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: Remasters vs. original issues (Read 44048 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Remasters vs. original issues

Reply #25 –
Well, I'm in the middle of a rip/encode/compare right now (borrowed a copy of the first album's remastered CD) and I've already found some interesting things:

1) The 1980s original issue was NOT normalized to 0 dB!  I opened the CD audio file in an editor and confirmed that the peak level was somewhere around -2 dB.  Weird.

2) The phase was inverted on the new issue compared to the original.

Once I corrected this (since I already had to normalize the 80s release, I went ahead and flipped the phase as well, but I have no idea which one is "correct") a quick A/B sounded very similar.  I'm going to go ahead and ABX the lossless as well as the lossy files, but when I get to the lossy files - will the phase reversal affect the encoding?  And does iTunes normalize when encoding?

Mods, should I start a new thread since this is diverging from the generic nature of the current one?

Remasters vs. original issues

Reply #26 –
Then there are the few who take maters into their own hands, such as Elbow's The Seldom Seen Kid. They mixed and mastered it themselves and it sounds great!

Remasters vs. original issues

Reply #27 –
Quote
What ticked me off was that the remasters for "Chicago 16" and "Chicago 17" replaced album cuts with radio edits. Fortunately, I still had the original issues. However, as with most reissues, bonus tracks were included on both re-releases. Since the remastered versions were noticeably louder than the original release, I ultimately burned myself a ReplayGain adjusted disc, inserting each bonus track at the end of its respective album.


I have an original copy of the 1984 pressing of Chicago 17. It's not only recorded well (many 80's record were) the mastering on it is nothing short of perfect. It really stinks that they went back remastered it and probably butchered it. 
budding I.T professional

Remasters vs. original issues

Reply #28 –
Quote from: Chjmu on
Then there are the few who take maters into their own hands, such as Elbow's The Seldom Seen Kid. They mixed and mastered it themselves and it sounds great!


and the funny part is, it's a relatively loud album by today's standards.  I bought it because it had the "Turn me up" logo in the liner notes, and when I ripped it and ran it through MP3 gain and Audacity, it made me wonder how it could be considered a dynamic album.  It does sound pretty fantastic though.

Another recent album I think (for the most part) sounds great is In Rainbows.  I think these are two examples of how albums can be somewhat loud and still have some dynamics to them.

Remasters vs. original issues

Reply #29 –
I haven't gotten all the data together yet, but the examination of Chicago remasters continues.  The first album definitely looks like the signal is phase-inverted and normalized vs. the original issue.  The second album looks nearly identical in waveform view between the editions.  I haven't done a complete ABX yet due to time constraints (maybe this weekend), but both remasters seem to be less aggressively filtered; i.e., I can detect tape hiss on the remasters, and I think the high end is correspondingly extended.  We'll see whether I can pick the X out every time. 

The real disappointments are the DTS mastering of II and the reissue of V.  I found a couple of reviews stating that the people who remastered II got the original multitracks and worked wonders with them.  Unfortunately, just from examining the waveform it's apparent that they also compressed the living h*ll out of them before authoring the discs.      It does sound like they got a cleaner transfer, but it's so "in-your-face" that a lot of the improvement is negated IMHO.  I bet I'll be able to pick the DTS copy out every time, and not just because of the 2-track 16-bit downmix.  V was treated similarly; if you zoom in on the waveform, there are lots of clipped peaks and the overall level looks much higher.  I think I'll be able to distinguish remaster from original here, but the remaster doesn't really sound improved to me.  The volume compression will probably be the smoking gun.

I'm posting from my Mac, so I may have to go back to the utility PC to get replaygain values and my ABX test results; I can't find a program on OS X that just displays replaygain without trying to change the data.    Also, the Mac version of abchr has been cantankerous and seems to require a lot of additional utilities to be installed before it'll recognize AIFF and M4A files.  I'm working on capturing the waveforms in the meantime.  Very enlightening...

Remasters vs. original issues

Reply #30 –
Yay, I was able to get abchr to work on my Mac, though I had to use WAV files only.  For some reason it can't recognize my faad install.  No biggie, I feel better comparing the source data anyway and I had the CDs to hand.

In short - I picked two tracks from Chicago II and ABX'ed the old vs. the remaster, then ABX'ed the remaster against the DTS 2-channel.  To my slight surprise, I scored 100% on all the comparisons; as I expected, the test does obliterate your biases.  The differences must really be significant for me to score so well.      The old vs. new were primarily distinguishable because of a very slight apparent improvement in clarity; I suspect it was due to more high end, or maybe the noise reduction was more carefully applied.  Cymbals and toms helped a lot, so I paid attention to them even more than the brass. 

