Jump to content

-18dbfs in the Digital domain = 0dB in Analogue? Eh??!


Recommended Posts

So, I get talking to this old guy in a bar on Xmas Eve & it turns out he ran a commercial 16-track studio 'back in the day' ('70s/'80s). I mentioned in passing that I do a bit of home recording. Anyway ... to cut a long, tipsy conversation short, the main nugget of information I took from our exchange was this:

 

"Be real careful with those input levels when recording digitally! That 0 dB level on digital input channel strips isn't the same as the 0 dB of VU metering in the analogue domain. Talking relatively, as a ballpark figure, when recording digitally you should aim for an average input signal level of around -18dBfs ... that would be around about the equivalent of tracking to 0dB in analogue".

 

This snippet of info has been percolating away with me over the entire festive season. Was it the beer talking or is this guy right in what he's saying? If so, I will need to completely rethink the way I meter input signals when recording!

 

Any further thoughts, insights on this much appreciated!

 

PS: For the record, for the past ... God knows how long, I have been recording with input levels of approx. -6dB average / -3dB peak. If what this guy's saying is right (ie: -18db in digital domain = equivalent of 0dB level-wise in the analogue) then I've been recording a whole 12dBs hotter than I should have been for the longest time now! No wonder I often fight with Channel Strips going into the red when I come to adding software plug-ins at mix time! Intrigued ....

Link to comment
Share on other sites

Well that certainly explains a few things - thanks for confirming that! OK, I've now got a lot of re-working (& questions!) to be going on with as a result of this simple 'EUREKA!' revelation. (I've been constantly fighting with trying to control my levels for the longest time now & now I know why!) There's no doubt about it, I've been recording way too 'hot' - silly me!

 

From just one tentative, new regime test recording session (trying!) to maintain lower input levels (lower average tracking levels by -15db as it turns out = YIKES!) off the top of my head I note the following changes:

 

- Waveforms are SO much smaller than what I'm used to

- Dry, non-processed recorded tracks are SO much quieter too!

- I have to turn my headphones volume up to 3 o'clock (previously it had been at 9 o'clock)

- Same goes for my monitor speakers volume. I've gone from about the 8 o'clock position on average to about 2 o'clock. (I've now got to be really careful I don't blow my speakers when switching to iTunes with my volume levels set this high - it's going take a bit of getting used to!)

- I haven't had a chance to do much recording yet but I did notice that a MIDI drum beat I quickly played in (just triggered a beat using my V-Drum kit then converted it using NI Abbey Rd 60s Drummer software) went straight into the RED! I'm really not sure why this is happening so i will need to investigate the whole MIDI levels side of things in more detail too!

- I added some processing to a test electric guitar part (Library>Electric Guitar>'Clean Echoes') and this stock preset put my low recorded dry track (which peaked at -16dB during tracking) into the red. That's not right either is it??!

 

WHOAH! Time to wipe the slate and start again methinks (New Year's Resolution!) In light of this single nugget of info, all along I'm now VERY aware I've been recording WAY too hot. I'm going to need to go away & completely re-think the way I do stuff ..... sigh!

Link to comment
Share on other sites

Although it is true, that doesn't mean you have to calibrate everything to VU scale.

 

Keep doing what you were doing, just make sure you don't clip your input or output. Unless you're using external analogue hardware processing or emulation plugins that are calibrated to VU levels (and not all are calibrated to the same scale, just to make it more confusing), then you really don't have to worry about it.

Link to comment
Share on other sites

Although it is true, that doesn't mean you have to calibrate everything to VU scale.

 

Keep doing what you were doing, just make sure you don't clip your input or output. Unless you're using external analogue hardware processing or emulation plugins that are calibrated to VU levels (and not all are calibrated to the same scale, just to make it more confusing), then you really don't have to worry about it.

 

Yes and no, record between -16 and -20 dBFS RMS will assure you to be in your converters sweet spot.

