Example of EAC's unreliable "Track quality 100.0 %" Options

[Akkurat]

Something that I wanted to share, haven't seen anything like this before with my rips nor here in HA.

CD: Metallica - Metallica (used in fairly good condition)

CODE

Metallica - Metallica - 12 - The Struggle Within.wav

Peak level 100.0 %
Track quality 100.0 %
Test CRC B92731B3
Copy CRC C856E97C
Cannot be verified as accurate (confidence 200)  [3DED4995], AccurateRip returned [C612BBDA]
Copy OK

NO re-reads at all.. that's why the track quality is 100% (EAC calculates this only from copy run.. I watched the test run and there were no re-reads either).

Same drive, another rip:

CODE

Metallica - Metallica - 12 - The Struggle Within.wav

Peak level 100.0 %
Track quality 99.9 %
Test CRC C1C95CE6
Copy CRC B92731B3
Accurately ripped (confidence 142)  [B37842EA]
Copy OK

Look at the copy CRC, same as in the first rip test run.

The first rip copy doesn't re-read, wrong CRC. Second rip copy, one bar re-read but correct CRC (nothing wrong with this).

Ignore the non-matching CRC's that AR returns, those are taken from different offsets.

Also, it's possible to get matching T&C CRC's even if the data is read incorrectly.. didn't catch this in these rip examples, can't have everything for a perfect example.

I've never seen EAC track quality 100% and AR mismatch before. Have you?

Without AR, user would be knee deep in the brown stuff.

I'm not so worried about this problem because AR saves the day (this time), but I'd be more than happy to hear if anyone can explain how this is possible.. EAC bug?

EDIT: oh yeah, forgot to add that I checked the first wrong rip with CUETools, no other pressings matched.

This post has been edited by Akkurat: Oct 9 2009, 18:55

[greynol]

Sharing the details about your drive and how EAC was configured to handle it would tell us a whole lot more. That way we won't have to engage in any unnecessary speculation.

Detailing the differences between the two rips could be useful as well.

This post has been edited by greynol: Oct 9 2009, 19:01

[Fandango]

Waht version of EAC do you use Akkurat? I remember there was a bug that took the Copy OK from the Test CRC run. It may be the same issue. Track quality is based on the Test rip and not on the final Copy rip.

IMHO there should be two Track quality lines for each one of the rips, test and copy.

[greynol]

It's clearly not the version that mixed up the test and copy CRCs as evidenced by the "Cannot be verified as accurate" message that did not exist in that version.

[JunkieXL]

It's happened to me in the past. I usually had that happen with bad pressings of CD's or CD's that were pretty damaged.

I also sometimes saw this behavior when I had been using my drive pretty heavily. i.e. ripping CD's for over a couple of hours continuously. After letting it cool back down and rebooting everything was working fine again.
JXL

[Akkurat]

Sharing the details about your drive and how EAC was configured to handle it would tell us a whole lot more.

Yes yes of course, damn, I always forget something relevant.

CODE

Used drive  : HL-DT-STCD-RW GCE-8520B   Adapter: 1  ID: 0

Read mode               : Secure
Utilize accurate stream : Yes
Defeat audio cache      : Yes
Make use of C2 pointers : No

Read offset correction                      : 6
Overread into Lead-In and Lead-Out          : Yes
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks   : No
Null samples used in CRC calculations       : Yes

Unfortunately I'm unable to provide any detailed info of the differences between the rips.

[spoon]

All you need is a constant error when the 2nd read is done and it will get through with 100% quality. c2 pointers might have spotted it.

[Antonski]

Today I had the same behavior with one a bit old disk.
I remember recently I've tried to rip it and I've got some entries in the log showing the suspicious positions of the problematic track.
Now I don't get this info.
I use the same version (0.99pb5) and I don't think I've done some changes in the settings since then.
Do you guys know how to get again the suspicious timestamps?

[Akkurat]

Waht version of EAC do you use

I guess that greynol already made this unnecessary but it's the latest 0.99pb5.

Track quality is based on the Test rip and not on the final Copy rip.

