considered as a system.

After you've learned everything else, you'll still be learning about gain staging.

Peeder, you rock!

Experimenting with differing preamp coloration is a way more interesting and effective use of time than attempting to track so that all my virtual faders end up in a line.

[Dean Roddey]

After you've learned everything else, you'll still be learning about gain staging.

You mean even after I've snatched the pebble from that old blind guy's hand?

[MarkRB]

Originally Posted by TheArchitect

Your forgetting that when I pull that fader down, the noise drops as well as the audio does it not? The ratio stays the same. By recording at a proper level, you get a better ratio.

OK, Imagine this or better yet try it if you have a desk.

Set your pre on a channel at a sensible level with your fader at zero and measure the noise.
Now turn your fader down to -10 and add an additional 10dB gain to the pre.
Which is noisier?

[synthoid]

Originally Posted by MarkRB

OK, Imagine this or better yet try it if you have a desk.

Set your pre on a channel at a sensible level with your fader at zero and measure the noise.
Now turn your fader down to -10 and add an additional 10dB gain to the pre.
Which is noisier?

You're comparing analog gain-staging to digital tracking levels; or at least I understood the original topic of this thread to be digital tracking levels.

When tracking digitally, you can set your pre to its best S/N ratio (or whatever other setting you are happy with) and track away, steering clear of digital overs of course. There's no need to artificially lower the gain on a pre (i.e., to settle for a suboptimal S/N ratio in the pre) simply because you are tracking a part that will ultimately have a low volume in the final mix.

Digital faders and digital summing don't work like an analog console. If you pull all the faders down to negative infinity in a DAW, the output will be silence (in the digital realm anyway). An analog console makes noise even if you pull every fader all the way down.

-synthoid

[TheArchitect]

Originally Posted by MarkRB

OK, Imagine this or better yet try it if you have a desk.

Set your pre on a channel at a sensible level with your fader at zero and measure the noise.
Now turn your fader down to -10 and add an additional 10dB gain to the pre.
Which is noisier?

These guys are talking about essentially premixing with input trims while while tracking so they can leave the faders in a nice straight row when mixing. This results in tracks that are recorded at far less than optimum for the "quiet parts" What difference does it make where you turn down the gain when track is recorded at a level far below optimum? Anything you might gain (minimal) is offset when you have to crank that channel because you changed your mind at mix time.

You might get away with this when you are tracking essentially complete performances live in the studio but when you are working with piecing together material a track at a time there is just no way you can be that sure of things.

[Paul Frindle]

Originally Posted by synthoid

You're comparing analog gain-staging to digital tracking levels; or at least I understood the original topic of this thread to be digital tracking levels.

When tracking digitally, you can set your pre to its best S/N ratio (or whatever other setting you are happy with) and track away, steering clear of digital overs of course. There's no need to artificially lower the gain on a pre (i.e., to settle for a suboptimal S/N ratio in the pre) simply because you are tracking a part that will ultimately have a low volume in the final mix.

Digital faders and digital summing don't work like an analog console. If you pull all the faders down to negative infinity in a DAW, the output will be silence (in the digital realm anyway). An analog console makes noise even if you pull every fader all the way down.

-synthoid

Yes - bravo for saying this.

Avoiding overs on your converter is obviously an extremely good idea :-) But deliberately under driving pre-amps and converters (below -10dB or so) is pointless and saves nothing.

Another important thing to note for very quiet instrument sources is that many (especially cheaper) Mic pres do not produce their best signal to noise ratios at lower gains - so pushing it up to get higher output and modulation level (without clipping) could make the recording significantly cleaner.

The business of creating headroom in your mix by dropping levels in the digital domain after the converter and before your processing, is an entirely different matter - and generally will not cost you significant extra noise since the SNR of the digital domain is always far better than your converters.

Now I know all this is tedious and a very unwelcome overhead when one is just trying to respond to the muso's, manage panicking producers and still keep a level head to get on with the job. But unfortunately, the naivety with which most W/S environments have been conceived makes it well worth while :-)

So this is what I would do:

1. Record through the Pre and ADC to get levels approaching -6dBFS or so - and record that to disc.

2. Drop the levels by a further -6dB or so at the head of your channels (on playback) when mixing to provide around 12dB of headroom (i.e. the amount something like an EQ or Dynamics can make the signal bigger without clipping).

3. Mix the whole thing without trying to max out the buss - aiming for -6dBFS or so is fine.

4. Boost your final mix output level (and apply limiting etc.) at the very last point in the signal chain on your master bus.

5. If you are maximising levels for the final mix - make really sure that nothing is done after your limiter - as this can push peak levels higher (even if you are 'removing stuff') and therefore reduce the loudness of your product.

[peeder]

In many DAWs, you can costlessly use a master fader to allow headroom at the bus while still keeping your faders around unity.

The master fader actually just pulls the other faders down for you invisibly.

[AndyUF2]

Originally Posted by Paul Frindle

5. If you are maximising levels for the final mix - make really sure that nothing is done after your limiter - as this can push peak levels higher (even if you are 'removing stuff') and therefore reduce the loudness of your product.

Interesting. But where would I then place a dithering plug-in? Wouldn't having it before the limiter increase noise? (yes, yes, I know, dither IS noise, technically, but still)

[Dean Roddey]

Most folks use a limiter/maximizer, that's doing both limiting, maximizing, and dither in the same plugin. However, sample rate conversion is often done by the DAW itself or the outboard converters, so that still get's done after maximization and dither (since that's done a buss in the DAW and the SRC gets done as part of the export.) Not sure if that's a bad thing or not.