Link to comment
Share on other sites

Thanks guys! Been out in my makeshift studio shack today trying stuff out. Unfortunately encountered my first 'sonic accident' as a result. I'm not used to having to have my headphone and monitor levels so high it seems! I was test recording some drums (using a method kindly brought to my attention by the good Reverend J Sleaze as it turns out!) In a nutshell I record my drums as such: I record (& monitor) a stereo audio track PLUS, simultaneously, a MIDI track (using a Roland TD3 V-Drum kit as my controller). I then hide the Audio track for reference & convert the MIDI to samples from NI's 'Abbey Road 60s Drummer' software. I have my headphones set to, well, just below FULL ON as it turns out. I normally have my cans volume much lower but, recording my V-Drum track with an average peak of -18dB, they needed to be that high in order for me to monitor my audio track drums along to my guide track. Thing is, post recording, when I converted my MIDI track to drum samples, the NI samples play back at 0dB and "BANG!" the first snare crack just about blew my ears out = "OUCH!" As a resultr, today's recording session has come to an abrupt end.

 

Any further tips as to how to best manage situations like this & how to process & control levels when adding processing to dry recorded tracks. I really wasn't expecting that! My ears are still ringing as I type here ....

Link to comment
Share on other sites

Yes and no, record between -16 and -20 dBFS RMS will assure you to be in your converters sweet spot.

 

Do you have a link to evidence of that? How is a converters' "sweet spot" measured?

 

Do you meter in RMS when recording? How does that work out with highly dynamic sources?

 

Well, you won't find one single link about that question as there's no definitive standard for the dB VU/dB FS equivalence. BUT there are recommendations from EBU for example that decide that 0 dB VU should be -18 dB for Europe, -20 dB for US and Australia.

 

Converters sweet spot aren't measured, they just follow these specs. And yes, you meter in RMS. With highly dynamic sources, the best way to do it is to compress before it hit the converter. And it doesn't have to be exactly at -18 all the time, it's not even possible anyway. Between -20 and -16 is ok.

Link to comment
Share on other sites

Well, you won't find one single link about that question as there's no definitive standard for the dB VU/dB FS equivalence. BUT there are recommendations from EBU for example that decide that 0 dB VU should be -18 dB for Europe, -20 dB for US and Australia.

 

Converters sweet spot aren't measured, they just follow these specs. And yes, you meter in RMS. With highly dynamic sources, the best way to do it is to compress before it hit the converter. And it doesn't have to be exactly at -18 all the time, it's not even possible anyway. Between -20 and -16 is ok.

 

How does that work with orchestral or classical ensemble recordings?

 

I know that VU levels vary between countries, but I was asking about optimum dBFS levels on input. I tried to find a white paper, but can't.

 

Have you got no links or a reason for this sweet spot? I don't understand how something like this can be known without being able to be measured...

Link to comment
Share on other sites

That works exactly how it did back in the days of analog recording. They tried to be as close as possible of the 0 VU. Analog tape handled pretty well levels beyond 0 VU so it wasn't an issue to go over that. (if not too much)

 

In digital recording, we try to be as close as possible of -18 dB FS and we have a limit (0 dB FS) so we have 18 dB of dynamic range to work with. If it's not enough (most of the time it is, maybe not in case of an orchestral recording, dunno i've never been involved in one), you'll have to lower the level ...

 

But once again, you don't necessarily have to work at -18, but it's the best spot to record to in a digital environment. Nowadays we don't have the same signal-to-noise ratio issue they had before, so you can totally record at -24 if you want to but in terms of levels it will be more consistent at -18.

For example if you use hardware for mixing, if everything has been recorded at -18, your gears will receive exactly the level they need as they suppose to work at 0 VU which is -18 dB FS. Same thing goes for plugins. You can easily overload a plugin with something recorded too hot.

 

And at last, if you want to measure it, you can ! Send anything with a constant level in the input of your choice at -18 on a Logic track and then insert on that track a VU Meter plugin. If it says 0 dB, your audio interface is "calibrated" at -18 dBFS. If not, adjust the level until it says 0 on your VU Meter to find out.

Link to comment
Share on other sites

Ok, this is getting quite confusing, so I'll try to find out where you're coming from in a piecemeal fashion:

 

That works exactly how it did back in the days of analog recording. They tried to be as close as possible of the 0 VU. Analog tape handled pretty well levels beyond 0 VU so it wasn't an issue to go over that. (if not too much)

 

:lol: "they"? There are lots of us still around that remember VU meters! And many who still record analogue! :D

 

In digital recording, we try to be as close as possible of -18 dB FS and we have a limit (0 dB FS) so we have 18 dB of dynamic range to work with. If it's not enough (most of the time it is, maybe not in case of an orchestral recording, dunno i've never been involved in one), you'll have to lower the level ...

 

Do we now? :wink:

 

If you're metering in RMS, how do you know how much headroom (which is what I guess you mean by "dynamic range") you have to work with?