Are you absolutely sure? I think that it's the other way around. It doesn't sound very wise to use the test run when logging/deciding on the track quality. Copy run is the one that is saved to the disc, not the test run. And what about if the user doesn't use test run? Then use copy run? Sounds hairy. I could be wrong but I've always thought that only the copy run decides the track quality.. I've an image in my head that I've verified this myself. Tried searching quickly, I found nothing that could verify this.. does anyone know which way it is?

[greynol]

Do you guys know how to get again the suspicious timestamps?

EAC reports suspicious positions when it cannot get the same result at least 8 times in a set of 16 re-reads. If you want to increase EAC's sensitivity set the error recovery quality to low, yes I said low.

Track quality is based on the Test rip and not on the final Copy rip.

Are you absolutely sure?

He's completely wrong on this, actually.

The issue was that AR results were recorded for the test rip rather than the copy rip in V0.99pb1 and was promptly fixed.

This post has been edited by greynol: Oct 9 2009, 22:56

[Akkurat]

All you need is a constant error when the 2nd read is done and it will get through with 100% quality.

This finally made me study the technicalities a bit more and widen my understanding of what EAC does in secure mode. This of course put a (new) dent in my confidence of EAC (Edit: and/or DAE). Ignorance is bliss. Does anyone know the approximate odds of this happening? Probably slim. Would make easier to weigh the significance of this issue.

This post has been edited by Akkurat: Oct 9 2009, 23:03

[greynol]

Depends on the disc and the drive. I doubt anyone will be able to give you approximate odds.

[Akkurat]

Just as I thought/feared.

Thanks for the replies.

[twostar]

I have a similar experience with one track of my prized CDs. Track quality is at 99.9% but it cannot bet verified with Accuraterip. I also can't get a matching CRC with burst, secure no C2/with C2, and even with foobar2000.

Here are the logs. Notice track01.

CODE

Exact Audio Copy V0.99 prebeta 5 from 4. May 2009

EAC extraction logfile from 8. October 2009, 23:01

Radiohead / OK Computer

Used drive : TSSTcorpCDDVDW SE-S224Q Adapter: 1 ID: 0

Read mode : Secure
Utilize accurate stream : Yes
Defeat audio cache : No
Make use of C2 pointers : No

Read offset correction : 6
Overread into Lead-In and Lead-Out : No
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks : No
Null samples used in CRC calculations : Yes
Used interface : Native Win32 interface for Win NT & 2000
Gap handling : Not detected, thus appended to previous track

Used output format : User Defined Encoder
Selected bitrate : 768 kBit/s
Quality : High
Add ID3 tag : No
Command line compressor : C:\Documents and Settings\ac25\My Documents\My Music\Apps\encoders\flac-1.2.1b\flac.exe
Additional command line options : -T "artist=%a" -T "title=%t" -T "album=%g" -T "date=%y" -T "tracknumber=%n" -T "genre=%m" -5 %s


TOC of the extracted CD

Track | Start | Length | Start sector | End sector
---------------------------------------------------------
1 | 0:00.00 | 4:44.30 | 0 | 21329
2 | 4:44.30 | 6:23.37 | 21330 | 50091
3 | 11:07.67 | 4:27.53 | 50092 | 70169
4 | 15:35.45 | 4:24.60 | 70170 | 90029
5 | 20:00.30 | 4:59.20 | 90030 | 112474
6 | 24:59.50 | 4:21.47 | 112475 | 132096
7 | 29:21.22 | 1:57.25 | 132097 | 140896
8 | 31:18.47 | 3:50.48 | 140897 | 158194
9 | 35:09.20 | 4:45.15 | 158195 | 179584
10 | 39:54.35 | 3:48.40 | 179585 | 196724
11 | 43:43.00 | 4:19.47 | 196725 | 216196
12 | 48:02.47 | 5:24.40 | 216197 | 240536


Track 1

Filename C:\Documents and Settings\admin\Desktop\01 - Airbag.wav

Peak level 98.5 %
Track quality 99.9 %
Copy CRC 21AEF401
Cannot be verified as accurate (confidence 200) [F305E4EA], AccurateRip returned [57E5B427]
Copy OK


No tracks could be verified as accurate
You may have a different pressing from the one(s) in the database

