| Author
|
Peak + 0.86 db - not clipping / cubase / protools
|
MuckyPuh
IsraTrance Junior Member
Started Topics :
14
Posts :
86
Posted : Jun 27, 2013 13:23:16
|
Hey mates °
the last days i have been checking mixes of friends and i came across something confusing.
In cubase there is the AUDIO - Statistics option that gives you information about an audio files peak and rms aplitude.
normaly when i prepare an mp3 for the web i export the final bounce so the peak is more or less - 0,03 db
as everything (digitaly) over 0 db would make the sound clip, right ?
If i look in cubase at the statistics of an ordenary mp3 i downloaded, the peak is most of the time > 1.5 db
i know that if you master or mix your music analog you can and should go through most mixing tables with +4 dbu
but if you then record the music again digitaly - and convert it to mp3 - how come it does not clip?
but the biggest confusion for me was that a song a friend mixed in pro tools with a brikwall limiter on the master - which did not clip in pro tools - appears in my cubase statistics with a peak at 0,86 db ...
after i told him - he doublechecked with izotope ozone - and says again the song does not go over 0 db ..
if i listen to it in winamp it does not sound clipped ... but i guess if i export it with cubase it then would ... how can this be ? its not like cubase is giving an extra 1 db to the file while importing ...
  http://www.muckymusic.blogspot.de/
http://soundcloud.com/muckypuh-muckymusic |
|
|
|
TimeTraveller
IsraTrance Full Member
Started Topics :
80
Posts :
3207
Posted : Jun 27, 2013 14:16
|
it is not a constant red on the channel fader but only sporadicaly I assume. It takes 3 channels in cubase when red is occasionaly to become red on the master.Just woke up need coffee for more details.
Im not sure but there was something about cubase real clipping level.. I think it is 0,86 dbfs where the headroom really ends (in cubase, means it is 0 dbfs when the master fader shows 0,86 dbfs but cubase shows 0 earlier, you than still have a bit), not sure though, but sure cubase gives you more than it shows, maybe someone can post it more properly detailed.mmmh I love that pro tools limiter.
https://soundcloud.com/shivagarden |
|
|
|
MuckyPuh
IsraTrance Junior Member
Started Topics :
14
Posts :
86
Posted : Jun 28, 2013 11:49
|
thanks for hint with the lights - but actualy i was not even looking at them.. i was just importing the final mp3 into a new cubase project and hit: "audio - statistics".
so if i finish my track in cubase i should level it to more or less + 0.85 db so it will be more or less -0,01 db "in real" ?
so how is it then with mastered mp3´s ? are they bounced from the DAW with less then 0 db and later in mastering put up a few more db ? ... if so .. how can i record the track digitaly again and convert it to mp3 without clipping ?
http://www.muckymusic.blogspot.de/
http://soundcloud.com/muckypuh-muckymusic |
|
|
|
TimeTraveller
IsraTrance Full Member
Started Topics :
80
Posts :
3207
Posted : Jun 28, 2013 18:21
|
Make your track unpeakable over 0 or better unclippable and master it or leave headroom for mastering engineer.
At the end set in maxim (protools) or any other brickall limiter output to -0,1 or what you want & anything will be alright, forget befor it audio statistics for later and the figures +0,85 or 0,86 etc all is ok.
I'm not sure I get your last 2 questions. I don't think many master really mp3's but generally speaking all tracks need headroom before mastering, so yes all the wav's need less then 0 dbFs for later mastering chain where at the bottom of it a brickwall limiter even puts them a few db's up.
https://soundcloud.com/shivagarden |
|
|
|
Babaluma
IsraTrance Junior Member
Started Topics :
18
Posts :
729
Posted : Jun 29, 2013 13:52
|
|
Colin OOOD
Moderator
Started Topics :
95
Posts :
5380
Posted : Jun 29, 2013 14:42
|
When you decode an mp3 to WAV, it reconstructs the waveform mathematically from a (relative) handful of data, and in doing so sometimes produces peaks greater than 0dB, especially with mastered audio which is usually hard up against the full-scale limit. I think you'll only see values >0dBFS in the Cubase stats if you have it set up to record (and, presumably, decode) into 32-bit audio, as they would otherwise just clip at 0dB. Neither the Cubase stats routine nor the Cubase meters catch inter-sample peaks as far as I know.
