IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
XLD vs. EAC cue sheet: first track pregrap
gregs1104
post Dec 16 2012, 01:57
Post #1





Group: Members
Posts: 5
Joined: 16-December 12
Member No.: 105206



I've been using Exact Audio Copy on Windows for a while now to extract audio from CDs, and I'm working on a switch to using XLD and Mac OS X instead. XLD is very good, and it seems to do everything important right. I have noticed a small difference in cue sheet formatting between the two.

Cue sheets and pregap detection are not very well defined standards. It looks like EAC treats the first track differently from XLD there. I was curious if anyone else was aware of this difference. I can't figure out if this is a problem to be fixed, or just a difference that doesn't really matter.

The most popular CD I've noticed this on is the US release of "Metallica", AKA the black album. (Napster bad!) EAC and XLD matched against AccurateRip, and their cue sheets match perfectly everywhere except the first track.

XLD 20121027 (142.1) sees the TOC start like this:

TOC of the extracted CD
Track | Start | Length | Start sector | End sector
---------------------------------------------------------
1 | 00:00:33 | 05:31:42 | 33 | 24899
2 | 05:32:00 | 05:24:43 | 24900 | 49242

It then turns the 33 second empty space at the CD start into an index position in the cue sheet:

FILE "01 Metallica - Enter Sandman.flac" WAVE
TRACK 01 AUDIO
TITLE "Enter Sandman"
INDEX 00 00:00:00
INDEX 01 00:00:33

EAC treats this differently. It considers that to be a pregap on track 1:

FILE "01 Metallica - Enter Sandman.flac" WAVE
TRACK 01 AUDIO
TITLE "Enter Sandman"
PREGAP 00:00:33
INDEX 01 00:00:00

I'm just short of knowing enough about this subject to tell if a CD burned from the two resulting cue sheets will be physically different or not. Making a perfect backup of rare discs was one of my ripping goals. It doesn't help my confidence the first CD I've tried with actual music in the track 1 pregap--Kamelot's "The Black Halo"--didn't come out of the XLD/Mac combo correctly. (I am suspicious of the drive just not handling that, the EAC rips were done on an old Plextor) Those are rare enough that I'm willing to just use EAC for them. I'd really like to switch to XLD for most of ripping though, the workflow is much simpler to automate.

Here are some other CDs I've found that have similar ~30 second track 1 pregap:
  • Van Halen - 5150
  • Marc Cohn - Marc Cohn
  • Ric Ocasek - Quick Change World

That's out of a library of 200 test rips, so this is at least 2% of my collection so far.

This post has been edited by gregs1104: Dec 16 2012, 02:00
Go to the top of the page
+Quote Post
kode54
post Dec 16 2012, 02:46
Post #2





Group: Admin
Posts: 4661
Joined: 15-December 02
Member No.: 4082



INDEX 00 implies that the pregap is included in the rip. PREGAP implies that it is digital silence and not included in the rip. The result produces the same disc if the pregap is indeed digital silence.
Go to the top of the page
+Quote Post
Pio2001
post Dec 16 2012, 16:11
Post #3


Moderator


Group: Super Moderator
Posts: 3936
Joined: 29-September 01
Member No.: 73



That's right. In this case, in this situation, XLD seems more accurate than EAC. In order to have a faithful backup with EAC, the CD should be ripped as one big file + cuesheet. This way, the gap before track one will be included.

Mind that this is not 33 seconds, but 33 sectors, which is about half a second. These sectors are not played back by CD players unless you start the playback, then rewind backwards at the beginning. The player then goes back before the start of track one and reads the gap.
Some rare CDs features a complete hidden track in a large gap, that can only be played back this way.

I said that XLD seemed to be more accurate, because there is an ambiguity about pregaps. XLD's cuesheet seems to assume that Track 0 index 0 equals sector 0. This is wrong, track 0 index 0 is in fact sector -150. But negative sectors cannot be reached with CD Burners. They can be read, forcing a large negative read offset into EAC, but not burned.
Go to the top of the page
+Quote Post
gregs1104
post Dec 16 2012, 18:34
Post #4





Group: Members
Posts: 5
Joined: 16-December 12
Member No.: 105206



Oops, I meant to say these titles have a ~30 sector pregap, my hands didn't type what my brain was thinking. A 30 second one would be a much larger issue. This seems relatively common, and I wanted to make sure it was being handled correctly. I know with EAC there's no good way to treat real music in the pre-gap without making a full-disc image. Individual tracks are so much easier to deal with in the general case that I'd like to stay with those for the bulk of extraction. As far as I can tell that doesn't cause a quality problem for most albums.

