Clipping: The Surprising Solution
Vložit
- čas přidán 27. 02. 2024
- ▶︎▶︎ Free 5-Step Mix Guide here: www.5stepmix.com
If you get clipping regularly, this video will help you gain a deeper understanding of how levels work, how to think about gain-staging, and how to solve your clipping problem forever.
*****
CONNECT WITH ME:
Instagram: / joegilder
Facebook: / joegildermusic
Twitter: / joegildermusic
Home Studio Corner: www.homestudiocorner.com
Music: www.joegildermusic.com
#Mixing #HomeStudioCorner - Hudba
▶︎▶︎ Free 5-Step Mix Guide here: www.5stepmix.com
The only thing I will say from a pro audio standpoint that a lot of people don’t get is that gain and volume are not the same. Gain is the amount of input signal and volume is the amount of output signal. Not to criticize, but they aren’t an interchangeable term. All that to say, this video is super helpful so far.
But it IS called “gain-staging,” and that typically refers to more than just input gain.
@@HomeStudioCorner I also think it depends on the application. In a live event setting gain and volume are clearly defined. In recording, there’s gain on mics for recording, gain in mixing with clip gain, gain on the channel strip, etc.
I came from the live event world and people in that world were sticklers about the difference between gain and volume. Unfortunately old habits die hard 😂
Thank you Joe! Always helpful!
Fantastic stuff, Joe ! Such a key point that catches so many people out 👍
Thanks Joe. I wanted to make sure I heard you so, I turned up my speakers!
Most excellent. Thanks Joe. 😊
Good Video Joe for those who are having those issues...We know we all go through that stage and finally understand which like you said!....Makes the process a whole lot easier and consistent.
Man you are so right on with your approach I guess sometimes us old guys just assume that up and coming engineers already know this stuff that sometimes we fail to teach this. I have throughly enjoyed your resent videos that cover this subject of a balance "static" mix is Sooooo important and will speed up your mixing time immensely.
Great info.
Next time a foh asks me to turn my amp down I'll say :"well that's, a complicated, demand". 😂😅
Thank you!!
Loud and clear !!
Good video. Easy to get wrapped up in a holy war when discussing this.
Truth is that you can have all tracks and busses red internally in the DAW and the results might(!) null when you export it, but you already cover that. Nulling is nulling, and that is usually part of the discussion. I like your point about how some level of discipline will ultimately give you more consistent mixes, and I totally agree with this. I think this is the reason why gain-staging discussions so many times devolve into religious wars.
However: You seem to gloss over that the compressor on the ADL 700 also changes gain, literally all the knobs in that section affect the gain. This is an interesting place to discuss gain staging, because it is one of the places where the gain drastically colors the sound, which might be what you want. This obviously also includes a bunch of plugins ITB and while most of them have internal gain if it is relevant to the processing, it might be interesting and helpful to explain why where you change the sound level affects the final mix in a way that cannot simply be "fixed" by pulling on a lever on the rendered file.
You do a great job of pointing out where and how you can change the level of the signal at different places in the signal path. I think a great follow-up video would be why there are so many places to change it and why you would want to change it for what effect. I think, even if most of us don't own an ADL 700, you could use that as an example.
All-in-all I like your take, it seems measured and not dogmatic. Thanks for the content.
Good video Joe. As a matter of interest, what levels do you go for on the master buss once you're ready to mix down or master? Same logic or something different?
Nice video Joe. You are always very clear. The only thing that I am feeling a little uneasy is to record with your “near 50%, just a bit more” rule with mics that have some higher self-noise levels (like some top Neumann tube mics, e.g.). I believe one needs to have more input gain in order to avoid the need to push too much the recording made with them afterwards.
With your specific example, if the microphone itself has inherent noise, that noise is happening before you set the preamp level, which means the signal-to-noise ratio won't change no matter what you set your preamp to. In fact, if the microphone itself has noise, then turning up the preamp turns up the noise as well as the signal.
@@HomeStudioCorner Thank you very much for your comment on that, Joe. I have never thought of it like that (SNR constant independent of preamp level).
how would i turn the overall volume back up for release on youtube and spotify instead of just listening for a clean mix myself?
Just to slightly elaborate for newer additions to the audio world..... "Clipping" in analoge will normally come across as distortion when driven very hard eg, on electric guitars, when you go into the digital realm, clipping will come across as sounding like a skipping or glitching CD. Thanks Joe.
Though they are certainly related intimately...gain and volume are technically not the same thing. Gain is signal level. Volume is "loudness level" (or how the human ear perceives loudness). Joe is correct though on ease of control regarding volume...monitor volume knobs. An easy way to think about all this in very general terms is: (1) Just make sure the signal is never clipping anywhere at any stage (including within plugins) staying in the -18 dB to -12 dB range and (2) Adjust the "loudness" your ears want to hear by raising/lowering your monitoring system rather than any stages which increase signal (gain) such as pre-amps, plugins, track/bus faders, etc. This methodology is a good reason for employing monitor volume hardware such as Mackie Big Knob, Presonus Monitor Station, Behringer Monitor Controllers, etc.
