Jump to content

export bitrate


greg54

Recommended Posts

There's no real advantage in exporting at 32 bit vs 24 bit.   

32 bit is a floating point format, whereas 24 bit is an integer format...  so although 32 bit may technically support a greater dynamic range,  for all practical purposes in a final export, they're equivalent in quality.

32 bit has an advantage when recording/mixing, because it's basically impossible to clip even when going to 1000's dB, and you can go incredibly quiet without losing information. This is why all DAW's use either 32 bit float / 64 bit float internally.  

This is essential when mixing multiple tracks, but of little use with a single exported file - unless of course the file is incredibly loud or incredibly quiet.

In 99.9% of cases, your final export will be in the normal dynamic range of human hearing.  So 24 bit is fine.  In fact I'd argue that 16 bit is more than adequate as a final export format, as it more than surpasses the dynamic range of human hearing.

  • Like 5
Link to comment
Share on other sites

2 hours ago, msmcleod said:

32 bit has an advantage when recording/mixing, because it's basically impossible to clip even when going to 1000's dB

The "1000's dB" caught my eye here. That will definitely smoke commercial hardware (why a limiter on the output buss is always advised) :) One of the most powerful (intentional) sonic transmissions is active SONAR, which can reach nearly 300 dB and will instantly kill animal life within close proximity. That kind of power is not a concern for going deaf... it is for being lethal.

  • Like 1
Link to comment
Share on other sites

5 minutes ago, mettelus said:

The "1000's dB" caught my eye here. That will definitely smoke commercial hardware (why a limiter on the output buss is always advised) :) One of the most powerful (intentional) sonic transmissions is active SONAR, which can reach nearly 300 dB and will instantly kill animal life within close proximity. That kind of power is not a concern for going deaf... it is for being lethal.

What is in your audio file is largely irrelevant during playback, as everything is clamped to 0db at the point it gets sent to your audio interface, as it needs to be converted to either 24 bit or 16 bit (depending on what your audio interface is set to).

 

  • Like 1
Link to comment
Share on other sites

I thought turning it up to 11 was great back in the day. Now your telling me we got a 1000db dynamic range!

I was going to create a few .1hz-1.0hz audio wave files and see what they would sound like at 1000db. Now you've ruined everything by this "everything is clamped to 0db" 😆

Link to comment
Share on other sites

Export several versions into 24bits, each time lowering the volume by -6dB.

Then load all of them as different tracks and bring back the level so all of them match in level.

Comparing one by one with original, notice at which point you can perceive the difference. Each 6dB is one bit, so -48dB is what you will get in case you save into 16bit WAV.

PS. you don't have to publish the truth, it is ok if you will have to go down to -60dB or even lower, no-one will know ;)

PS.PS. publishing you already start notice if you change dithering algo on 24bit master (maxed) export make little sense (on this forum) 😏

PS.PS.PS. note that processing (EQ + compression + effects + maximizing) can amplify some parts with ridiculously high ratio, so the results of the "test" are valid for final rendering only. Keep intermediate renderings in 32bit.

  • Like 1
Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...