DSOTM Black Triangle or Japan Harvest=Same Mastering?

Just what the name says.
User avatar
lukpac
Top Dog and Sellout
Posts: 4591
Joined: Wed Apr 02, 2003 11:51 pm
Location: Madison, WI
Contact:

Postby lukpac » Thu Mar 24, 2005 10:35 pm

I'll have to compare the MFSL to the "old" stock Capitol again. I didn't notice much difference the last time I compared them...
"I know because it is impossible for a tape to hold the compression levels of these treble boosted MFSL's like Something/Anything. The metal particulate on the tape would shatter and all you'd hear is distortion if even that." - VD

User avatar
Yesman
Posts: 9
Joined: Thu May 06, 2004 5:21 pm
What color are leaves?: Green
Spam?: No

Postby Yesman » Sun Apr 10, 2005 5:41 pm

krabapple wrote:Yes, DSotM is an instance where the CD layer is different from the SACD layer.


Really? Can you hear a difference?

http://www.stevehoffman.tv/forums/showp ... stcount=66

User avatar
krabapple
Posts: 1615
Joined: Wed Apr 16, 2003 4:19 pm

Postby krabapple » Mon Apr 11, 2005 3:07 pm

I don't know; I haven't ABX's them. But the measured difference [url=http://www.stereophile.com/news/11649/]
demonstrated[/url] for 'Money' is such that I would not be terribly surprised if someone positively ABXed them, even after level-matching. Doesn't mean *everyone* could, of course...inlcuding me. Atkinson errs in implying that multiple instances --even hundred -- of clipped peaks *must* be 'fatiguing' of themselves -- when AIUI it's more likely they're audible when they're runs of *consecutive* peaks over 0 db FS. IIRC Cool Edit reports a 'possibly clipped' signal when two consecutive samples are deemed likely 'overs', but I'm not sure two-sample overs are *necessarily* audible. I would expect the level difference that Atkinson shows existing even after peak-matching, to be in the range of potentially audible, though. I intend to get around to transferring all of DSotM from SACD--> PCM eventually, and after I do I'll give it a shot.

Btw, was your comparison blind? What were the sucess/failure difference stats?
"I recommend that you delete the Rancid Snakepit" - Grant

User avatar
lukpac
Top Dog and Sellout
Posts: 4591
Joined: Wed Apr 02, 2003 11:51 pm
Location: Madison, WI
Contact:

Postby lukpac » Mon Apr 11, 2005 3:41 pm

krabapple wrote:I don't know; I haven't ABX's them. But the measured difference [url=http://www.stereophile.com/news/11649/]
demonstrated[/url] for 'Money' is such that I would not be terribly surprised if someone positively ABXed them, even after level-matching.


Rather than compare a rip to a 96/24 dub, shouldn't he have done 96/24 dubs of both?
"I know because it is impossible for a tape to hold the compression levels of these treble boosted MFSL's like Something/Anything. The metal particulate on the tape would shatter and all you'd hear is distortion if even that." - VD

User avatar
krabapple
Posts: 1615
Joined: Wed Apr 16, 2003 4:19 pm

Postby krabapple » Mon Apr 11, 2005 4:09 pm

Yes, he should show that his A-D dubbing itself doesn't do anything too funky to the files. But it seems unlikely to me. In my rather less elaborate setup, the Cool Edit stats for the Rolling Stones CD vs SACD weren't very different at all. I would be very surprised if his setup and methods introduced the sort of peak limiting he saw in his SACD transfers, not to mention the apparent average level differences betweenn CD vs DSD. He claimed to have monitored the transfer levels such that it would not clip during digitization , then normalized the file to O dB. It's interesting to note therefore that the left channel 'flattops' at a bit less then 0 dB, while the flattops in the right channel peak at 0. I *hope* his channels aren't imbalanced somewhere along the line, thought that still wouldnt' account for the flattopping observed, nor the overall level differences...


Atkinson expolain it thus:

The right channel in the second solo bangs up hard against 0dBFS. This is where the 362 clipped samples reported by Cool Edit Pro come from. Fig.5 therefore shows a very short example from this solo. Yes, the transients in the right channel are flat-topped, suggesting digital clipping. But so are the transients in the left channel, and as this occurs about 0.5dB below 0dBFS, it can't be clipping in the digital domain. It looks as if the mastering engineer who did the transfer for the CD layer ran the analog tape machine into an aggressive peak limiter. He didn't do so for the SACD layer, however, as there is no evidence of flat-topping of the waveform in the SACD version of "Money."


