OK, I admit it - I’m somewhat of a n00b audio engineer (though not a horrible musician/composer). But this forum is so helpful, I don’t mind submitting basic questions like this. You guys are always helpful. This post goes in a few directions (several questions), but it’s all kinda related.
I understand that clipping is bad… so I bring levels down to avoid it and make sure there is no clipping even after export. But I don’t really understand what is considered the “ideal” level, if there is such a thing, or even the negative values on the meters. Meaning I don’t know how close to the clipping line I should be pushing my music to. For instance, in SoundForge, why do they show level markers at -6.0 (see screenshot). Is that a meaningful number, or just a guideline? If my audio is say at -0.5, is that bad? I actually have some background music at -0.1 after exporting, is that crazy stupid?
I do music/sfx for games. Isn’t it possible that some combination of background music + 1, 2 (or more) sfx playing at the same time could cause “clipping” dynamically, even though each of those didn’t have clipping individually? If so, I guess it’s as simple as testing the music/sfx at once in the DAW and ensuring there is no clipping, but with so many combinations to test… is this the way others do it?
I know you can blow speakers by applying too much power (pushing more than what the speaker is rated for). But can’t it also happen from clipping, because the speaker is not oscillating correctly?
CONCLUSION: My iPad recently has blown speakers (just recently and I’ve playtested on it for years), and it could be complete coincidence, but I’m kinda worried that maybe it’s because of some dynamic clipping occurring in a game I’ve done music for… nothing that I can nail down specifically, because like I said everything is exported below clipping line. But perhaps the levels are too close to the clipping line individually, and combined… a problem? Usually I run through a mastering tutorial (that I hardly understand), that I know brings the overall level down a bit. But this recent game, I did not do that, and so perhaps (combined) the levels of my music and sfx are too much at certain points in the game?
If it is something I’ve done, obviously I want to fix it before launch to prevent blowing speakers on anyone else’s device.
By default your DAW shows peaklevels, that is the level of the loudest sound in your recording. It does not show the average level, which does say much more about the perceived loudness of a piece of music.
But yes, your peaklevel should never exceed 0 dB otherwise you get clipping.
All your tracks can be below 0 dB but combined they can make your mixbus clip, so you have to bring back the levels of all your tracks.
Speakers " don’t like" clipping so yes they can be damaged but if that is what happened to the ipad I do not know.
My answer is not complete because there is a lot to say about this subject. Scan this forum or the entire internet and you will find a great deal of information about this.
You might want to look for a beginner’s book on audio engineering. I can’t really recommend anything specific since all the books I have on the topic are about ancient technology. But I’m sure reading some reviews on Amazon will turn up a number of good selections.
Also Berklee Collage of Music (via Coursera) is offering a free class, Introduction to Music Production, starting July 19. I’ve taken the Berklee/Coursera classes on Songwriting (really lyrics) and Improvisation which were both great and very much exceeded my expectations. I’m certain you’d find a ton of useful info in this class. And check out their other offerings, lots of good stuff - and all of it free.
I enrolled in that Coursera course on Music Production. After reading/watching the intro, I think simply from my experience I will know much of the material covered (I’ved used both Cubase and ProTools and other audio tools for years), but I’m sure there will be pointers or explanations to things that I simply didn’t know because I lack the formal education. Thanks very much for suggesting this! I will also check out those books.
I think this is a possibility for what happened to my iPad speakers. There are parts in a particular game I’m working on where multiple sfx can be triggered at once, so those combined with the background music (all of which are pretty high level just under clipping line) likely pushed the mixbus one too many times recently. I guess in general it’s best to leave some headroom for all the sfx and music so that combinations aren’t deadly, because it’s kinda difficult to preview all dynamic audio scenarios that could happen in the game.
No one else working on this game has blown speakers on their devices, so it could just be product wear… but just in case I’ve backed levels down a bit across the board.
In conclusion, is there a typcial level that I should shoot for? Like -2.0 or something? Or does it not matter at all? I only remember -2.0 because of some mastering tutorial I have that trims it to that, but I have no idea if that level is significant in some way.
Really anything on your output below 0dB is fine. I usually stick a limiter on my output, the brickwall limiter in C7 is great for this, and set it to -1 to -2dB, but its not a magic number - just an easy one.