 

But once again, you don't necessarily have to work at -18, but it's the best spot to record to in a digital environment. Nowadays we don't have the same signal-to-noise ratio issue they had before, so you can totally record at -24 if you want to but in terms of levels it will be more consistent at -18.

 

Not sure what you mean by "consistent". Are you just talking about routing to-and-from analogue gear? When you talked about a converter's "sweet spot", I presumed you meant it sounded better there.

 

For example if you use hardware for mixing, if everything has been recorded at -18, your gears will receive exactly the level they need as they suppose to work at 0 VU which is -18 dB FS. Same thing goes for plugins. You can easily overload a plugin with something recorded too hot.

 

Yes, I mentioned these two examples earlier. However, this has nothing to do with any kind of sweet spot on an A/D converter. You could attenuate the signal in the digital realm before hitting your D/A converter and it would make no odds.

 

Same goes for plugins, in terms of digital attenuation. And different analogue emulations are calibrated differently. These calibrations are arbritrary because average voltage in an analogue audio system and the digital scale are not comparable.

 

And at last, if you want to measure it, you can ! Send anything with a constant level in the input of your choice at -18 on a Logic track and then insert on that track a VU Meter plugin. If it says 0 dB, your audio interface is "calibrated" at -18 dBFS. If not, adjust the level until it says 0 on your VU Meter to find out.

 

Ok, here's where it gets even more confusing :?

 

So, you're using a software coders' arbitrary calibration of an emulation of an RMS voltage meter to determine the input levels of your DAC?

 

There are very good reasons why analogue audio gear measures signal in voltage, and digital audio gear measures in digital scale. The two are not directly related or comparable.

 

Not to mention the particular behaviour of different VU meters, the way transient spike distortion can be too fast for the meters to even show etc.

 

We seem to have gone from "You gotta average out your incoming voltage to 1.23 volts RMS", to "you can set it wherever you like as long as you're not clipping - whilst being mindful of levels when routing the audio through analogue hardware, or some plugins that emulate analogue hardware (though, most of those plugins have input attenuation, which works just as well) - which is what I said in the first place.

 

There seems to be an increasing trend in people trying to reintroduce certain limitaions of analogue audio into their digital audio workflow. I can't for the life of me figure out why!

Link to comment
Share on other sites

Ok, you seem to be a technical person which i'm not. Plus, English is not my native language as you may have noticed :D

You also seem to be a lot concern by the whole level thing. As i said, there's not one answer to that question because, there's no official value for 0 VU in the digital domain. -18 seems to be the most common though.

 

Ok, this is getting quite confusing, so I'll try to find out where you're coming from in a piecemeal fashion

What i meant here was:

In analog, you try to record at 0 VU (+ 4dBu). Most analog gear clip approximately at +20-24 dBu, so you have between 16 to 20 dB of headroom.

In digital, you try to record at -18 dBFS RMS and your limit is 0 dBFS, so you have 18 dB of headroom. This should keep your transients safe :)

 

 

Do we now?

Yes ! Nobody's gonna die if you're under -18 !

 

 

If you're metering in RMS, how do you know how much headroom (which is what I guess you mean by "dynamic range") you have to work with?

You're not metering in RMS only. You're metering the Peak level at the same time. And yes i meant headroom sorry ...

 

 

Not sure what you mean by "consistent". Are you just talking about routing to-and-from analogue gear? When you talked about a converter's "sweet spot", I presumed you meant it sounded better there.

Yes i'm talking about routing to and from analog gear ! And yes, by sweet spot, i meant the ideal level to work at for your converters.

 

 

So, you're using a software coders' arbitrary calibration of an emulation of an RMS voltage meter to determine the input levels of your DAC?

I choose a plugin with a -18 dBFS calibration option. But you're right, i had to trust the guy who did the plugin ! But anyway it was just for fun. All you have to do is to record at or around -18 dBFS RMS and you know you're good to go.

 

 

There are very good reasons why analogue audio gear measures signal in voltage, and digital audio gear measures in digital scale. The two are not directly related or comparable.

Absolutely and that's why you should not try to measure it and stuff. In analog, you KNOW 0 VU = 1,23 volts = + 4 dBu. I wish it was that simple in digital.

 

 

Not to mention the particular behaviour of different VU meters, the way transient spike distortion can be too fast for the meters to even show etc.

Well, aren't VU Meters supposed to show RMS levels for the most part ?

