PDA

View Full Version : Re: Which tracks are "best" AKA 'to remaster or notremaster'



Mark Lanctot
2006-01-17, 16:56
If they were both ripped using the same settings,
they should be identical.

zooropa320 wrote:
> Hi Mark & thanks for your detailed reply. The
tracks in question have
> already been ripped to VBR MP3s using EAC & LAME
with the alt-preset
> standard switch. This is more a question of how to
identify which of
> the resulting files are "better" when there are
duplicates.
>
>

--
___________________________________


Mark Lanctot
___________________________________

zooropa320
2006-01-17, 18:17
Looks like your reply ended up as a new thread.

Keep in mind that these are different discs (potentially) mastered at different times but they have been ripped and encoded with the same settings.

With all due respect Mark, I think you may be simplifying the potential differences. Think about the original release of The Rolling Stones release of 12 X 5 with Time Is On My Side vs. the Forty Licks 40 years later. Potential editing differences aside (to fit 40 tracks onto 2 cds) I believe they will sound different even though they were ripped the same way. A more agregious example would be if this song appeared on a time-life (or lesser quality) compilation disc. I would think there are sonic differences even though they are both being ripped from CDs in the same way.

My source discs are original releases, remasters, various artist compilations and soundtracks. I have identified over 2,500 duplicates and would like to know if there's an objective way to determine the best versions (sonically speaking) without having to listen to them all individually.

pfarrell
2006-01-17, 18:35
zooropa320 wrote:
> Keep in mind that these are different discs (potentially) mastered at
> different times but they have been ripped and encoded with the same
> settings.

Its actually worse than that. Not easier as others have implied.

Back when I was ripping 40,000 CDs we had a fair number
ordering mistakes that got us two or more copies of the same
CD. Same artist, album, songs, etc. Same UPC code, two or more
physical CDs.

Sometimes the songs had different bits.
We didn't listen to them all, but most of the time
the sound was the same. But a MD5 or SHA1 of the
data was different.

> Potential editing differences aside (to fit 40 tracks onto 2
> cds) I believe they will sound different even though they were ripped
> the same way.

This is, or course, a far more serious difference. I expect them
to sound different. Differing masters, "remastering" etc
are supposed to sound different.


> My source discs are original releases, remasters, various artist
> compilations and soundtracks. I have identified over 2,500 duplicates
> and would like to know if there's an objective way to determine the
> best versions (sonically speaking) without having to listen to them all
> individually.

I don't know of any, and can't imagine how you could do it.
If the remastered version has more of the room reverb, or
is compressed so that the vocals are more forward, how
can you analyze it without listening and 'seeing' which
one you think is better?

You could tell simple things, say one is 3dB down relative
to the other that has lots of 0dBfs. But how can you
tell if the difference is just caused by multiplying the
values, and thus just has the same signal and the same signal/noise
ratio? or something else?

I'm not sure you can reliably separate a few. Your ears
are going to fatigue long before you can listen to 2000 tunes.

Pat

--
Pat
http://www.pfarrell.com/music/slimserver/slimsoftware.html

zooropa320
2006-01-17, 20:40
Thanks for the input Pat. Yes, fatigue was what drove me to start this thread in hopes that there was a more reliable and consistent way to separate the wheat from the chaff. I guess I'll need to listen in small bursts to make the most accurate decisions.