But there's another explanation -- that *after* digital peak limiting, the overall level of the left track was brought down a little (it's still the overall louder of the two DSD channels, according to the stats). OR, his own transfer rig has a channel imbalance. Might have been good to show the transfer 'native', that is, before it has been normalized....
Last edited by krabapple on Tue Apr 19, 2005 11:54 am, edited 1 time in total.
"I recommend that you delete the Rancid Snakepit" - Grant

User avatar
krabapple
Posts: 1615
Joined: Wed Apr 16, 2003 4:19 pm

Postby krabapple » Mon Apr 11, 2005 4:37 pm

And btw, James, since the SHtv folk seem to be floundering around trying to locate the link to the DSotM measurements, and I'm currently persona non grata there...perhaps you would repost the link I provided to the Stereophile article, ot SHtv.

In the meantime we await the results of Michael St. Clair's jitter comparions of UDI and UDII discs....upon which all the Nightgorf's hopes must rest...


:roll:
"I recommend that you delete the Rancid Snakepit" - Grant

User avatar
Yesman
Posts: 9
Joined: Thu May 06, 2004 5:21 pm
What color are leaves?: Green
Spam?: No

Postby Yesman » Mon Apr 11, 2005 8:10 pm

lukpac wrote:Rather than compare a rip to a 96/24 dub, shouldn't he have done 96/24 dubs of both?


That's where this test is flawed. We don't know what that D/A A/D conversion is doing to the DSD signal.

Aslo, he recorded the SACD layer around -3dB. How'd he lower the signal? Did he lower it in the analog domain or did he hit the A/D converter at full volume, lower the signal inside the recorder and then normalize back up to 0? That kind of tnkering should've been done to both sample for an actual accurate test.

devhda
Posts: 1
Joined: Mon Apr 18, 2005 11:37 pm

Postby devhda » Mon Apr 18, 2005 11:42 pm

May I suggest you to post your EAC logs in a thread I just opened at Steve Hoffmanforum

User avatar
Xenu
Sellout
Posts: 2209
Joined: Mon Apr 07, 2003 8:15 pm

Postby Xenu » Tue Apr 19, 2005 12:08 am

Problematically, those EAC logs don't necessarily take into account the phenomenon of "same master, different volume." IE EAC logs would report that the version of Julia on the Lennon "Imagine" soundtrack was different from the version on the White Album, where in reality it's the same master with the volume boosted slightly. EAC will *not* tell you about masters derived from *each other*.
-------------
"Fuckin' Koreans" - Reno 911

User avatar
Crummy Old Label Avatar
Posts: 1226
Joined: Wed Aug 25, 2004 5:55 pm
Location: Out of my fucking mind

Postby Crummy Old Label Avatar » Tue Apr 19, 2005 9:45 am

Xenu wrote:Problematically, those EAC logs don't necessarily take into account the phenomenon of "same master, different volume." IE EAC logs would report that the version of Julia on the Lennon "Imagine" soundtrack was different from the version on the White Album, where in reality it's the same master with the volume boosted slightly. EAC will *not* tell you about masters derived from *each other*.


Why does no one ever look at the ISRC codes? EAC will report these on the CUE sheet if you enable the feature under "EAC Options." Shouldn't comparing the ISRC number of the same track from different discs tell you precisely whether or not the track is indeed a different mastering?

In theory the ISRC should tell you what you need to know, but I have no idea how it works out in actual practice. Do the disc manufacturers actually keep up with this? (I've never sat down and compared ISRC codes.) According to the ISRC Handbook, the rule for remastered music appears ambiguous:

4.9.10 Re-mastering

When a track is re-mastered for the purpose of reproduction on a new carrier without restoration of sound quality (also see Section 4.9.1 Re-mixes/ Edits / Takes), then no new ISRC is required.

It is nevertheless the Registrant’s responsibility to decide where to draw the line between sound restoration (full re-mastering) and simple re-mastering.


["Full" remastering vs. "simple" remastering. Go figure.]