Link to comment
Share on other sites

All very interestin' for sure. On a purely practical level coming from someone who's been recording WAY too hot since making the transition from analogue recording to digital (I just thought 0db was 0db regardless = "D'oh!") I would like to seriously give this a go. I've been fighting with plug-ins going into the red for what seems like the longest time now so the idea of recording considerably lower rather appeals to me (regardless of all the science behind it all). In light of this (and my technical limitations!) I'm wondering, going back to my original posts, if you guys can offer me some user help to enable me to experiment this end? From my preliminary tests to date, here's some points I'm having issues with:

 

1. I was tracking a 'dry' dynamic electric guitar part really trying to get it to register an RMS value of -18dBfs (= DIFFICULT! I thought I'd set my peak level correctly but, when I hit record & actually began to play in what I hoped would be the 'definitive' part, as the song progressed & I got more & more into it then noticed the channel strip in Logic was indicating -9dBfs. It took me quite a few goes of replaying to finally contain the part with a max. peak of -16dBfs). When setting levels (regardless of what you're aiming for) is there any way of getting Logic to read a constant, real time input level? I work in isolation so need to play and engineer simultaneously. The way my channel strip meters are configured in Logic (Pro 9) it gives me what I presume to be a peak level readout then just stays there. I would find it easier to set my levels when playing if the input level readout (above the lighting meters) fluctuated rather than just reaching the peak and staying there.

 

2. As much as I like this (new to me) idea of recording much lower than I have been, I'm seriously concerned as to just how high I seemingly have to have both my headphones & studio monitors levels set in order to hear my playbacks at a 'normal' cans/room listening level! Is this right?

 

3. In the past (ie: for the past 5 years right up to last week!) I would somewhat rely on the visual representation of the sounds being recorded. Now, my waveforms are so minuscule, I have to keep double checking I'm actually recording something. My waveforms are TINY (think visual straight lines with a few dots to indicate spikes). This doesn't seem right somehow. I realise I can magnify them but I'd only be seeing a few lanes of tracks in the project window to be able to read the screen how I used to.

 

4. So ... I have my recorded guitar part and my headphone volume is on full with the track playing back with the fader on the channel strip set at 0dB. I add some MIDI drum samples and they're WAY louder than my recorded guitar part! Hmm ... where can I get a VU meter plug-in to calibrate my interface with Logic?

 

5. Any advice on adding plug-ins to sounds recorded with an RMS of -18dBfs would be much appreciated too. (I haven't got to that stage yet!)

 

PHEW ....

Link to comment
Share on other sites

Ok, you seem to be a technical person which i'm not. Plus, English is not my native language as you may have noticed :D

You also seem to be a lot concern by the whole level thing. As i said, there's not one answer to that question because, there's no official value for 0 VU in the digital domain. -18 seems to be the most common though.

 

I'm not really a technical person, my eyes glaze over when I look at formulae :)

 

My only concern with levels is that the people at home reading this don't go away with the notion that they have to add unnecessary layers of complexity to the digital recording process.

 

Your input metering workflow works for you, and that's fine, there's nothing wrong with what you're doing. But, there's nothing wrong with doing things differently either ;)

 

At the end of the day, for people recording through an audio interface into a DAW, all you have to worry about is that your input does not clip your converters. Any "sweet spot" depends entirely on the dynamic range of your source.

 

In analog, you try to record at 0 VU (+ 4dBu). Most analog gear clip approximately at +20-24 dBu, so you have between 16 to 20 dB of headroom.

In digital, you try to record at -18 dBFS RMS and your limit is 0 dBFS, so you have 18 dB of headroom. This should keep your transients safe :)

 

That's fine for a general rule of thumb, but it's not set in stone. Headroom, noise floor and "sweet spots" are a concern with analogue gear. With digital recording, these things are far less of a concern.

 

As I said above, the dynamic range of your source is what will determine an RMS level measurement. As long as your highest peak is below 0dBFS, the RMS value is not really relevant.

 

With non-destructive processing in a floating-point digital audio engine (ie. plugins in Logic), it makes no difference what your recording levels were, because you can attenuate the signal upstream of any plugins that are calibrated to emulate analogue levels.

 

Nobody's gonna die if you're under -18 !

 

Nope! And no-ones gonna die if you record hotter than -18dBFS! :D

 

Yes i'm talking about routing to and from analog gear ! And yes, by sweet spot, i meant the ideal level to work at for your converters.