Mastering - http://mastering.OOOD.net :: www.is.gd/mastering
OOOD 5th album 'You Think You Are' - www.is.gd/tobuyoood :: www.OOOD.net
www.facebook.com/OOOD.music :: www.soundcloud.com/oood
Contact for bookings/mastering - colin@oood.net |
|
|
|
Melange5738
IsraTrance Junior Member
Started Topics :
13
Posts :
142
Posted : Jun 29, 2013 22:07
|
|
Actually, if you read the Cubase manual it says something like because it uses floating integers to handle the peaks it should not necessarily clip just because you go above 0 dBFS and that it can handle significant values above 0 dBFS before clipping occurs. |
|
|
|
Colin OOOD
Moderator
Started Topics :
95
Posts :
5380
Posted : Jun 29, 2013 22:32
|
Quote:
|
On 2013-06-29 22:07, Melange5738 wrote:
Actually, if you read the Cubase manual it says something like because it uses floating integers to handle the peaks it should not necessarily clip just because you go above 0 dBFS and that it can handle significant values above 0 dBFS before clipping occurs.
|
|
Yes, that's because it works at 32-bit float internally. Virtually unclippable. However whether or not a decoded mp3 file can contain these overs depends on the bitdepth of the WAV file created by the decoding process (which is what happens when you drag an mp3 file into Cubase); the only way an audio file will give stats showing a maximum sample value >0dB is if that audio file is 32-bit float. If the mp3 we're talking about is decoded into a fixed-point audio file, that file may well clip if the inter-sample peaks are large.
Mastering - http://mastering.OOOD.net :: www.is.gd/mastering
OOOD 5th album 'You Think You Are' - www.is.gd/tobuyoood :: www.OOOD.net
www.facebook.com/OOOD.music :: www.soundcloud.com/oood
Contact for bookings/mastering - colin@oood.net |
|
|
|
frisbeehead
IsraTrance Junior Member
Started Topics :
10
Posts :
1352
Posted : Jun 30, 2013 01:16
|
|
What if the option for detecting inter sample clips was ticked in Ozone? Would this still be the case? |
|
|
|
Colin OOOD
Moderator
Started Topics :
95
Posts :
5380
Posted : Jun 30, 2013 06:12
|
|
frisbeehead
IsraTrance Junior Member
Started Topics :
10
Posts :
1352
Posted : Jul 1, 2013 17:17
|
It's funny that thread. Also agree that it changes the sound a lot and it's mostly unpleasant.
Back on topic: even using Ozone and making sure there's nothing going overboard, that's really not the end of the chain if you bounce the file down in some daw (which one was it?) with different options in bit depth or dithering in Ozone and then on the daw. Was any of these things taking place?
Because, correct me if I'm wrong here, the moment you select the bit depth you want, say 24 or 16 bit, the 32 bit floating point that Cubase works internally, is out of the equation for good, and Ozone is processing the sound with the available headroom you've selected, right?
|
|
|
|
Colin OOOD
Moderator
Started Topics :
95
Posts :
5380
Posted : Jul 4, 2013 03:46
|
If I get you right: adding dither of a particular bit-depth does not in and of itself reduce the bit-depth, it simply adds noise on the least-significant bit of the depth selected. I'm not an Ozone user but according to my research, Ozone is similar to every other dither plugin I've used in that it does not truncate the bit-depth as the audio passes through it; rather this process must be done manually in your audio editor or DAW. If you run 32-bit audio through it you'll get 32-bit audio out of it, even with 16-bit dither applied.
Mastering - http://mastering.OOOD.net :: www.is.gd/mastering
OOOD 5th album 'You Think You Are' - www.is.gd/tobuyoood :: www.OOOD.net
www.facebook.com/OOOD.music :: www.soundcloud.com/oood
Contact for bookings/mastering - colin@oood.net |
|
|
|
frisbeehead
IsraTrance Junior Member
Started Topics :
10
Posts :
1352
Posted : Jul 4, 2013 16:22
|
Yeah, I know dithering is some form of noise shapping treatment to cover from bitdepth reduction: when you go from 24 to 16 bit, you're changing the headroom and thus the signal-to-noise ratio as well, right?
What I meant was if such a problem couldn't arise from using different settings on Ozone and on the exporting options of the DAW couldn't be the cause for such problems in the first place.
I think (I'm not sure) that if you set it to 16bit you're immediately affecting the way it will perceive the audio and the headroom on the output, specially where dithering is applied. It will interpret the "noise to signal ratio" differently and if these is applied and then applied again exporting on the DAW with other options, it can produce artifacts and irregularities in the files.
|
|
|
|
Colin OOOD
Moderator
Started Topics :
95
Posts :
5380
Posted : Jul 4, 2013 17:15
|
|
frisbeehead
IsraTrance Junior Member
Started Topics :
10
Posts :
1352
Posted : Jul 5, 2013 00:08
|
No particular parameter. I know a plug-in can't actually change the audio format, but when it applies dither to a file, it zeros out the bits bellow the bit depth chosen. And dithering is one of those things you only want to do once, right?
Second step should be choosing the bit depth on the host app and bouncing down - so that the audio file is truncated, leaving out the bits that are uncessesary, right?
I was asking if introducing two levels of dithering, and perhaps level changes after it's been applied could be the cause for the intersample peaks - and not just the difference in headroom between 32 bit (practicly endless) and 24 or 16 bit.  |
|
|
|
