Skip to main content
Topic: 6 missing samples in EAC compared to CDex - Why? (Read 410 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

6 missing samples in EAC compared to CDex - Why?

I were ripping some CDs that got some errors with EAC.
So to be sure it was correct I also ripped them with CDex (paranoia mode).
Then I compared the samples (wav-files) to check.

All the files where identical except the EAC ones where missing 6 samples at 0:00
(and that the last track on the had diffences in the samples at the very end, around 600 samples).

My drive is an Asus BD/DVD/CD writer, and Accurate rip has it set for a +6 offset.
Is that just a coincidence, or does that +6 have anything to do with the 6 missing samples?

And why the 600 differnt samples at the end of the last track of the CD.

PS!  Tested with several other CDs as well and go the same missing 6 samples on all (otherwise files are equal).
Lars-Erik Østerud

Re: 6 missing samples in EAC compared to CDex - Why?

Reply #1
http://www.dbpoweramp.com/spoons-audio-guide-cd-ripping.htm
http://wiki.hydrogenaud.io/index.php?title=AccurateRip

If the EAC rip has read offset correction applied and the CDex rip has no correction then the rips are starting at different positions. The first link has a graphic to illustrate.

The second issue can occur when Overread into Lead-In and Lead-Out is enabled in EAC for a drive that has positive (+) read offset correction and does not support Overread into Lead-Out.
korth

Re: 6 missing samples in EAC compared to CDex - Why?

Reply #2
My drive is listed with +6 as offset. What does that really mean?
Does it start to read 6 samples to early so that the 6 first is skipped?
CDex offeset is 0 and that wav-files has 6 more samples at the start.
Will the wav-files produced with CDex then have 6 samples BEFORE the audio track really start then?
Should I add an offset of 6 to the CDex settings then to make it read from the right start position?

The last track is strange anyhow. One thing is the difference in data at the very end of the track.
But in EAC I very often get a "sync error" on the last track, and it cannot be verified, and also get a low %.
But when compared to the rip from CDex they are still equal except a few samples at the very end.
And there is no glitches in the audio when I listen to it.
I use a program to strip of silence, and on theses last tracks it always strips of quite a bit silence at the end.
Lars-Erik Østerud

Re: 6 missing samples in EAC compared to CDex - Why?

Reply #3
*The following assumes the goal is an AccurateRip verifiable lossless rip of all tracks using a drive with +6 read offset correction. Some of the information below is rephrased from the links or response I've already provided above but should cover the additional questions as well. This stuff has been covered repeatedly so please SEARCH for answers within the forums before posting questions.

A drive that requires a +6 read offset correction actually reads 6 samples (24 bytes) ahead of (earlier than) the position used by the adopted reference. With +6 read offset correction applied, the software essentially ignores the first 6 samples but does read the total number of complete frames for each track (588 samples per frame (aka CD sector), 75 frames per second).

When the drive physically reaches what would be the start of the 2nd track (actual position without correction) the software continues to read for an additional 6 samples to complete the frame. Track 2 is read starting from that point forward until what would be the physical start of the 3rd track plus an additional 6 samples and so on.

When the drive physically reaches the end of the last track (most drives with positive (+) read offset correction are not capable of reading beyond that point and cannot over-read into the Lead-Out area) the rip ends 6 samples short of the total samples required to complete the last frame of the last track. The properly configured software will pad 6 samples of silence to the end of the rip so that the last track has the correct number of samples to complete the last frame. Not to worry, the missing 6 samples were probably inaudible anyway.

If you have that rare drive that is capable of reading beyond the end of of the last track (can over-read into the Lead-Out area), EAC can be configured to allow over-reading and the rip will end after the final 6 samples are read. However, if EAC is configured to allow over-reading and the drive is not capable, the rip may show errors at the end of the last track and non-null samples in the last chunk of data read may be replaced by null ones (which can prevent the last track from verifying as accurate in the database(s)).

Quote
Should I add an offset of 6 to the CDex settings then to make it read from the right start position?
To use the AccurateRip plugin in EAC, read offset correction needs to be applied. AFAIK CDex doesn't have AccurateRip verification but a rip that wasn't read offset corrected can still be verified by other programs (assuming no samples are missing and you have additional info such as track 1 pre-gap or CD-extra data track length if part of the original TOC). However, if you plan to use both EAC and CDex with the same drive, it would probably be useful to configure both programs the same way.
korth

Re: 6 missing samples in EAC compared to CDex - Why?

Reply #4
Very good explaination, and reflects what I'm seeing.

When I setup CDex with an offset of 6 the files where equal to those from EAC.
I got a "File is longer 0:00:000" on the EAC files, but no compare errors.

On the last track on discs I get a couple of different samples at the very end on the CDex and EAC files.
But if that is in the "over-read" area this values could be random or "grabage" and then of course they would be different.
Anyway I always trim the files with "mpTrim" and the last track always have lot of extra silence at the end.
Lars-Erik Østerud

 
SimplePortal 1.0.0 RC1 © 2008-2020