Jump to content

Which are the recommended OPTIMAL WAV / MP3 bounce settings?


Recommended Posts

I have always exported WAV at 44 KHz, 24 bit, Interleaved, normalize OFF and no dithering. (both for mixing and the final mastered bounce)

 

Would you consider these settings optimal or which settings would you recommend?

 

ALSO, should I keep exporting in WAV or move to AIFF?

 

A FEW more questions...

 

1) Should I use dithering? If so, both on the mixing export and the final master export? Pow-r 1, 2 or 3?

 

I read that if you are exporting a WAV for master on 24 bit, you don't need to use dithering UNTIL you export the final master in 16 bit?

 

2) Should I export at 48 KHz instead of 44 KHz? (I read 48 is used for video?)

 

3) When exporting an MP3... shoud Bit Rate Mono and Bit Rate Stereo be the same? Should I use "Best Encoding" and "Filter Frequencies below 10 hz"? Is Joint Stereo the best option or Normal?

 

I'd TRULY, TRULY, TRULY appreciate the information. I want to gather a general consensus out of this. Thanks for your time! :shock:

Link to comment
Share on other sites

I have always exported WAV at 44 KHz, 24 bit, Interleaved, normalize OFF and no dithering. (both for mixing and the final mastered bounce)

 

Would you consider these settings optimal or which settings would you recommend?

If you're working at 44.1 kHz then yes:

 

WAV

24 bit

44.1 kHz

Interleaved

Normalize = Off

*No dithering

 

*Logic adds TPDF (flat dither) going from the internal 32 bit float signal to a 24 bit fixed file, but not to 16 bit where you need to choose your own algorithm or Logic will simply hard truncate. The choice of dither at the 24 bit stage has virtually no significance, though you should generally avoid adding noise shaped dither twice, which would make POW-r#1 the best alternative early in the chain (if you insist).

 

ALSO, should I keep exporting in WAV or move to AIFF?

WAV is more cross-platform compatible and less error-prone - especially with Logic Pro. Though you will probably never experience this, there's an old and very unfortunate AIFF bug in Logic that can cause the AIFF file headers to corrupt and give you what appears to be unreadable 8 bit files. It's fixable, but it requires you to manual edit the ressource part of the audio files using HexEdit or similar.

 

I've never had that bug myself, but I know a couple of people who have, as recent as a couple of weeks ago, though it was fixed with the help of a ressource editor. I switched from AIFF to WAV/BWF when I switched to Intel-based Macs and never looked back.

 

A FEW more questions...

 

1) Should I use dithering? If so, both on the mixing export and the final master export? Pow-r 1, 2 or 3?

Mix bounce for mastering: use the above mentioned settings.

 

Final master export (16 bit WAV): up to you. The noise-shaped POW-r#3 is theoretically the most powerful of the four (including UV22HR) which is why I use it for some very dynamic sources such as classical music, but for most other purposes I prefer the shaped dither of POW-r #1. There are even better dither algos out there, but it's of little significance.

 

I read that if you are exporting a WAV for master on 24 bit, you don't need to use dithering UNTIL you export the final master in 16 bit?

Not quite correct, though correct in practice in Logic Pro. You should theoretically apply dither at any stage where you reduce bit depth in order to avoid quantizing noise. Noise-shaped dither (a special dithering technique) should be reserved for the final bout as mentioned earlier. Also, as mentioned, Logic adds TPDF from 32 bit float to 24 bit fixed which means we don't need to worry about adding a custom dither to our 24 bit files.

 

2) Should I export at 48 KHz instead of 44 KHz? (I read 48 is used for video?)

Only if you're already working in a 48 kHz project or the output of a 44.1 kHz project needs to end up in 48 kHz, such as a special tv/video/movie/broadcast version of a song.

 

Be aware that sample rate converting (including bouncing to a different sample rate) will usually cause small overloads with brickwall limited material and you should increase headroom by at least a few fractions of a dB in those cases.

 

In any case you should theoretically have at least 1 dB of headroom when bouncing to a lossy format such as MP3 or AAC to account for conversion overs and inter-sample peaks, but this is often ignored in order to maximize loudness.

