whats up with my recording input level?

when I record the audio of my sound modules into Logic, the sound is extremely low
I'm not talking about bouncing, just how it's compared to everything else, it's significantly low after recording
so far I've found a good balance for recording my MPC, mixing tracks with tracks created in Logic
but the mods are annoying me

for instance I was using my Korg TR-Rack
I did a bass loop, messed with the midi parameters... sounded great with everything playing, a nice level
opened up an audio track and recorded it
after recording it's 10x more quiet than it sounded when I layed down the MIDI and previewed it

basically, I have the mod turned up almost max volume + the makeup gain on the input level of my interface
any louder and the speakers will blow... it shouldn't be this low when I look at the wavelength of the audio...

Which interface and how is everything routed? Which version of Logic?

What are the levels you're getting in Logic with pre-fader metering on (if your version allows it)? What about levels in your interface's DSP Mixer? You say that the waveforms look normal which means you likely recorded at a healthy level. If the waveforms looks correct, the problem isn't the modules it's likely somewhere in the monitoring...inside Logic, your interface's mixer, or afterward. The level going in does not translate to the loudness you hear from your monitors, in case that's what you mean.

Which interface and how is everything routed? Which version of Logic?

Logic 9 with a Motu 828mk3 (8 inputs) - MPC and 3 sound modules going in

What are the levels you're getting in Logic with pre-fader metering on?

I went to SEND to select Pre Fader, but it's greyed out... won't let me unclick No Send
any idea why that is? not good I'd suppose...

If the waveforms looks correct, the problem isn't the modules it's likely somewhere in the monitoring...inside Logic, your interface's mixer, or afterward. The level going in does not translate to the loudness you hear from your monitors, in case that's what you mean.

actually the waveforms I don't think are too good, which is apparent in the way it sounds on playback

here is a comparison of a kick drum on the MPC and one on the Trinity Rack
as you can see, the kick and the snare on the TR are like half the volume compared to the MPC
the snare looks like volume reduced hi-hats or something..

MPC



Trinity Rack


and this is how high it's going on playback, where the yellow bars are
the MPC playback goes almost to the top




I hate turning up something to where you can hear the volume's high without any sound playing...

Looks like your modules may be operating at -10 dBV or other consumer reference and the interface at +4 dBu, which may at least explain some of the signal discrepancy. I'm surprised that the output gain of the modules can't compensate for the difference, even if it's not good practice (noise). Maybe the MPC just has more gain? I'm not sure if all MPCs operate at the same nominal level, I did a search and couldn't find out if they differ, and you didn't mention which you were using. Regardless, you may want to try toggling your interface to operate at -10 dBV if that's what your modules are working at.

Ronald wrote:I went to SEND to select Pre Fader, but it's greyed out... won't let me unclick No Send
any idea why that is? not good I'd suppose...

You're in the wrong place. What you have in your screen capture is the pop-up menu to select whether you want your "sends" to be pre-fader or post panner/fader. A few ways to enable pre-fader metering is by going to the Options > Audio menu in Logic or to customize your transport and enable it from there, etc. At this point it's not really relevant though since the waveforms recorded from your modules are indeed low, in comparison. However, you do have that MPC a bit loud by looking at the meters.



Just out of curiosity, is there a reason you have Input Monitoring on during playback of already recorded tracks/channels? I doubt if that's the culprit though, since it's apparent your waveforms were recorded with weak levels.

I hate turning up something to where you can hear the volume's high without any sound playing...

