Posts: 30
Threads: 7
Joined: Mar 2008
so basically even if a track is mixed below 0db, once its played for example on a cd, the level could actually end up over 0db beacuse of the DA converters in the cd player? the same with the sound that comes through the speakers via the DA converters in the soundcard?
Wouldnt this be solved in the mastering stage? so say for example, I sent a wav to you for mastering, what db should it be about? - quite a few below 0db ?
:confused:
Posts: 1,251
Threads: 137
Joined: Jun 2007
Yes this is solved in the mastering phase if it's done correctly, something I do on every track. The level you send me doesn't matter as long as it's not clipping.
Posts: 413
Threads: 26
Joined: Jun 2007
Medway Wrote:That was part of Paul's point about keeping levels low just so you can concentrate on the mix and not worry about this stuff.
Cool, and so it shall be.
Posts: 1,251
Threads: 137
Joined: Jun 2007
Also note that this applies for rolling off low end of even subtractive eq, you can end up with a louder signal than when you started which is opposite of what you'd expect.
For instance a square wave at 200hz, then apply a steep HiPass to it around 100hz or a bit higher. In my test I encountered a rise in 8.5db of level, from -6db to +2.6, and this just from cutting below the frequency of the actual wave. This is one place where a Linear Phase eq is helpful as it won't suffer from these effects as they increase in value is from changing phase.
Posts: 413
Threads: 26
Joined: Jun 2007
Damn, 8.5dB, that's quite a lot.
Posts: 281
Threads: 49
Joined: Dec 2007
Medway Wrote:Also note that this applies for rolling off low end of even subtractive eq, you can end up with a louder signal than when you started which is opposite of what you'd expect.
For instance a square wave at 200hz, then apply a steep HiPass to it around 100hz or a bit higher. In my test I encountered a rise in 8.5db of level, from -6db to +2.6, and this just from cutting below the frequency of the actual wave. This is one place where a Linear Phase eq is helpful as it won't suffer from these effects as they increase in value is from changing phase.
So does that mean in some cases we shouldn't use a low cut filter on our basslines, but instead cut out the very bottom end with an eq?
And how can we identify when this happening? My best guess is to look at the a frequency graph on an eq and then a/b with the low cut on and off?
Posts: 1,251
Threads: 137
Joined: Jun 2007
Technically an eq and filter are the same thing, just different terminology. But generally a low cut filter is more aggressive or 'steep' and less adjustable than a shelving one. You should find that more gradual slopes/cuts will have less of an effect as described above. A good way to check is just note the peak level before and after. The idea when cutting is to create more headroom just see if your channel is displaying more signal with the cut applied as without.
If you have access to a linear phase eq this shouldnt be a problem although with those you can have pre and post echos which might result in 'smearing'. It's a case of the lesser of two evils really.
Also note the above test was on a pure square wave which is a worse case scenario. So just keep an eye out on the peak read outs and you should get a feel for when it's an issue and when it's not.