Here's what I'm seeing XLD do when there's real music in the pre-gap:

FILE "01 Kamelot - March Of Mephisto.flac" WAVE
TRACK 01 AUDIO
TITLE "March Of Mephisto"
ISRC DEA450406740
FLAGS DCP
INDEX 00 00:00:00
INDEX 01 01:20:62

I can easily write a script that finds all cue sheets that look like this, and flag them for manual review or disk image style extraction. There are so few titles like this that it's not a large problem.

While the cue sheet seems reasonable, the problem with the resulting audio on this disc is that there's a hiccup when the music goes from the pre-gap song to the track that's the normal start of the CD. I'm not sure whether that skip is from XLD or the drive in my Mac yet. I'm planning some work to try and isolate that next.
Go to the top of the page
+Quote Post
JJZolx
post Jan 28 2014, 04:29
Post #5





Group: Members
Posts: 406
Joined: 26-November 04
Member No.: 18345



I just noticed this same thing in many of my EAC CUE sheets. Is this 33 sector pregap a quirk of EAC, or is it some kind of CD mastering habit? I have almost 150 CD rips with exactly the same thing. Looking at the logs for several of them, they show 'Pre-gap length 0:00:02.33' for track 1. Other tracks also show a pregap in the log, but not in the CUE sheet.

This post has been edited by JJZolx: Jan 28 2014, 04:31
Go to the top of the page
+Quote Post
mjb2006
post Jan 28 2014, 05:59
Post #6





Group: Members
Posts: 847
Joined: 12-May 06
From: Colorado, USA
Member No.: 30694



CD mastering habit, an artifact of certain mastering hardware. Have a look at the last pie chart at http://db.cuetools.net/stats.php

This post has been edited by mjb2006: Jan 28 2014, 06:00
Go to the top of the page
+Quote Post
JJZolx
post Jan 28 2014, 06:09
Post #7





Group: Members
Posts: 406
Joined: 26-November 04
Member No.: 18345



Thanks. Given the pie chart for Agents, it could just as easily be an EAC quirk, as EAC accounts for more 91% of the submissions.
Go to the top of the page
+Quote Post
greynol
post Jan 28 2014, 06:23
Post #8





Group: Super Moderator
Posts: 10245
Joined: 1-April 04
From: San Francisco
Member No.: 13167



All the major players capable of showing this information will say the same thing.

What about a 32-frame pregap or a 37-frame pregap? You would think this quirk would be consistent.

Is it also a quirk that my CD player allows me to rewind these discs into negative time but not discs where track 1 index 1 starts at 0?

This post has been edited by greynol: Jan 28 2014, 06:23


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
JJZolx
post Jan 28 2014, 06:32
Post #9





Group: Members
Posts: 406
Joined: 26-November 04
Member No.: 18345



Makes sense.

What's the relationship between the 00:02:33 pregaps shown in the logs and the 00:33 pregaps shown in the CUE sheets? (Is that 2 seconds plus 33 frames or sectors?) Is the 2 seconds actually present as a pregap or is it just assumed by EAC? From the brief discussion above, it appears that the 33 isn't prepended to the first track, but what about the 2 seconds?

This post has been edited by JJZolx: Jan 28 2014, 06:34
Go to the top of the page
+Quote Post
greynol
post Jan 28 2014, 06:51
Post #10





Group: Super Moderator
Posts: 10245
Joined: 1-April 04
From: San Francisco
Member No.: 13167



I think Pio may have covered that. IIRC, the first 150 sectors is the lead-in which includes the TOC. After that is the beginning of the audio data which doesn't necessarily have to start with index 1.

Pragmatically speakng, in the case of a 32-frame HTOA (as an example), whether track 1 index 1 starts at 00:00:32 or 00:02:32 shouldn't be of any consequence provided you're able to burn a properly timed replica, or achieve whatever is your goal.

Was that wishy-washy enough for you? wink.gif


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 22nd November 2014 - 04:00