Another thing to makes sure of is that the actual digital volume inside of your patches aren't too low. So for example, if you make a sound in one of your modules, don't leave the volume in the parameters at 50% and then try to turn the output of the rack up. That may not be the problem, but you want to double check that.

after looking for more information, I came across something
...and I think I have solved the problem...

this guy was describing pretty much the same thing I was
and he was told to go to 'Preferences > Display > Mixer tab'
and where Level Meters is, choose the Scale selection
and change from Exponential to Sectional db-Linear



after I did this, with the volume up only halfway on the module
and the gain on the interface actually down a pinch
I got what appears to be a nice looking level of bass on the meter and the waveform





I would guess this was the solution?... unless I'm just super f**ked up...



one more question for you, if you don't mind...
I assume you use Logic as well? if so, does this picture look okay for a track to Bounce?
because this is usually what I do, to test the way it sounds on my iPod, in the car, etc.
just wanted to get someone elses input on this... because I don't know anybody else
by the way, thanks for helpin me configure all this sh*t...

Ronald wrote:after looking for more information, I came across something
...and I think I have solved the problem...

this guy was describing pretty much the same thing I was
and he was told to go to 'Preferences > Display > Mixer tab'
and where Level Meters is, choose the Scale selection
and change from Exponential to Sectional db-Linear

If changing the meter scaling solves your issue, then there's nothing wrong with that. You may have just been over or undershooting because of the way Exponential metering works, which would explain some of the weaker looking waveforms. Do know though that changing the scale of the metering from Exponential to Sectional db-linear does not change the actual levels coming in nor going out (it doesn't increase gain). It only changes the way levels are displayed. Meaning, exponential is biased toward higher levels so that you can see finer measurements/metering near 0 dBFS, while Sectional tries to make all levels more equally distributed visually.

Those are some hot levels.

one more question for you, if you don't mind...
I assume you use Logic as well? if so, does this picture look okay for a track to Bounce?
because this is usually what I do, to test the way it sounds on my iPod, in the car, etc.
just wanted to get someone elses input on this... because I don't know anybody else
by the way, thanks for helpin me configure all this sh*t...

I use Logic quite a bit, along with other software, but I'm no expert by any means. And no, I don't mind. I'd hope someone would help me out too when I have legitimate questions.

If you're trying to bounce with plans to get it mastered later on or are archiving your project for backup, then those settings are correct. You want to keep it 24 bit and at the original sample rate that the project was recorded at for archival or if planning to get it mastered. However, if I'm not mistaken most iPods (maybe not all) do not support 24bit audio. If you're planning to play it on an iPod or actually burn the file to a CD-R yourself then your settings should be Wave or AIFF, 16 bit 44.1 kHz with Dithering ON (using either of the Weiss POW-r dithering settings) or None (if dithering using a third-party software as your last plug-in). If you just want to have a quick listen on the iPod but won't use the file for anything else, you can also select M4A and change encoding to Apple Lossless which will give you the same quality as a PCM Wave/AIFF file at 16 bit 44.1 kHz, except about half the filesize. If you want the file even smaller and need it just for judging balance but not fidelity, you can obviously use M4A:AAC or MP3, where AAC is superior at each equivalent bit-rate. If you're planning to have it in iTunes/iPod, you may want to turn on Add to iTunes Library though.

For the other settings. I'd also personally keep Include Audio Tail off and just manually select the area I want to bounce beyond any reverb tails or such but you may want to turn it on to avoid having any effects cut off that may be still playing beyond your Locators (cycle area), if you don't want to manually do so. Only change Interleaved to Split if you plan to send the files to someone using Pro Tools or another software that uses such files. Normalize is evil. Definitely keep it off unless you have a good reason and using it to make a mix "louder" not only doesn't work well but is also not a good reason. Just in case you didn't know.

can't thank you enough for the detailed response... really helping me out

Undertow wrote:

Those are some hot levels.

yea?... I guess I have the wrong idea then, because I thought that was the level to aim for, close to the top
...mostly because when I would bounce tracks, I was annoyed at the sound being so low
(I know it wasn't mastered and mixed, but it just sounded so weak compared to playback in Logic...)
in your opinion, would you say the level should peak at about where the '0.0' bar is in the picture?
or before it turns from orange to yellow?

the reason for so many is questions is because I finally got my little home studio set up
I've never done stuff like this before (using the DAW with all the equipment and interfaces, etc.)
been practicing in Logic for about a month now, trying to get familiar with it
so I'm trying to work out the kinks before I actually get down to business for real
...again, thanks a lot really..

