Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Converting a 24/192 FLAC to 16/44.1 (Read 17843 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Converting a 24/192 FLAC to 16/44.1

Is there any issue with doing this?  I know the extra data can probably only be heard by your pet dog and bat, so I figured, why waste the space on the HD.

Converting a 24/192 FLAC to 16/44.1

Reply #1 –
Indeed, and depending on your sound hardware and OS audio settings, you're probably doing some conversion when you play the 24/192 files anyway.

IMHO the biggest risk of audible problems would be due to using a poor sample rate converter, one that introduces aliasing artifacts. There's a comparison of converters at http://src.infinitewave.ca/ ... the best ones keep aliasing to a minimum. SoX, for example, has a good one. I personally use the format converter built into foobar2000. I use its built-in features for bit depth reduction with dither, and for sample rate conversion I use foo_dsp_resampler. This component adds "Resampler (SoX)" to foobar2000's DSPs; you can enable it in the converter settings. I set the DSP to use "Best" quality and left the other settings alone.

For bit depth reduction from 24 to 16 it is very unlikely you will hear any difference no matter what you do. The extreme quietest parts of pure digital recordings (as found in field recordings or classical, not any kind of pop/dance) may sound excessively noisy or silent in 16-bit, when played with the volume cranked way, way up. If you choose to use dither, those parts will be somewhat preserved, albeit in a lossy way with added noise. Without dither, they'll just be pure noise or silence.

If these are vinyl rips, the deafening roar of the groove walls rushing past the stylus is in the 12-bit range, tops; there's no signal to preserve below that noise floor, thus 16-bit offers plenty of headroom. Pop/rock/dance music is all way too loud for the bit depth to matter, especially if analog gear was used at any point in the chain; the lowest bits are just noise. This is the inspiration for lossyWAV: since the lowest bits aren't really used, you can make them be all zeroes so the music compresses better with lossless codecs.

Converting a 24/192 FLAC to 16/44.1

Reply #2 –
Isn't the new SSRC/dBpa resampler from foobar2000 1.3.9 beta better than SoX?

Converting a 24/192 FLAC to 16/44.1

Reply #3 –
If you look at the 1KHz tone and Passband on http://src.infinitewave.ca/, I would say SSRC/dBpoweramp is better than SOX (all modes).  dBpoweramp is listed there, and the foobar beta is using the same as the listed R15.1. Saying that some of the issues for Sox are -120dB, you will not hear that...

Converting a 24/192 FLAC to 16/44.1

Reply #4 –
Better because the noise at -170dB is different is very far fetched. The stopband may be set higher in dbpoweramp to values you need for nothing.
One thing i remember with SSRC is that it added more clipping as other resamplers when i played with it years back. This means i couldn't normalize as loud.
I don't know it counts with the dbpa implementation.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Converting a 24/192 FLAC to 16/44.1

Reply #5 –
Quote from: mjb2006 on
If these are vinyl rips, the deafening roar of the groove walls rushing past the stylus is in the 12-bit range, tops; there's no signal to preserve below that noise floor, thus 16-bit offers plenty of headroom. Pop/rock/dance music is all way too loud for the bit depth to matter, especially if analog gear was used at any point in the chain; the lowest bits are just noise. This is the inspiration for lossyWAV: since the lowest bits aren't really used, you can make them be all zeroes so the music compresses better with lossless codecs.


Nope.  Nothing is a vinyl rip.  Just stuff I got off of hdtracks.com, because I wanted a FLAC and couldn't find any place else to buy it other than getting off my lazy but and driving to a record store.

I never got the point to vinyl snobbery.  The sheer volume of stuff being released on limited edition vinyl kind of boggles my mind.

Converting a 24/192 FLAC to 16/44.1

Reply #6 –
If you don't listen to extremely dynamic music at loud levels in the quietest environments or don't ride your volume knob in less than perfect environments to the point that you would cause hearing damage with the loudest portions without backing off, then there is zero need for 24-bits in the delivery format.

Also, if there is no detail to be heard in the least significant 8 bits then you don't need them (e.g.: vinyl, music that is not highly dynamic).

Converting a 24/192 FLAC to 16/44.1

Reply #7 –
Quote from: Wombat on
Better because the noise at -170dB is different is very far fetched. The stopband may be set higher in dbpoweramp to values you need for nothing.
One thing i remember with SSRC is that it added more clipping as other resamplers when i played with it years back. This means i couldn't normalize as loud.
I don't know it counts with the dbpa implementation.


Clipping would not be problem if you:

1) Set the bit depth to 32 bit float
2) Resample
3) Run Peak to Peak Normalization to bring back within -1.0 to 1.0
4) Reduce bit depth to original 16 or 24 bit, dithering as required.

Converting a 24/192 FLAC to 16/44.1

Reply #8 –
Not what i wanted to know. If you do 24/96 to 16/44.1 you need to make it a more silent file because of more clipping added as with a standard SoX setting. This was with Naoki Shibatas SSRC.

Edit: will install foobar beta and try one maybe
Edit 2: To be fair i tested the lvqcl SoX plugin against the SSRC in foobar 1.3.9b3 and found no real difference regarding clipping. Still it is debatable what parameters with resampling should be viewed as being "better"
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Converting a 24/192 FLAC to 16/44.1

Reply #9 –
Is it more foolproof to go 24/96 to 16/48?

(By "more foolproof" I absolutely don't mean it matters if done right.)

Converting a 24/192 FLAC to 16/44.1

Reply #10 –
No, because decent resamplers do not do simple decimation (which would alias like crazy), instead they increase the frequency then reduce.

Converting a 24/192 FLAC to 16/44.1

Reply #11 –
On another thread from last year, I was given these instructions to convert a hi-res flac so I could ABX it.

Quote from: Apesbrain on
Converting Hi-Res to 16/48
These instructions are for Windows; all software is free

1. Do a "Full" installation of foobar2000 and its Encoder Pack

2. Open foobar2000 "File" > "Preferences" - "Components" and drop the SoX Resampler installation zip file on the open window; click "Apply" and restart when prompted

3. Drag your hi-res files to foobar2000 main window

4. Configure your file conversion:

- select all music files in foobar2000
- right-click and select "Convert" > "..."
- "Output Format": Format: FLAC, level 6; Output: 16-bit; Dither: always
- "Destination": Output folder: ask later. Output type: tracks into individual files. File name pattern: /%artist%/%album%/%filename%
- "Processing": Resampler (SoX): configure Resampler for 48000 "Target Sample Rate"
- "Other": When finished: ReplayGain-scan output files as albums. Optionally, select "Transfer attached pictures" if your files have embedded artwork.  If you have a cover art JPG in the folder, put *.JPG in the box labelled "Copy other files..."
* IMPORTANT - Save this configuration as "Hi-Res To 16/48" preset so you can easily use it again
- Press "Convert" -- BE SURE TO SPECIFY AN OUTPUT FOLDER DIFFERENT FROM WHERE THE ORIGINAL FILES ARE LOCATED


Is there any reason I should stick with 48K and not  use 44.1K?

Converting a 24/192 FLAC to 16/44.1

Reply #12 –
Quote from: apastuszak on
Is there any reason I should stick with 48K and not  use 44.1K?

Probably not but since you know how to ABX why not test it for yourself?  Let us know your results.