I've noticed that sometimes the audio never red lines but phones will still turn down the volume like the song had clipped. Even with the daw saying there is as much as 10db of overhead still remaining. Just wondering why?
6:30 I think we can manipulate the volume like 5 ways.
Unrelated: I was mixing a track recently and did a lot of really hard panning. I got it sounding amazing, and then listened to it in mono and it sounded like hot garbage. Do you have a video about mixes that don't suck in mono? If not, can you make one?
The solution that I've found is to do all of my initial level setting and EQ in mono, then to pan and adjust from there.
A hobbyist, but I hate when I export and it clips when I thought I had the levels pretty good in Reaper. (hides due to 32-bit being mentioned)
For speed mixing, I like to export/render to 32-bit float PCM via Reaper. I then load it in Audacity and then Audacity actually automatically has the normalization levels automatically filled. That being said, if you have more than a few places of peaking, you should adjust your levels in the DAW. Also, before I adjust the levels for peaking, I will do the 32-bit export
Another thing is that it might be a good idea to normalize according to a LUFS level if you are going to upload the audio to a service like CZcams anyways. So, if the waveform render isn't perfectly filled, it might not be too bad of a thing.
Around 2008 when my hair starting to recede, I realised that the solution to getting loud, punchy mixes was the same as the solution to my follicle deficiency issues: shave it off. I’ve never looked back!
whats your take on quantization error/bit depth in relation to the level being fed to the AD?
I know you weren’t asking me, but the answer is that with a 24-bit DAC, you won’t get any. The analogue noise in any modern hardware is still enough to effectively randomise the quantisation error and make it a non-issue. (This is also sometimes called “self-dither”)
Even with 6-12 dB of peak headroom at the input, you’ll be fine 👍
@@ProductionAdvice thanks for your answer! I agree with you, thats why im asking. Joe proposes a fairly reasonable criteria but also says 'record low level, if the instrument is going to be low volume in mix'
So the question was...how lo will you go? And how you find (in your experience) that low level translates in sound regarding quant error....is it really an issue? Or not at all....or what
@@srTraseroRojo Well, I'm told by people I trust that even the best converters have a self-noise floor equivalent to 20-bit audio, so in a 24-bit converter we have 4 bits "spare".
So even if we play it safe and say we don't want to waste more than 3 bits of headroom, at 6 dB per bit that means we could have signals peaking at -18 dBFS and still be completely comfortable !
And in reality the mics and gear we plug in are likely to have noise at a much higher level than this, so in practise I think with 24-bit converters it's not something to really worry about.
Personally I prefer to see peaks around -9 to -6, but in theory even this isn't necessary. Hope that helps ?
What about clipping inside plugins? Some of them have clipping indicator, some don't, but can they both clip ?🤔
Depends on the plugin.
"If you having Clipping problems I feel bad for you son. I got 99 problems but gain staging ain't one"
Looking for mixer levelling masterclass 😌
I counted 9 ways that level could be adjusted, but I didn't account for speaker level in my count. I counted "Event Gain" and "Clip Gain" as two separate options on the waveform itself. Mission failed successfully, I guess?
Over-achiever. 😊
I clip no matter what. My interface knobs clip at 1-3, the Focusrite desktop Control mixer thing the faders can't get past -20, & even with my DAW faders down to -20 to -24db I still clip if I hit the drums harder than a mouse tip toeing
where do you clip? Inputs, Master or even the mics itself?
and at how high is your bit depth of your devices? try 24 bit, or even 32.
Right from the beginning, & I'm on 24 bit. Its way too hot from the very beginning. But my output volume is way too low even though my headphones & monitors are at full blast.
What mike are you using? Does it have a -10 db pad? My first guess would be a faulty mike - they can clip too.
@@andrewjacobs5579 ah ok. So first, make sure you dont clip your recording, even if that means, that you dont have to gain up your mics. you also can use pad switches on the mics itself. And then make sure, that you dont have some kind of trim down the signalpath. Set all the faders to unitiy and check also your trims in your interface control software and other (like sound id Reference if you use it).
So now i understand why Presonus build in a dirt stage in their Revelator series.😂
Just make sure, when you turn your headphones up, you make sure your computer or device has notifications muted. 😅
if the question still stands, I would say 14 also with the "print"
"Based on what you can see on the screen, how many ways can we adjust that volume, nothing theoretical"
The answer: 8 + all the stuff off the screen.
The answer is 8 Joe. There are 8 things on the screen Joe.
You’re right. 😊
There were at least 6 places the volume could be affected on that highlighted channel.
Only if everything on that DAW including plugins and console emulations is linear. If it's not then there are only two, the channel and the master fader. If there are non-linear pathways the gain attuentors will affect volume AND tone, which obviously is not the same thing.