 

A/D converters' sweet spot is below 0dBFS! If you're recording at -40dBFS you might have problems with preamp noise, but your converters will still be happy.

 

Well, aren't VU Meters supposed to show RMS levels for the most part ?

 

Yes, but they are not a perfect representation of average voltage.

 

Here's a good primer on metering: http://www.soundonsound.com/sos/jun00/articles/metring.htm

 

On VU meters from the above article:

 

"Because the VU meter measures 'average' levels, a sustained sound reads much higher than a brief percussive one, even when both sounds have the same maximum voltage level: the reading is dependent on both the amplitude and the duration of peaks in the signal. In addition, the standard VU response and fallback times (around 300 milliseconds each) exaggerate this effect, so transients and percussive sounds barely register at all and can cause unexpected overloads."

 

"Occasionally, you might notice the VU meters on different equipment reacting differently to an identical audio signal, particularly when professional and budget units are used side by side. This is because, though VU meters are supposed to be sensitive to both the positive and negative half-cycles of audio signals, many budget units are sensitive only to one half of the waveform. This can lead to considerable differences between VU readings, as many audio signals are asymmetrical."

 

With analogue recording, compressing on input can be a necessary step with highly dynamic material, because you're battling headroom and noise-floor concerns. But in digital recording there is no need to compress anything before it hits your converters.

Link to comment
Share on other sites

Spock, you're jumping down a rabbit hole that will just unnecessarily complicate recording for you.

 

1. I was tracking a 'dry' dynamic electric guitar part really trying to get it to register an RMS value of -18dBfs (= DIFFICULT! I thought I'd set my peak level correctly but, when I hit record & actually began to play in what I hoped would be the 'definitive' part, as the song progressed & I got more & more into it then noticed the channel strip in Logic was indicating -9dBfs. It took me quite a few goes of replaying to finally contain the part with a max. peak of -16dBfs). When setting levels (regardless of what you're aiming for) is there any way of getting Logic to read a constant, real time input level? I work in isolation so need to play and engineer simultaneously. The way my channel strip meters are configured in Logic (Pro 9) it gives me what I presume to be a peak level readout then just stays there. I would find it easier to set my levels when playing if the input level readout (above the lighting meters) fluctuated rather than just reaching the peak and staying there.

 

Don't worry about it!

 

If the input meter in Logic is indicating -9dBFS peak, you're all good!

 

2. As much as I like this (new to me) idea of recording much lower than I have been, I'm seriously concerned as to just how high I seemingly have to have both my headphones & studio monitors levels set in order to hear my playbacks at a 'normal' cans/room listening level! Is this right?

 

Add a gain plugin to your output strip and turn it up. It's that simple.

 

3. In the past (ie: for the past 5 years right up to last week!) I would somewhat rely on the visual representation of the sounds being recorded. Now, my waveforms are so minuscule, I have to keep double checking I'm actually recording something. My waveforms are TINY (think visual straight lines with a few dots to indicate spikes). This doesn't seem right somehow. I realise I can magnify them but I'd only be seeing a few lanes of tracks in the project window to be able to read the screen how I used to.

 

There is a seperate "waveform zoom" control in Logic. It used to be just to the right of the horizontal scroll bar in the arrange window. It may have moved in Logic X, I don't own it to check. This lets you enlarge audio region waveforms without having to make the track lanes any bigger.

 

4. So ... I have my recorded guitar part and my headphone volume is on full with the track playing back with the fader on the channel strip set at 0dB. I add some MIDI drum samples and they're WAY louder than my recorded guitar part! Hmm ... where can I get a VU meter plug-in to calibrate my interface with Logic?

 

You don't need no steenkin' calibration! Use the peak meters on your audio interface!

 

5. Any advice on adding plug-ins to sounds recorded with an RMS of -18dBfs would be much appreciated too. (I haven't got to that stage yet!)

 

Not sure what you mean by this...

Link to comment
Share on other sites

 

5. Any advice on adding plug-ins to sounds recorded with an RMS of -18dBfs would be much appreciated too. (I haven't got to that stage yet!)

 

Not sure what you mean by this...

