Skip to main content
Topic: Exact audio copy, flac to alac conversion question? (Read 4883 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Exact audio copy, flac to alac conversion question?

Hi,
I rip CD's using EAC,then convert them to Apple Lossless and listen through Itunes. I used to rip to FLAC, then manually use DBpoweramp to convert them to ALAC. Recently I discovered how to make EAC rip them straight to ALAC using ffmpeg. So my question is: is there any difference between using ffmpeg versus dbpoweramp / is one preferable over the other? I'm not exactly an audiophile but I want my files to be as error-free as possible.

More generally, do files lose quality when being converted from one lossless codec to another? Say I convert a song from FLAC to ALAC, then back to FLAC - will it be just as good as the original file? I've noticed that there's usually a slight difference in file size between every conversion but I'm not sure why.

Thank you!
last.fm/user/ihatepretending

Re: Exact audio copy, flac to alac conversion question?

Reply #1
There is absolutely no difference transcoding lossless formats; a little difference in size is to be expected.  I'm sure what you're doing is fine, but I'd use qaac as it uses the same QuickTime components as iTunes:
http://wiki.hydrogenaud.io/index.php?title=EAC_and_QAAC

Re: Exact audio copy, flac to alac conversion question?

Reply #2
Cool, thanks!
last.fm/user/ihatepretending

Re: Exact audio copy, flac to alac conversion question?

Reply #3
By the way, shouldn't the "Important Note" on the EAC and QAAC page be removed now? Looks like the strings have been updated.
last.fm/user/ihatepretending

Re: Exact audio copy, flac to alac conversion question?

Reply #4
Does EAC verify the written audio? I always check when I convert, just to ensure nothing has gone wrong (disk full or overbusy, blah blah blah). 

uselessladder:
you should take note that ALAC-in-mp4 is not checksummed, so any file corruption occurring later, would be harder to to detect. FLAC is checksummed, and there are tools (like foobar2000) which verify that the audio still is the same as when it was written.

I've noticed that there's usually a slight difference in file size between every conversion but I'm not sure why.

Lots of reasons, none of them affecting audio quality; the latter of the following could be somewhat annoying though:

Lossless codecs compress the same audio differently (which has nothing to do with whether the data is audio or something else - try 7-zip and see the difference between .zip and .7z), and e.g. FLAC has different settings where one can use more CPU to pack the same audio harder.

Lossless streams are packed in containers. They differ in how much space they do use for e.g. metadata. And even from instance to instance of the same format with the same compression settings, padding settings may make files differ slightly in size. (Padding: some empty space (zeroes) between metadata and audio, so that a change in tags will not have to trigger a complete overwrite.)

And finally, depending on formats and tool: you may or may not have embedded pictures transferred. (Meaning, if you have the album art as separate picture files and they are copied - fine, for pictures that are embedded into the audio files, it is unfortunately not so that every tool will transfer every picture.)

“It sounded bad to me. Digital. They have digital. What is digital? And it’s very complicated, you have to be Albert Einstein to figure it out.”
- Donald Trump, May 2017

Re: Exact audio copy, flac to alac conversion question?

Reply #5
Thanks for the explanation! I always use the AccurateRip plugin when I rip or convert to verify the resulting M4As.

Also a somewhat related question: I have noticed that when I play lossless files through my laptop plugged into my stereo's aux port, it appears slightly quieter than playing the same music directly from CD. This is normal, right? (just want to make sure that the volume is the only difference and nothing else gets lost in transit).
last.fm/user/ihatepretending

Re: Exact audio copy, flac to alac conversion question?

Reply #6
It's normal; no data loss. You may be able to raise the laptop output level in its volume app.  Or use a software player like foobar2000 that has a "pre-amp" function.

Re: Exact audio copy, flac to alac conversion question?

Reply #7
Seeing no mention of highly compressed tracks and replaygain...
a software player like foobar2000 that has a "pre-amp" function
will likely drive the signal into clipping before it gets converted back to analog.
Is 24-bit/192kHz good enough for your lo-fi vinyl, or do you need 32/384?

Re: Exact audio copy, flac to alac conversion question?

Reply #8
Of course, you are right.  It prompted me to look at all the CDs I've ripped from the past two years: none of them has a positive Album Gain and the worst is -11.05 dB!

Re: Exact audio copy, flac to alac conversion question?

Reply #9
Of the discs released in the '80s, simply using replaygain with the default reference is enough to drive many of them into clipping.

However, we have absolutely no idea how loud a CD with an album gain of -11dB played from OP's laptop without the correction will be compared to playing it with his CD player.
Is 24-bit/192kHz good enough for your lo-fi vinyl, or do you need 32/384?

Re: Exact audio copy, flac to alac conversion question?

Reply #10
Thanks guys!
last.fm/user/ihatepretending

Re: Exact audio copy, flac to alac conversion question?

Reply #11
Of course, you are right.  It prompted me to look at all the CDs I've ripped from the past two years: none of them has a positive Album Gain and the worst is -11.05 dB!

Of some four hundred classical CDs with the Album Gain tag:
more than 75 percent has positive Album Gain,
more than half has +2 dB or more,
nine percent has +6 dB or more
two CDs (both flute works) have more than +15 dB. They peak at about 0.3, but a 15 dB boost ...


Of my non-classical CDs with album gain (I have deleted the album gain/album peak from many compilations):
About 3 percent has positive album gain. More than half of those have album peak of 0.9000 or above.
About 1.5 percent has album gain of at least +1 dB. Most of those have album peak of 0.8000 (that is about a decibel) or above.
About half a percent has album gain of at least 3 dB.
A very few albums have album gain > +5 dB, the most popular ones are likely Jennifer Warnes: Famous Blue Raincoat and Terje Rypdal: Blue.


I tell fb2k to prevent clipping according to peak.
“It sounded bad to me. Digital. They have digital. What is digital? And it’s very complicated, you have to be Albert Einstein to figure it out.”
- Donald Trump, May 2017

 
SimplePortal 1.0.0 RC1 © 2008-2018