No errors occurred

End of status report

CODE

Exact Audio Copy V0.99 prebeta 5 from 4. May 2009

EAC extraction logfile from 22. August 2009, 1:48

Radiohead / OK Computer

Used drive : TSSTcorpCDDVDW SE-S224Q Adapter: 1 ID: 0

Read mode : Secure
Utilize accurate stream : Yes
Defeat audio cache : No
Make use of C2 pointers : No

Read offset correction : 6
Overread into Lead-In and Lead-Out : No
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks : No
Null samples used in CRC calculations : Yes
Used interface : Native Win32 interface for Win NT & 2000
Gap handling : Not detected, thus appended to previous track

Used output format : User Defined Encoder
Selected bitrate : 768 kBit/s
Quality : High
Add ID3 tag : No
Command line compressor : C:\Documents and Settings\ac25\My Documents\My Music\Apps\encoders\flac-1.2.1b\flac.exe
Additional command line options : -T "artist=%a" -T "title=%t" -T "album=%g" -T "date=%y" -T "tracknumber=%n" -T "genre=%m" -5 %s


TOC of the extracted CD

Track | Start | Length | Start sector | End sector
---------------------------------------------------------
1 | 0:00.00 | 4:44.30 | 0 | 21329
2 | 4:44.30 | 6:23.37 | 21330 | 50091
3 | 11:07.67 | 4:27.53 | 50092 | 70169
4 | 15:35.45 | 4:24.60 | 70170 | 90029
5 | 20:00.30 | 4:59.20 | 90030 | 112474
6 | 24:59.50 | 4:21.47 | 112475 | 132096
7 | 29:21.22 | 1:57.25 | 132097 | 140896
8 | 31:18.47 | 3:50.48 | 140897 | 158194
9 | 35:09.20 | 4:45.15 | 158195 | 179584
10 | 39:54.35 | 3:48.40 | 179585 | 196724
11 | 43:43.00 | 4:19.47 | 196725 | 216196
12 | 48:02.47 | 5:24.40 | 216197 | 240536


Track 1

Filename C:\Documents and Settings\admin\Desktop\noc2\01 - Airbag.wav

Peak level 98.5 %
Track quality 99.9 %
Copy CRC 11345540
Cannot be verified as accurate (confidence 200) [FF375AC3], AccurateRip returned [57E5B427]
Copy OK

Track 2

Filename C:\Documents and Settings\admin\Desktop\noc2\02 - Paranoid Android.wav

Peak level 98.7 %
Track quality 100.0 %
Copy CRC 2807C46F
Accurately ripped (confidence 200) [FE12781B]
Copy OK

Track 3

Filename C:\Documents and Settings\admin\Desktop\noc2\03 - Subterranean Homesick Alien.wav

Peak level 98.6 %
Track quality 100.0 %
Copy CRC 889CEC18
Accurately ripped (confidence 200) [6DDB1E82]
Copy OK

Track 4

Filename C:\Documents and Settings\admin\Desktop\noc2\04 - Exit Music (For A Film).wav

Peak level 98.4 %
Track quality 100.0 %
Copy CRC 39E44B5D
Cannot be verified as accurate (confidence 200) [E5DCECA0], AccurateRip returned [D6E44ACB]
Copy OK

Track 5

Filename C:\Documents and Settings\admin\Desktop\noc2\05 - Let Down.wav

Peak level 98.6 %
Track quality 100.0 %
Copy CRC 96379566
Accurately ripped (confidence 200) [0852CFE4]
Copy OK

Track 6

Filename C:\Documents and Settings\admin\Desktop\noc2\06 - Karma Police.wav

Suspicious position 0:04:21

Peak level 99.2 %
Track quality 99.7 %
Copy CRC 92C54BF0
Accurately ripped (confidence 2) [D99FE099]
Copy finished

Track 7

Filename C:\Documents and Settings\admin\Desktop\noc2\07 - Fitter Happier.wav

Peak level 98.0 %
Track quality 100.0 %
Copy CRC 66772E6F
Accurately ripped (confidence 200) [18D59A8E]
Copy OK

Track 8

Filename C:\Documents and Settings\admin\Desktop\noc2\08 - Electioneering.wav

Peak level 98.5 %
Track quality 100.0 %
Copy CRC 48BE730D
Accurately ripped (confidence 200) [7C13429B]
Copy OK