Then to see if I was deluding myself, I picked 3 or 4 subsections of each track and ran an ABX test on each clip, so I probably was harder on myself than if I'd focused on the same part of each track for every test sequence.  This actually made it easier for me to identify the "sound" of each track, though.  The old vs. new remasters were very close.  The DTS mix just jumped out at me because of the louder bass - apparently they cranked up the bass band a lot when they compressed the tracks.  It seemed marginally cleaner, though, even in the clips with little or no bass.  Abchr did a good job of normalizing the apparent volumes, because the visually rather dramatic compression wasn't too apparent during listening.  I actually preferred the DTS mix for one track, because the brass sounded less muddy or more distinct.

So I learned a lot; I guess my 55-year-old ears aren't entirely gone yet.  On the iPod where I listen to music most of the time, I will probably pick the DTS copy because it will punch through the background noise a little better and it's not an unpleasant sounding mix.  On the big stereo, though, it'll probably be the 2002 remaster.

Next I'm going to try the CTA CDs and report on them.  And I'll try to get the waveform pictures uploaded as well.

Remasters vs. original issues

Reply #31 –
Here's the original 1980s Chicago II, track 1, beginning:

Remasters vs. original issues

Reply #32 –
OK, this time the first track from the CTA (first) album.  Comparing the very beginning of the track in waveform view, first the original CD issue:



And then the 2002 remaster:



Note the slight level normalization and phase reversal.  Again I managed 100% on the ABX test, but surprisingly, I think I like the sound of the original release slightly better.  The cymbals are a bit clearer; on the remaster I seem to hear a sort of distortion on cymbal crashes that easily let me choose the correct file.

Remasters vs. original issues

Reply #33 –
With CD I go with the original releases purely because it hasn't been messed with, is way cheaper as it is second hand, and you know what you are getting.

The term remastering is so tainted nowadays that it is nothing more than a big BEWARE sticker.

I usually look for the vinyl as well as that is even cheaper and solves all these unkowns and the bonus is that it gives you the sheer physical prescence of the artwork and liner notes.

Remasters vs. original issues

Reply #34 –
Quote from: mattdm11 on
Quote from: Chjmu on

Then there are the few who take maters into their own hands, such as Elbow's The Seldom Seen Kid. They mixed and mastered it themselves and it sounds great!


and the funny part is, it's a relatively loud album by today's standards.  I bought it because it had the "Turn me up" logo in the liner notes, and when I ripped it and ran it through MP3 gain and Audacity, it made me wonder how it could be considered a dynamic album.  It does sound pretty fantastic though.

Another recent album I think (for the most part) sounds great is In Rainbows.  I think these are two examples of how albums can be somewhat loud and still have some dynamics to them.


Yeah, I recently got the debut album by the Maccabees. I could hear the drums and guitars very clearly throughout the album, but surprisingly, RG was -10.18!

Another album is Konk by the Kooks. Has RG of -9.xx but it's still dynamic.
 

Remasters vs. original issues

Reply #35 –
Quote from: Chjmu on
I generally prefer the originals, unless the remastered is remixed or something (such as Killing is my Business)

Thumbs up for a being a Megadeth fan! Regards

Remasters vs. original issues

Reply #36 –
Quote from: mattdm11 on
Quote from: Chjmu on
Then there are the few who take maters into their own hands, such as Elbow's The Seldom Seen Kid. They mixed and mastered it themselves and it sounds great!

and the funny part is, it's a relatively loud album by today's standards.  I bought it because it had the "Turn me up" logo in the liner notes, and when I ripped it and ran it through MP3 gain and Audacity, it made me wonder how it could be considered a dynamic album.  It does sound pretty fantastic though.

I just got my copy (for a bargain 6.49 € from Play.com) and I spotted the "Turn me up" text (no logo!) in the booklet. The whole text is as follows:

Turn Me Up!™
To preserve the excitement, emotion and dynamics of the original performances this record is intentionally quieter than some. For full enjoyment simply Turn Me Up! (TurnMeUp.org)

The album RG is -9.71dB. Loudest track is -11.16dB. Quieter than some? Quieter than Oasis? Yes, slightly. But it's very loud! But sounds ok (quickly listened).

I wonder couple of things:

1) Can an album with that high RG values have dynamics? (sorry if this a stupid question)
2) Has the "Turn Me Up" certified label text have poor choice of words? If dynamic != quiet ??

Remasters vs. original issues

Reply #37 –
It would seem stupid a TurnMeUp logo on a Oasis-esque "ReplayGain value album". All Oasis albums pratically hit the -12.00dB value. For one thing I know, if the distance of the loud peaks from the quiet peaks are *big* then perhaps it would be a loud dynamic album.

Can you please share the waveform graphs with us?

Remasters vs. original issues

Reply #38 –
Quote from: Akkurat on
I wonder couple of things:

1) Can an album with that high RG values have dynamics? (sorry if this a stupid question)
2) Has the "Turn Me Up" certified label text have poor choice of words? If dynamic != quiet ??

