Old 07-03-2021, 08:40 PM   #1
ReaperStudent1
Human being with feelings
 
Join Date: Jan 2016
Posts: 152
Default Problems porting 44.1K projec to 48K

I recently ported a Kontakt project (using string sample libraries from Spitfire and Native Instruments) from 44.1K to 48k, while keeping a safety 44.1 K copy of the project.

I found that mp3’s Rendered from the 48k project sounded glassy and shrill, even though the actual project sounds pretty much the same at 48k and 44.1k

Has anyone run into this problem?…Specifically mp3’s generated at 44.1 sounding different from 44k even though the underlying projects (48k and 44.1K) don’t sound different?

This project was pure midi...never bounced to audio. All fields in the Project Settings indicated it was properly playing at 48 k. No errrmessages form Kontakt or anything.
The rendering was “default” 320K , “Extreme High Quality” settings.

Thanks for any help...
ReaperStudent1 is offline   Reply With Quote
Old 07-11-2021, 08:59 PM   #2
PsychoCircuitry
Human being with feelings
 
Join Date: Sep 2018
Posts: 60
Default

Some thoughts on this issue. Firstly, I dont use reapers built in mp3 export. These days I'm exporting at 88.2 32bit fp and then using external software to perform sample rate conversion and mp3 encoding when necessary. However, I do believe reapers built in option uses the lame encoder, which I do know some about. Firstly, lame's algorithms are not very well tuned for 48khz material, there is info about this elsewhere if you want to look into it. Secondly, although the mp3 specs do support 48khz encoding, it does not expand the encoding bit rate based on higher sampling rate. So while you are getting more windows of time in your file, those extra samples still have to fit into a 320kbps stream, thus less data is available per sample. Mp3 is already pretty heavily compressed even at max rates, so losing more fidelity to a higher sampling rate could reduce the quality to, most definitely, a perceptible level. It's almost always recommended to encode mp3s at 44.1Khz for broad frequency spectrum material, to ensure the highest possible quality from the resulting file. There is nothing wrong with working in 48khz and then resampling to 44.1 for mp3. The quality loss in the sample rate conversion will be far less than the encoding at the higher sampling rate. I hope this is helpful and makes sense.
PsychoCircuitry is offline   Reply With Quote
Old 07-11-2021, 10:37 PM   #3
RovingEye
Human being with feelings
 
Join Date: Jun 2021
Posts: 55
Default

Quote:
Originally Posted by PsychoCircuitry View Post
Some thoughts on this issue. Firstly, I dont use reapers built in mp3 export. These days I'm exporting at 88.2 32bit fp and then using external software to perform sample rate conversion and mp3 encoding when necessary. However, I do believe reapers built in option uses the lame encoder, which I do know some about. Firstly, lame's algorithms are not very well tuned for 48khz material, there is info about this elsewhere if you want to look into it. Secondly, although the mp3 specs do support 48khz encoding, it does not expand the encoding bit rate based on higher sampling rate. So while you are getting more windows of time in your file, those extra samples still have to fit into a 320kbps stream, thus less data is available per sample. Mp3 is already pretty heavily compressed even at max rates, so losing more fidelity to a higher sampling rate could reduce the quality to, most definitely, a perceptible level. It's almost always recommended to encode mp3s at 44.1Khz for broad frequency spectrum material, to ensure the highest possible quality from the resulting file. There is nothing wrong with working in 48khz and then resampling to 44.1 for mp3. The quality loss in the sample rate conversion will be far less than the encoding at the higher sampling rate. I hope this is helpful and makes sense.
Enlightening, thanks!
RovingEye is offline   Reply With Quote
Old 07-14-2021, 07:30 AM   #4
serr
Human being with feelings
 
Join Date: Sep 2010
Posts: 12,557
Default

If there's a way to get access to the source audio the mp3 copy was made from, that would be the thing to do. If the only copy of some mix elements is a mp3 copy and if it's in a fragile state (eg already converted back and forth a few times or whatever happened) and degrades from just one sample rate conversion from 44.1k to 48k, then keep the project at 44.1k! (Convert any new audio to 44.1k)

If the final master needs to be 48k for bluray format or something, you probably will be none the wiser converting the full final mix at that point. The mp3 source elements would be baked in and shouldn't be a problem.

There's pursuing nuance and keeping audio HD from recording to mix to master and delivering full HD audio to the end consumer and that's all well and good. But if you already have not only SD audio but only lossy compressed mp3 copies of it to begin with, the HD ship has sailed! Keep the artifacts at bay.
serr is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 04:55 AM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.