Pink Floyd - Dark Side of the Moon - Analysis

I keep wishing that SHF had a Like button like Facebook!

 

Thanks for all the work and explanation. I really enjoyed this.

 

I also want to thank you for posting this data. It's very interesting to see.

So far i have purchased only one of the newly remastered discs - Obscured By Clouds (my fave by Pink Floyd) and will purchase all of them in the next few months.

 

So, what does all this have to do with the Beatles?

JcS (who has NO idea what most of this means, but can tell it took a lot of work.)

 

The Beatles' 1987 & 2009 remasters could benefit from similar scrutiny, me thinks ...

 

foobar - Really great idea of yours using the exported spectrum data! I didn't know you could do such a thing...

...so to take your idea for a spin, I compared two different versions of a-ha's "Take On Me": one from Razor & Tie's 2-CD Totally '80s (1993, mastered by our host) and one from Warner Special Products ' 2-CD Night Beat (1988). I've suspected that Steve's Razor & Tie compilations were based on masterings from earlier, existing CDs. I'd even used the language that Totally '80s "uses the same analog transfer" as Night Beat. Well, the data looks even closer than that!

I can now say that for "Take On Me", Totally '80s (1993) sounds pretty much exactly like Night Beat (1988), only about 1 dB louder.

Your graphs and plots look so much better than mine - any tips for spectral analysis settings (I used Audacity), plot settings (I used Excel), or graphics settings (I did a cut/paste into some Paint program and exported a png file)? The actual spectral analysis is really easy, but turning the plots into a pretty picture is a pain.

 

Great work, very illustrative.

So the 2011 Redbook is just a downsampled version of some hirez source used for the BluRay 96/24. I'll bet dollars to donuts that you'll hear people claiming the clearly audible superiority of the BluRay though! I've heard both and they do sound identical to me. It is nice to see the 96 kHz light on the DAC but if you blindfolded me there is no way I could tell the difference. Since I never bought the Sax master I can't compare, but it looks like they are on par with each other. Regardless I feel this mastering is about as good as it can get - I doubt you can squeeze any more fidelity out of that nearly 40 year old ribbon of rust.

 

I'm glad you appreciated that adjustment. However, I am at total loss as to why you think is was necessary? I can see the same information on both charts just fine.

 

Just like I said, I wanna be able to easily compare tracks equalization only without gain differences.

 

I can tell the difference even between the CD and the LPCM 48kHz stereo track on the DVD. It's the added 24bit depth in dynamic range that makes the difference.

But I agree that they'll probably never top this mastering.

 

Not to put any pressure on ya foobar but could you do "Wish You Were Here" next?

Very cool stuff! I have always loved the original 1983 mastering of DSOTM, but I must now make time for an unbiased comparison with the TO in the near future (I have all versions minus the 2011). Also, for those on here who like to give Dave a tough time (in fun of course), I am pretty sure his preference has always been for the TO. This thread validates what he's said all along.

 

Couple things.

First off, the TT DR Meter score is not from EAC, neither are the ReplayGain Values. The TT DR Meter is a stand-alone app, the ReplayGain numbers are from foobar2000.

Second, volume is not compression. The 2011 Guthrie redbook has a total of two clips two samples or longer. That means he used up the full dynamic range available on the CD. That is not a bad thing. Even though it makes the waveform look larger.

Anyway, I don't fully understand what goes into a TT DR "score" I just thought I'd include those numbers for completeness. In any case, I just scanned the folder again to double check, and those are the correct numbers. Here is the full TT DR Meter report:

Code:

 ----------------------------------------------------------------------------------------------
 DR   	   Peak     	   RMS      	 Filename
----------------------------------------------------------------------------------------------

 DR11     -4.96 dB   -19.34 dB   Money - 1983 Sony.wav
 DR11      over      -13.02 dB   Money - 1984 EMI.wav
 DR11     -2.79 dB   -17.45 dB   Money - 1985 Capitol.wav
 DR10     -4.98 dB   -18.05 dB   Money - 1988 MFSL.wav
 DR10     -0.28 dB   -13.48 dB   Money - 1993 Sax.wav
 DR8       over      -10.69 dB   Money - 2003 Guthrie-RedBook.wav
 DR9      -0.05 dB   -12.66 dB   Money - 2003 Guthrie-SACD.wav
 DR11     -0.34 dB   -14.16 dB   Money - 2011 Guthrie-BluRay.wav
 DR10      over      -12.41 dB   Money - 2011 Guthrie-RedBook.wav

----------------------------------------------------------------------------------------------
 Number of files:    8
============================================================================================== 

(note that this is for my five minute clips)

 

I did do charts for three or four different Beatles albums in 2009. They were in each album's remaster thread.

 

TT DR, I believe, measures average difference between the track's current peak & RMS values (in dB).

 

You may view it as an odd anti-social quirk of mine, but I really don't care to tell people what to like. Now that you know what exists, try a few and let your ears decide. None of these are terribly expensive, if you don't care which pressing to get.

 

I used a Blackmann-Harris window and an 8192 FFT size. Make sure you highlight the whole clip, and press scan. Otherwise you're just getting the spectral balance at the cursor.

Getting pretty graphics out of Excel is a royal pain. I use a VB script called "Export Chart" to export a png. Still, it's like trying to play golf from orbit.

 

Nice work, Foo!

 

Like I said, I did it already, and don't think I'll re-visit it until the Immersion comes out.

Dave can certainly like whatever he wants. And change his mind whenever he wants too. I think the one he likes though is the non-TO version, that would be what I call the 1983 Sony. (Dave can of course correct me if I'm wrong.) The TO version is what I call the 1984 EMI. The TO means it was made by Toshiba, and not Sony, and the Toshiba made discs had the same mastering that the first West German CDs did.

It's confusing however. Japan the US and Europe all had both these very different masterings available concurrently. There are even some TO discs made by Toshiba with the Sony mastering. It's difficult to know just which of the two masterings you have.

One thing Dave likes to do is compare different pressings of the same data, which of course all this analysis completely ignores. So, he may have additional sound-quality comments that these measurements don't show.

 

Call me Bar, please.

 

That's what I mean foobar, after Immersion comes out!

 

I have every mastering available as US versions too (I ran them all through EAC and even stuck a small cue sheet in each case so I can keep up with the different versions) so yes it can be confusing!

I remember being surprised by Dave's choice of his preferred DSOTM because it wasn't a first pressing from Japan. That's why it stuck out in my mind, but maybe I was dreaming . I'm sure he'll be here soon to let us know!

 

Maybe I'm wrong then, or he changed his mind.

 

It's probably my memory that's failing

Seriously though foobar, this is a great thread, thanks so much

 

http://stevehoffman.tv/forums/showpost.php?p=5227173&postcount=9

CP35-3017 with "no TO in the matrix" would be what I call "1983 Sony".

However, copies with or without CDP 7 in the matrix are digitally identical, so they would both be the same in my analysis.

Of course Dave is free to like what he wants, or to change his mind since the post I linked.

 

You were correct, I was wrong. Dave just confirmed it in a PM to me. Sorry for the confusion!

At any rate, I am gonna listen critically to the TO tomorrow! Thanks again foobar for the great info!