I don't think that there are any hard rules, if there are I don't know them but I usually try to get my average levels around -12 dBFS when recording. Basically I try to keep the bulk of my signal between -12 and -6 dBFS. Some people like to record as hot as possible but I believe this is just a practice carried over from when the gear was limited to 16 bits and done to maximize resolution and for noise reasons. This is not so much needed with 24 bit and above recordings. I wasn't recording back when everything was 16 bit so there may be other reasons also that I just don't know about and because of this I can't invalidate any other legitimate reasons that may exists for wanting to record as hot as possible. As for the meters in Logic, I think they transition from yellow to orange at -5 dBFS so I guess that means I stay in the yellow with maybe only an occasional peak into the orange.

I do this because it leaves me room for processing, which itself sounds better with a bit of headroom, in my setup.

Here's how -12 dBFS would appear running a test tone through the channels, first with Expontential metering then with Sectional db-linear.



I wouldn't worry about un-mixed and un-mastered music sounding as loud as commercial records, personally. Just turn it up when listening to yours. Have fun with the setup, man.

I usually try to get my average levels around -12 dBFS when recording. I try to keep the bulk of my signal between -12 and -6 dBFS

how come when using the Software Instrument for a track in Logic
a lot of the instruments you can select are at about -12 and then some are way louder like this?




so trying to do it now like you suggested..
I turned the modules's sound lower + tweaked the interface gain, and got this level off a kick drum
but looking at the waveform, doesn't that look kinda low, or is it just fine?
since you said the last ones I posted were too hot, I've been conscious about it...


^ the light grey is the kick, the other on the left is a snyth
both audio input from the Korg


KICK


SYNTH


the recordings that looked very hot to you... when I listened back on my moniters, or iPod, or in the car stereo through iPod, it sounded decent I guess... that's why I said I had the wrong idea of where the input level should be at, because I figured the loud input would translate to a decent sounding unmastered track when bounced....and I know you said not to really worry about an unmastered track's playback, and just turn the volume up to listen back, since it's not unmastered and it's not supposed to sound that way (seems like the common sense).... so that's why this aspect is confusing to me as well.... I might be rambling a bit, hopefully you get what I mean

I guess the thing here is that you're dealing with very predictable material like synthesis and samples. The levels of these don't usually fluctuate as much as acoustic signals since those sources aren't usually as dynamic or because they are programmed via MIDI velocity before recording. So, with these you can easily push them up to around -6 dBFS or so. For me the benefit of keeping my levels between -6 and -12 dBFS is that I have a safety area between the average level, transient peaks and clipping point. Also I find that in my case, EQs, compressors and such have headroom to work better. Remember though, I'm talking about the average level of the audio being around -12 dBFS or a little higher not the peaks.



I'm not on my computer with Logic, but hopefully Pro Tools will get the point across. As you can see in the image above, the dense portion, I guess we can say the average RMS level, is near -12 dBFS while the occasional peaks have actually reached -3 dBFS on it's highest transient. This is what I meant. If you're dealing with synths or samples which are predictable there's not really much risk in having rogue transients cause clipping, as you can just adjust and optimize to the correct gain while it's playing back.

As for listening to them on the iPod, most anything louder gives the illusion of sounding better and more powerful but this is only aural illusion. If you turn the volume up, it should sound just as good. Take into consideration that mastered records have a lot more done to them to make them that loud, including often having their dynamics mashed with brickwall limiters and other molestation. I know you probably already know that though.

yea I know what you mean...

so what I'm getting is in my case I should just keep it around -6 and adjust volume accordingly in that range to even it out?

There's is no right or wrong way but that's what I would do when recording/tracking. Others may have a different and equally valid opinion though. Either way, I hope it helps somehow, man.