 

3) When exporting an MP3... shoud Bit Rate Mono and Bit Rate Stereo be the same? Should I use "Best Encoding" and "Filter Frequencies below 10 hz"? Is Joint Stereo the best option or Normal?

The mono bit rate setting isn't applicable unless you're actually bouncing to a single mono file on the disk. This means that the setting doesn't kick in with mono material in a stereo bounce, where the setting is instead completely ignored.

 

You can bounce to a real mono file by changing the Stereo Output channel strip in the mixer to mono (one circle) and clicking the bounce button directly on the output channel strip - not using the bounce menu option, which will always bounce to stereo regardless of the output channel strip setting.

 

Yes, use Best Encoding.

 

No, don't filter frequencies below 10 Hz with a loud mastered file. While it seems to make sense to filter out inaudible sub sonics this filter will change the relations between the frequencies which can cause more unnecessary overs during the lossy MP3 conversion as the waveshape integrity is further deformed. This is mainly a problem with loud masters, having been subjected to a squaring of the waveform from brickwall limiting or intentional clipping. With dynamic material with at least a couple dBs of headroom you can turn on this option without worrying about overloads, though it will always impact the waveform in some way.

 

Use Joint Stereo. Despite some myths about Joint Stereo and Normal Stereo (especially on audiophile websites), Joint Stereo is almost always the best choice because it intelligently uses M/S to get the best quality out of the encoded bits.

  • Like 1
Link to comment
Share on other sites

  • 3 weeks later...
  • 2 years later...

Holgar,  you said this in a post---

 

"No, don't filter frequencies below 10 Hz with a loud mastered file. While it seems to make sense to filter out inaudible sub sonics this filter will change the relations between the frequencies which can cause more unnecessary overs during the lossy MP3 conversion as the waveshape integrity is further deformed. This is mainly a problem with loud masters, having been subjected to a squaring of the waveform from brickwall limiting or intentional clipping. With dynamic material with at least a couple dBs of headroom you can turn on this option without worrying about overloads, though it will always impact the waveform in some way."

 

I understand that you are talking about wav - mp3 here, but I take it that there is no such problem at all in a

 

1)  32 - 24 bounce mix?

or a

2)  24 - 24 bounce copy?

 

thank you for so often sharing your expertise,

Kaiser

Link to comment
Share on other sites

Holgar,  you said this in a post---

 

"No, don't filter frequencies below 10 Hz with a loud mastered file. While it seems to make sense to filter out inaudible sub sonics this filter will change the relations between the frequencies which can cause more unnecessary overs during the lossy MP3 conversion as the waveshape integrity is further deformed. This is mainly a problem with loud masters, having been subjected to a squaring of the waveform from brickwall limiting or intentional clipping. With dynamic material with at least a couple dBs of headroom you can turn on this option without worrying about overloads, though it will always impact the waveform in some way."

 

I understand that you are talking about wav - mp3 here, but I take it that there is no such problem at all in a

 

1)  32 - 24 bounce mix?

or a

2)  24 - 24 bounce copy?

 

thank you for so often sharing your expertise,

Kaiser

He was talking about the "filter frequencies below 10hz" switch for MP3 export, 24bit bounces don't have that options. But yes, especially non-linear phase filters WILL affect phase relations and in consequently, waveform. (Linear phase introduce ringing, which also affects waveform). In general, if you plan to do filtering on the master, do it PRE-LIMITER, or do have at least 3dB head room.

Link to comment
Share on other sites

In "virtually anything, audio or video," there are always three levels in which you can find yourself working ... but, "once you go down, you can never go back up."

 

(1) "Floating-point, 64-bit integers" ... the stratosphere.  The original-production software is talking to itself, in its own terms.

 

(2) "The widest possible integer-only format that you can manage."  (If faced with several such targets, target each one individually, not as a chain.)

 

(3) "Anything and everything" that involves still-tinier (16-bit?) integers, or especially, compression.

 

Whenever possible, always try to work from the most-expansive source to the least destination.  Never use a "compressed" source as the input to any subsequent destination.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...