Just wanted to say that I took that Coursera course on Music Production (it ends next week), and it was really good. Only about a third of the videos had new info for me, but considering there are like 60+ videos, that’s pretty good Very useful for anyone that wants to brush up on formal audio engineering basics. You don’t have to do any of the assignments, the video material alone is useful. I’m sure they’ll offer it again.
" 2. All your tracks can be below 0 dB but combined they can make your mixbus clip, so you have to bring back the levels of all your tracks."
That’s an incorrect statement, don’t follow that advice.
Just make sure your mix bus stays out of the red, with or without a master limiter.
If you go into the red, you’ll know it, since Cubase lights up a big red “over” light.
I set my meter preference to hold that warning infinitely long, so I never miss it.
Now if you take the time and think about the quote, you’ ll see it says exactly what you are saying: If the master clips, you need to bring the levels down…
Correct, valid and good advice…
That’ s not incorrect, it’ s one of two ways to do it. And in case you have some fixed point plugins in the master inserts, it’ s the only correct way of the two (using the input gain being the second). And please let’s not discuss if having plugins in the master inserts is correct or incorrect.
“That’ s not incorrect, it’ s one of two ways to do it.”
I guess if you have a master bus plugin you’re worried about, there’s a third way as well.
You can bring down the trim control on Cubase’s master bus, since it’s before the inserts.
If you have a plugin that distorts in a master insert, the trim is handy for getting it to distort “just enough”.
Yes, my mistake, I took that to mean that the input channels had to be lowered. -Sorry.
So, a final re-statement for clarity: If you like your mix, but it’s too hot, you can just bring down the master bus instead of re-balancing your input faders.
You can do that with the master trim if you want to alter the master plugin drive, or with the master fader if you just want to set the level for the output file.
Cubase’s floating point math will compensate for any unintended (plugin related) overages within the mixing bus, as well as elsewhere.
I agree with that, and would just add a word of caution - just because Cubase can handle overages internally as a result of its 32 bit floating point math, that doesn’t guarantee that all the plug ins can. I learned that by trial and error, and the help of the good people on this forum - I’d be getting distortion because the input to Plug in “X” on track “Y” was too high … even though the master out was way below 0 dBFS.
As a result of that, I’m trying to record so my peaks on any given track are around -12 dBFS.
There’s a nice review of gain-staging in the September Sound on Sound magazine.
"One of the things we like about analogue gear is the soft saturation distortion it can add when you push it hard. And plugins that emulate this effect do the same.
So if the whole mix is running up into the red on the channels, and you’re using some processing that works in this way, you may end up with much more saturation and distortion than you intended. Better to stay on the safe side and keep the levels lower to avoid this happening by mistake.
And that means running all the channels in the mix at a lower level, not just pulling back the master fader. 32-bit float will prevent clipping, but if all your channel faders are maxing out, their plugins could be quite heavily overcooked – and pulling the level back at the last minute won’t help with that."
So after discussing master trim and its output fader, and then the channel trim and its output fader,
it appears that the real art of recording is in the way we hit the limiter and overdrive plugs.
BTW, that’s what I’ve always done, even in the analog days.
Generally, I get used to the way my limiters sound vs their meters, and when they start getting too far out of the norm, I consider cutting or boosting my trim.
This gives an additional bonus: if I always set my faders to zero, I can quickly test fader changes, and later I can visually scan the mixer for any faders I inadvertently forgot to return to zero.
I note the fader’s offset, and if i liked it, I adjust the channel’s plugin to reflect the offset. A control click then sends the fader back to zero, overcoming one of Cubase’s longtime weaknesses- the ability to undo level changes.
I’ve gotten so reliant on this crutch that I insert a UAD Cambridge (an EQ with level trim) last on commonly tweaked channels, like vocals.
A side note: on an old Neve console, I used to keep the fader high on the snare etc channels, with the mic trim low, because I knew for a fact that the snare was always clipping otherwise. (I could see it on a scope.)
No one else did that, and I felt smug because I was more technically oriented than they were.
Bottom line, it turned out that the clipping and transformer distortion made the snare sound better when I just did what everyone else did.
Draw your own conclusions.
But I’m sure reading some reviews on Amazon will turn up a number of good selections.
Very useful for anyone that wants to brush up on formal audio engineering basics. You don’t have to do any of the assignments, the video material alone is useful. I’m sure they’ll offer it again.