 

Thanks for the breakdown Rev, again much appreciated! Going back to my original post, 'this' (ie: rethinking the way I record) came about from talking with a recording engineer in a bar & me complaining about my constant fight with recording in the digital domain. Thing is, I'm from yesteryear (LOL!) The digital recording thing is only something I've entertained in the last half decade. Prior to that my amateur recording career had been analogue-based. I had some fundamental basic training 'back in the day' to get me started with recording in analogue (I started off on domestic 4-track portastudios then graduated through the years to an 8-track reel-to-reel). I was contemplating trading in my 8-track for a 16-track when I was 'persuaded' by a knowledgeable musician friend of mine to 'go digital'. Thing is, to date, I've never really felt entirely comfortable with the whole recording in digital thing (although I really appreciate those post-production editing capabilities!) After half a decade of what has felt like 'fighting' with this medium I finally felt I'd found the answer over Xmas from talking with this studio engineer ... now I'm just doubly confused & kind of wish I'd never started this post. Ho! Ho!

 

On a positive point, I am beginning to realise the error of my ways with digital recording. Bottom-line: I've been applying my analogue principles & techniques to digital recording and treating 0dB on my Logic channel strips the same as I did 0dB on my old analogue console. Yes, a big no, NO! I realise now so that's something! I've been recording WAY too hot!!!

 

Now I have (finally!) realised this I now need to know the 'correct' way of adding processing to my (dull? lifeless? 'DRY') recordings to prevent my signals again going into the red .... basically I'm just sick of recording stuff into the red and constantly fighting with it! I know this sounds really basic Dude but help me out here - right now I feel like THE least technical person on this forum - a total frickin' plankton in fact!

 

Do I take a dry track (recorded at a lower level than what i have been recording to date ie: nowhere near 0dB!) then for each plug-in I might add to said track (eg: if it were an electric guitar I might add: an amp from Logic's 'Amp Designer'; emulated 'Compressor', 'Delay' & 'Chorus' pedals from Logic's Pedalboard; a 'Sample Delay') adjust the output of each individual plug-in? And, if so, by how much? Man, I'm feeling completely lost this end (no doubt exacerbated by the fact that yesterday I somehow managed to wipe my external HD containing practically every musical idea I've had over the last few years = "D'OH!")

 

New year .... onwards & upwards!

Link to comment
Share on other sites

On a positive point, I am beginning to realise the error of my ways with digital recording. Bottom-line: I've been applying my analogue principles & techniques to digital recording and treating 0dB on my Logic channel strips the same as I did 0dB on my old analogue console. Yes, a big no, NO! I realise now so that's something! I've been recording WAY too hot!!!

 

Now I have (finally!) realised this I now need to know the 'correct' way of adding processing to my (dull? lifeless? 'DRY') recordings to prevent my signals again going into the red .... basically I'm just sick of recording stuff into the red and constantly fighting with it! I know this sounds really basic Dude but help me out here - right now I feel like THE least technical person on this forum - a total frickin' plankton in fact!

 

Do I take a dry track (recorded at a lower level than what i have been recording to date ie: nowhere near 0dB!) then for each plug-in I might add to said track (eg: if it were an electric guitar I might add: an amp from Logic's 'Amp Designer'; emulated 'Compressor', 'Delay' & 'Chorus' pedals from Logic's Pedalboard; a 'Sample Delay') adjust the output of each individual plug-in? And, if so, by how much? Man, I'm feeling completely lost this end (no doubt exacerbated by the fact that yesterday I somehow managed to wipe my external HD containing practically every musical idea I've had over the last few years = "D'OH!")

 

New year .... onwards & upwards!

 

Hehe!

 

Don't sweat it, I was the same when I started recording to digital.

 

The thing to remember is that you have TONS more wiggle room with digital. You don't have to worry about noise floor and bounce degredation. With analogue, you stay around the border of red, and for rock n' roll it doesn't matter if you go over a bit. With digital, you can record at much lower levels, and you NEVER want to go over 0dBFS.

 

My advice would be just to play louder than you would for the part you're recording, make sure that's a few dB below clipping, then forget about levels and concentrate on playing the part as well as you can.

 

For gain staging: yes. adjust input and output on plugins. If a plugin doesn't have gain controls, insert a gain plugin and use that. Even if you're using analogue emulation plugins, because they can be calibrated to anywhere between -18dBFS and -9dBFS, you should just use your ears to determine the correct level going into them. It's all fake anyway, and in "real world" analogue situations you'd also use your ears to decide on the appropriate level, how much breakup you want etc...