Someone who has several international CD versions of Dark Side of the Moon ought to compare the ISRC results, just to see what that reveals. Most indie releases don't utilize ISRC codes (you have to pay for the privilege, after all), but my guess is fairly certain that all Pink Floyd CDs use them.
If you love Hi-REZ TAPE HISS, you're REALLY going to love Stereo Central

User avatar
krabapple
Posts: 1615
Joined: Wed Apr 16, 2003 4:19 pm

Postby krabapple » Tue Apr 19, 2005 11:52 am

Yesman wrote:
lukpac wrote:Rather than compare a rip to a 96/24 dub, shouldn't he have done 96/24 dubs of both?


That's where this test is flawed. We don't know what that D/A A/D conversion is doing to the DSD signal.


True, but I've done conversions on rather less expensive equiupment, and I see no evidence that it *introduces* compression or clipping.

Aslo, he recorded the SACD layer around -3dB. How'd he lower the signal? Did he lower it in the analog domain or did he hit the A/D converter at full volume, lower the signal inside the recorder and then normalize back up to 0? That kind of tnkering should've been done to both sample for an actual accurate test.


True, he should have showed what an A/D conversion of the CD layer analog output looked like too.

When I do my comparisons, I lower the volume after D/A conversion in the player, in the analog domain, using an in-line preamp that feeds the soundcard.

Still, I'd be surprised if any of these procedures made the compression/limiting seen in the CD layer 'go away'. (Or make it appear in the SACD layer.)
"I recommend that you delete the Rancid Snakepit" - Grant

User avatar
krabapple
Posts: 1615
Joined: Wed Apr 16, 2003 4:19 pm

Postby krabapple » Tue Apr 19, 2005 11:59 am

Too good not to share (from the alt.music.pink-floyd newsgroup today; hasn't shown up on Google yet) .
I know 'Gil' but AFAIK he doesn't know anything about FLO...yet.

--


Tue, 19 Apr 2005 11:59:36 alt.music.pink-floyd Thread 8 of 8
Lines 21 Re: PROJECT: Document every copy of Pink Floyd's Dark Side of the Moon CDs RespNo 2 of 2
Gil Ulrik <rushomancy@gmail.com> at the John Weber Memorial Organization: Line

Newsgroups: alt.music.pink-floyd

devhdadevhda@gmail.com come on down:

>Go here to see the Project results and to post your EAC logs:
>
>http://www.stevehoffman.tv/forums/showthread.php?p=1036372
>
>
>The Mission
>
>Objective #1: To find and document every single copy of Pink Floyd's
>Dark Side of the Moon.
>Objective #2: To find the best sounding master, AKA The Holy Grial.

Boy, that's incredibly fucking stupid. Then again, I guess it is
stevehoffman.tv; a bigger group of self-obsessed audiophile wankers you
couldn't ask for.


--
"Margerate Thare (can't spel tonite) PM of Britian at the time. Considered
conserative."- Plasyd offers profound insight into _The Final Cut_
"I recommend that you delete the Rancid Snakepit" - Grant

User avatar
krabapple
Posts: 1615
Joined: Wed Apr 16, 2003 4:19 pm

Postby krabapple » Tue Apr 19, 2005 12:03 pm

Xenu wrote:Problematically, those EAC logs don't necessarily take into account the phenomenon of "same master, different volume." IE EAC logs would report that the version of Julia on the Lennon "Imagine" soundtrack was different from the version on the White Album, where in reality it's the same master with the volume boosted slightly. EAC will *not* tell you about masters derived from *each other*.


However, if you normalize them (assuming it's a simple level boost, and not also compressed/limited) then they should match virtually exactly afterwards in their peak/avg statistics.

And, too, one could compare the frequency profiles after normalizing both...I've alredy seen cases where the *only* difference turns out to be overall level (IOW, 0 difference at all frequencies from 20 - 20 kHz) ...which is fantastically unlikely if in fact two different masters have been used.
"I recommend that you delete the Rancid Snakepit" - Grant

User avatar
Xenu
Sellout
Posts: 2209
Joined: Mon Apr 07, 2003 8:15 pm

Postby Xenu » Tue Apr 19, 2005 12:51 pm

Or you could forego EAC logs at all and just compare digitally in Audition one track per disc. I don't think it would take that much longer.
-------------

"Fuckin' Koreans" - Reno 911

User avatar
krabapple
Posts: 1615
Joined: Wed Apr 16, 2003 4:19 pm

Postby krabapple » Tue Apr 19, 2005 12:57 pm

Me neither. But that ISRC tack is an interesting idea.
"I recommend that you delete the Rancid Snakepit" - Grant