Just for fun I decided to remaster some quiet music, namely Mahler's Urlicht from his 2nd Symphony.

The original has a peak value of -16.3 dB, so the 2nd version is just the original increased by 16.3 dB (ringed in red), for the sake of comparison.

I compressed and limited the hell out of it and got it to a Replay Gain of -11.13 dB from +11.30 dB. Yet it's still dynamic (not everything is rammed against the ceiling), it's just not as dynamic as it should be (and it sounds like shit).



To hear this "remaster" (atrocity) the 3 files (30 secs FLACs zipped) are here.

IMO for this Turn Me Up initiative to be of any value there should be a RG limit. Something like -7 dB. Anything over that, you don't need to Turn It Up.

C.
PC = TAK + LossyWAV  ::  Portable = Opus (130)

Remasters vs. original issues

Reply #39 –
Quote from: Neasden on
Can you please share the waveform graphs with us?

Here are the highest and lowest RG value tracks (not counting the bonus track which has +7.53 dB RG):



Lots of clipping. Could somebody with more knowledge analyse this image?

What I understand, the higher the RG value, the less dynamics, right? Is it this straightforward?

I wonder why this Elbow album has the Turn Me Up text. Is it completely bollocks with such high RG values and clipping?

Remasters vs. original issues

Reply #40 –
Quote from: Akkurat on
What I understand, the higher the RG value, the less dynamics, right? Is it this straightforward?

No. That was the point of my post above. The first and second versions are identical in their dynamics - the 2nd one has been maximised according to its peak (as ringed in red). Yet the difference in replay gain values is huge:

The original Replay Gain = +11.30 dB
The one with max vol according to peak has a Replay Gain of -5.00 dB.

Yet they have identical dynamics.

However to get to the kind of loudness wars RG range (-11, -12), one has to smash the music (limiting, DR compression etc. --- see the 3rd version - and that's where the dynamics suffer).

C.

[EDIT: last paragraph, minor addition]
PC = TAK + LossyWAV  ::  Portable = Opus (130)

Remasters vs. original issues

Reply #41 –
Quote from: carpman on
Quote from: Akkurat on
What I understand, the higher the RG value, the less dynamics, right? Is it this straightforward?
No. That was the point of my post above.

Sorry, I was actually thinking only high RG values (beyond something like -5 dB or your mentioned -7 dB level). I listened to your samples and the 3rd one was really bad, poorly "remastered".

Remasters vs. original issues

Reply #42 –
Was there a general consensus on the recent Depeche Mode remasters (good vs. bad)?  I have only Black Celebration so far, but it was probably the "cloudiest" sounding original master, anyway.  Do the new ones suffer from the same maxed out compression as "Playing the Angel"?

Remasters vs. original issues

Reply #43 –
I forgot this topic. I see that nobody analyzed the waveforms nor answered my questions (post #40). Experts needed.  Thanks.

Remasters vs. original issues

Reply #44 –
Quote from: Akkurat on
What I understand, the higher the RG value, the less dynamics, right?
This is often the case, but not necessarily true.

Quote from: Akkurat on
Is it this straightforward?
No, although the RG algorithm  is pretty straightforward, it does not directly measure dynamics.
http://replaygain.hydrogenaudio.org/calculating_rg.html

Quote from: Akkurat on
Is it completely bollocks with such high RG values and clipping?
Does it sound bad to you?  To me, that's all that matters.

Remasters vs. original issues

Reply #45 –
Quote from: greynol on
Does it sound bad to you?  To me, that's all that matters.

Also to me. It sounds ok.. I guess, no listening fatigue, hard to say, I can't instantly say that it's ok, something is bugging me with the sound, don't know what.. so I guess I've to say it's ok. One track (#7) has some distortions.. is it intended, dunno.

You cropped my question, the full question was: "I wonder why this Elbow album has the Turn Me Up text. Is it completely bollocks with such high RG values and clipping?" I questioned the "Turn Me Up" initiative, as to what grounds this record was "certified" (was it really or did the Elbow guys (or their record company) just stick the "certificate" to the booklet without requesting an audition?). The "certification system" is not ready yet, from the "Progress Report":

"1. Determining an objective method of measuring the dynamics of a record.
We're currently considering a measurement that will be precise, scientific, and 100% repeatable. It will not be a subjective or interpretative measurement in any way."

The album RG is -9.71dB. Loudest track is -11.16dB. And if I've understood correctly, with such high RG values, one can say that there's hardly any dynamics. Or am I talking utter rubbish here?

I also wanted somebody to analyze the waveforms; could the 2nd track be counted as dynamic and what about the clipping?

Thanks for humoring a dunce like me.

Remasters vs. original issues

Reply #47 –
Quote from: greynol on
RG was intended to measure loudness, not dynamics.

Yes, I learned that already, but:

(very) high RG = loud = loudness war = limiting/compressing/? = not much dynamics  ??