 

There is no true "optimum level" for digital processing. It's a good habit not to clip your channels, but if it sounds good don't worry about it. Just make sure that your stereo output isn't clipping. Use faders and gain plugs to keep your mix under control.

 

Because of Logic's floating-point audio engine, you can be clipping every track in your mix, but then stick a gain plug on your stereo-out to attenuate it below 0dBFS and your audio will not suffer at all.

 

PS. Always keep a backup of projects! ;)

Link to comment
Share on other sites

Recording between -9 and -6 in Logic is perfect!

 

Don't let it hit red in Logic, that's all.

 

OK. That's pretty much what I'd been doing ie: recording between -9 and -6. I'm now beginning to think my issues lie with processing & the use of plug-ins. I've not been turning stuff down within the plug-in just pulling the channel strip faders down (a bit) instead. Right, back to the drawing board .....

 

Thank you! :oops:

Link to comment
Share on other sites

Recording between -9 and -6 in Logic is perfect!

 

Don't let it hit red in Logic, that's all.

 

OK. That's pretty much what I'd been doing ie: recording between -9 and -6. I'm now beginning to think my issues lie with processing & the use of plug-ins. I've not been turning stuff down within the plug-in just pulling the channel strip faders down (a bit) instead. Right, back to the drawing board .....

 

Thank you! :oops:

 

With every Logic plugin except the bitcrusher that is absolutely fine. Some emulation plugins break up at different levels, but just use your ears to judge what level sounds good.

Link to comment
Share on other sites

Yeah but still, there's an optimum level as there is in the analog domain and it should be meter in RMS. If you're around -9 dBFS and -6 dBFS in Logic, you're safe ok but you don't know what your RMS level is.

And we know 0 dB VU is somewhere around -18 dBFS, so why shouldn't we use that information and meter in RMS when recording ?

Link to comment
Share on other sites

OK, two recording tests this end:

 

TEST 1: A lead vocal trying for an RMS record input level of -18dBfs (NB: I found this difficult for some reason & ended up with a max. peak registering at -14dBfs)

TEST 2: Similar lead vocal take but this time recorded at my usual vox level setting namely an RMS record input level of -6dBfs (with one isolated peak at -4.7dBfs)

 

Both takes were recorded dry then the following processing preset was applied post-recording:

 

Logic presets library/06 Voice/02 Female Voice/'Female Ambient Lead Vocal' (featuring the following preset chain: Channel EQ; Comp; Limiter; Ensemble; Space Designer

 

(I've used this chain before on a few recordings and it seems to work okay for my voice - yeah, I know, 'Female' Ambient Lead! Move over Jimmy Somerville!)

 

With the above chain applied to the channel strip of each track I noted the following read-outs on playback:

 

TEST 1 [the -18dBfs RMS recording] had a processed peak channel strip playback level of: -3.3dB

TEST 2 [the -6dBfs RMS recording] showed a processed peak channel strip playback level of: -2.6dB

 

I'm not very good at maths but that seems a lot closer than I would have expected seeing that there's an unprocessed recording input level difference of 12dBs between the two test tracks.

 

Having now read Lagerfeldt's excellent paper (link given above - THANKS!) it seems that both processed tracks are peaking too much here right? (Lagerfeldt suggests to aim for a peak of -6dB for individual effected channels).

 

So ... how do I rectify this? Do I have to go into each & every plug-in in the chain and lower certain settings? If so, what do I change?

 

As for the SOUND (the most important bit!) of each test track .... I think I actually prefer the vocal recorded with an RMS input recording level of -18dBfs. Doing a side-by-side direct playback comparison it just seems to sound 'purer' to my ears.

 

Hope this all makes some kind of sense! I'm not the most technical person I have to admit (but you'll have already guessed that!)

 

PS: Thanks for all your help here folks - I'll crack this recording lark if it kills me! That said, I'm still very much waiting for the penny to drop this end ....

Edited by MrSpock
Link to comment
Share on other sites

As for the SOUND (the most important bit!) of each test track. I think I actually prefer the vocal recorded ay -18dBfs to the same vocal recorded at -6dB. It seems to sound 'purer' to my ears.

Maybe because you clipped your transients at -6dB. I also noticed that kind of difference when I start recording at -18. Some sort of distorsion disappeared in my recordings.

 

 

Whats the best way to do this? Multimeter plug-in on channel strip with Software Monitoring enabled perhaps?

I use FreeG from Sonalksis. It seems to be a little more accurate than the Logic plugin. It's free.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...