Track 9

Filename C:\Documents and Settings\admin\Desktop\noc2\09 - Climbing Up the Walls.wav

Peak level 98.5 %
Track quality 100.0 %
Copy CRC B7E538D0
Accurately ripped (confidence 200) [FC3CE3DE]
Copy OK

Track 10

Filename C:\Documents and Settings\admin\Desktop\noc2\10 - No Surprises.wav

Peak level 98.7 %
Track quality 100.0 %
Copy CRC 67DC09B9
Accurately ripped (confidence 200) [CADC3EFC]
Copy OK

Track 11

Filename C:\Documents and Settings\admin\Desktop\noc2\11 - Lucky.wav

Peak level 98.9 %
Track quality 100.0 %
Copy CRC 9F03412F
Accurately ripped (confidence 200) [B6290C35]
Copy OK

Track 12

Filename C:\Documents and Settings\admin\Desktop\noc2\12 - The Tourist.wav

Peak level 98.8 %
Track quality 100.0 %
Copy CRC D132BA5A
Accurately ripped (confidence 200) [164B3CA6]
Copy OK


10 track(s) accurately ripped
2 track(s) could not be verified as accurate

Some tracks could not be verified as accurate

There were errors

End of status report

This post has been edited by twostar: Oct 10 2009, 00:02

[Antonski]

Do you guys know how to get again the suspicious timestamps?

EAC reports suspicious positions when it cannot get the same result at least 8 times in a set of 16 re-reads. If you want to increase EAC's sensitivity set the error recovery quality to low, yes I said low.

Yes, that makes sense, thanks a lot greynol!
BTW, I succeeded to rip the disk by accident, the workaround probably works on laptops only.
I just forget to plug the laptops charger to the socket, so the CD drive was obviously working at lower speed.
I got really amazed to see that the track was ripped with track quality 100.0 % and equal test and copy checksums!

[Akkurat]

I just forget to plug the laptops charger to the socket, so the CD drive was obviously working at lower speed.

Lower rev speed? Sounds familiar. Check my Nero DiscSpeed trick. I've been promoting this trick I recently found because I really would like to hear if that works with others too.

[Akkurat]

Notice track01.

I'd be more interested about tracks 4 and 6.

Track 4: Track quality 100.0% but no AR.
Track 6: Suspicious position, lot's of re-reads (track quality 99.7%), Copy finished (not OK), yet still AR, BUT with confidence 2 (others are 200 (EAC limit I guess, means 200 OR over, right?))

I don't find the logs for track 1 odd at all. It just seems that there's an error detected every time you rip it, that's all.

[Andavari]

I've never seen EAC track quality 100% and AR mismatch before. Have you?

I have seen that before and think if someone has a big enough CD collection (as I do) something is bound to pop up that's a little odd and unexpected.

The old saying of trust your ears is what I always go by to manually detect any audible errors that slip through so I listen to all rips carefully - with headphones of course. Some of those "errors" I suspect are built right into the original recordings so no way to fix those unless getting very clever with a wave editor.

For me EAC's matching CRC's are a "comfort zone" however they're definitely not a 100% ironclad guarantee you have a good rip. My main reason for using EAC's secure mode is to detect issues with problem discs that need special attention to get a good rip. Special attention meaning I may have to do a Burst mode test & copy, use a different drive, or use another ripping program altogether like CDex or BonkEnc with CD Full Paranoia enabled - all in the goal to get the best possible rip, albeit it may not be "perfect."

[twostar]

Notice track01.

I'd be more interested about tracks 4 and 6.

They do seem odd. I didn't pay attention to them before since I was able to get Accuraterip confidences of 200 for all tracks except track01 with another rip.

[Antonski]

I just forget to plug the laptops charger to the socket, so the CD drive was obviously working at lower speed.

Lower rev speed? Sounds familiar. Check my Nero DiscSpeed trick. I've been promoting this trick I recently found because I really would like to hear if that works with others too.

I guess yes, lower rev speed. I checked the link but I was not able to reproduce your trick, usually one of the program (sometimes both) get stuck, even the CD drive refuse to recognize the CD anymore (Lenovo R61). However, I think that I was able to lower the rev speed on my old Fujitsu-Siemens (very nice machine!) using the described trick. I have to dig it out and check some day.