Problem with CD extraction

[liveman]

Hello to all,
today, I have tried to extract a song from a CD-R I received in a trade. The CD-R brand is Verbatim. I have two drives: 1. my burner 2. my cd/dvd-rom reader. Both have offset corrected with EAC.

I extracted the song (this is track 1 by the way ) with the two drives and the song gives two result:

with my burner (my offset is really 0):

Used drive : HL-DT-STCD-RW GCE-8520B Adapter: 1 ID: 0
Read mode : Secure with C2, accurate stream, NO disable cache
Read offset correction : 0
Overread into Lead-In and Lead-Out : No

Used output format : Internal WAV Routines
44.100 Hz; 16 Bit; Stereo

Other options :
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks : No
Native Win32 interface for Win NT & 2000


Track 1
Filename C:\burning\Track01.wav

Peak level 100.0 %
Track quality 100.0 %
Test CRC 1EA53AFA
Copy CRC 1EA53AFA
Copy OK

No errors occured


End of status report


with my DVD-ROM reader:

Used drive : HL-DT-STDVD-ROM GDR8161B Adapter: 1 ID: 1
Read mode : Secure with C2, accurate stream, NO disable cache
Read offset correction : 96
Overread into Lead-In and Lead-Out : No

Used output format : Internal WAV Routines
44.100 Hz; 16 Bit; Stereo

Other options :
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks : No
Native Win32 interface for Win NT & 2000


Track 1
Filename C:\burning\Track01.wav

Peak level 100.0 %
Track quality 100.0 %
Test CRC 98B0742F
Copy CRC 98B0742F
Copy OK

No errors occured


End of status report


When I compare the two tracks with EAC, I have the following report:

36 repeated samples 0:00:00.097

I have verified my EAC offset for the two drives (comparing original wav files with files extracted from an audio CD, burnt on FUJI japanese discs, extracted with both drives) and everything works fine.

What's wrong? Is this because it is the first track of the CD? How can I know wich drive is the best to extract the song if both gives 100% track quality but different CRC signature?

thanks!
J-F

[Five]

bizarre...

have you ever had different crc signatures when ripping with the two drives using another disc?

[liveman]

bizarre...

have you ever had different crc signatures when ripping with the two drives using another disc?

To be honest, I didn't do the test before today (except when I originally setup the offset correction with EAC). I just did another test using an original silver CD, extracting track 1, both CRC signatures are identical. Curious.

[Five]

hmm..

as for which drive is best to extract with, EAC should tell you that when you run the wizard.

Is the offset for your burner 36 by any chance?

also, it might be worth it to compare the two tracks using foobar2000 (I expect they will turn out to be different, but it's worth a try). How to do this is midway down this thread:
http://www.thetradersden.org/forums/showthread.php?t=1112

Also, if you have a WAV editor that can tell you how many samples long each track is, I'd be curious to see if one is 36 samples longer than the other.

[liveman]

Offset of my burner should be really "0". As I have same CRC results using the same setting (offset 0 burner / offset 96 dvdrom reader) with the two other extraction tests I made.
---------------------

I've made a foobar bitcompare test of the two tracks that give different crc:


INFO (foo_bitcompare) : first different sample found
INFO (foo_bitcompare) : differences found: 31307647 sample(s), starting at

9.741497e-002 second(s), peak: 1.964478 at 15.01472 second(s), 2ch


on Cool Edit Pro both track has length: 15816024 samples

[Five]

okay, using coolpro you can see the exact difference.

1. load both versions into CEP
2. select all and copy the first one then
3. go to the second and select all
4. "mixpaste" (shift+ctrl+v) and invert the phase of both channels.

Now you can see the exact location of the different samples and exactly what is different. I can't figure out how you got two different extractions, but you should be able to use this information to select the better extraction (and the most accurate drive)

[RainDawg]

This is a truly strange on, and I am interested to hear what others think. Seems like something at the VERY start (9.741497e-002 seconds) of the audio is having a hard time being read. I'll read around a bit, but hopefully someone with knowledge of this will chime in here too.

[liveman]

I must do something wrong with Cool Edit "mixpaste" option, because I don't know how to see the difference. I have a very limited use of this software...

However, I think I found something. I checked the WAV file and there is a blank at the very beginning of the track from 0:00.000 to 0:07.399, maybe one drive read this blank part differently?

[Five]

I don't have cep right in front of me so I can't put up a screenshot.

what you want to do is copy one complete file and then paste it out of phase over the other. everything that's identical will become a flat line. anything that's not the flat line is different.

if the difference is indeed during the silent beginning then all of the music has been extracted perfectly (good news!). I still find it strange that you got two different crc's

[liveman]

Just done a test with another track from the Verbatim CDR and now the CRC codes are identical! The key is maybe indeed in the blank...

I'll check CEP again to see if I can do the mixpaste test...

[liveman]

I think the problem could be from the blank part. The CDR I have is made of 3 different concerts. The first track has the blank but there's another track with a blank at the very beginning (it is the "track 4", which is the first track of the 3rd concert on my CDR). This other track also has two different CRC.

All other tracks (that give 100% extraction track quality) have identical CRC except 3 tracks who are not 100% quality (EAC has some diffciculty to read them), in that case, I have not only different CRC from drive to drive but from the TEST CRC / COPY CRC on a same drive. I guess it is normal because my CDR is not a high quality brand and is not clean @ 100% (though there's no scrathes on it).

[Five]

when I have trouble getting 100.00% on some tracks Paranoid mode has often helped me (in drive options). Remember to put it back to Secure after you're done. They say that Paranoid mode could cause drive damage (!!??) but I've used it quite a few times with no probs.

[liveman]

Thanks "Five" for the tip, I'll try it!

[westgroveG]

I can see a problem already
- The GCE-8520B caches audio data, you did not tick disable cache

I have that LG drive & it has produced similar behavior on troublesome CD's; various pairs of matching CRC's for same track.

[liveman]

Thanks westgroveG for pointing the possible trouble! I'll check it when I come back to home.

[westgroveG]

I also suggest you try using a sample offset value of -36 with the GCE-8520B drive.