Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: EAC vs. CDex (Read 6514 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

EAC vs. CDex

[span style='font-size:14pt;line-height:100%']Introduction[/span]

I wanted to test EAC’s secure mode, old secure mode (Paranoid), CDex’s Paranoia implementation at recovering errors to do this with accurate results I first needed to find a drive which 1. Does not jitter & 2. Does not cache (or buffer) audio data because these things make accurate error recovery more complicated & currently CDex does not support drives which cache audio & jitter. My Pioneer 106s was my most suited drive so chose it for the test. Second I needed to find a CD with a track that had unrecoverable errors I chose a pressed CD with light & heavy scratches that I had lying around.. I would later need to clean the CD to create a reference copy of the track.

[span style='font-size:14pt;line-height:100%']Conditions[/span]

Drive used was a Pioneer 106s DVD-ROM with PIO transfer mode set to disable the drives audio caching


The software & setting I used were,

EAC V0.95pb5
  • Secure Mode
  • No disable Cache (no correction used)
  • Yes Accurate Stream (no correction used)
  • No C2
EAC V0.95pb5
  • Paranoid mode
  • No disable cache
CDex 1.51
  • Paranoia, Full
Track was ripped with each program, I then cleaned the CD & extracted the track with my Lite-on 811s using EAC secure mode & test & copy (achieved a CRC match) which I used as a reference file. I then bit compared each track from each extraction program against the reference file using foobar.

[span style='font-size:14pt;line-height:100%']Results[/span]

Error Reporting Results

Of course EAC’s secure mode reported errors
[span style='font-size:8pt;line-height:100%']
Code: [Select]
     Suspicious position 0:01:25 - 0:01:26
     Suspicious position 0:01:28 - 0:01:48

     Peak level 99.8 %
     Track quality 94.7 %
     Copy CRC DE218CDC
     Copy finished

There were errors
[/span]

EAC's old secure mode (Paranoid) reported errors
[span style='font-size:8pt;line-height:100%']
Code: [Select]
     Suspicious position 0:01:26
     Suspicious position 0:01:29 - 0:01:32
     Suspicious position 0:01:34 - 0:01:43
     Suspicious position 0:01:46 - 0:01:47

     Peak level 99.8 %
     Track quality 39.8 %
     Copy CRC 093BC072
     Copy finished

There were errors
[/span]

Errors were reported by CDex 1.51 Paranoia, Full
[span style='font-size:8pt;line-height:100%']
Code: [Select]
Jitter error detected (not corrected!) near 1:25 [m:s] (range 85373 - 85626 [ms], absolute 102203 - 102222 [sectors])

Jitter error detected (not corrected!) near 1:25 [m:s] (range 85720 - 85973 [ms], absolute 102229 - 102248 [sectors])

Jitter error detected (not corrected!) near 1:26 [m:s] (range 86066 - 86320 [ms], absolute 102255 - 102274 [sectors])

Jitter error detected (not corrected!) near 1:46 [m:s] (range 106866 - 107120 [ms], absolute 103815 - 103834 [sectors])

Jitter error detected (not corrected!) near 1:47 [m:s] (range 107213 - 107466 [ms], absolute 103841 - 103860 [sectors])

total duration: 00:05:56
[/span]

All programs & modes reported errors at similar positions. EAC's old secure mode seems to have missed a few errors though.


Error Recovery Results

EAC was best at recovering errors, CDex was at a close second & EAC's old secure mode (Paranoid) was worst.

http://www.hydrogenaudio.org/forums/index....ype=post&id=973


Extraction Speed Results

EAC's old secure mode (Paranoid) was the fastest, CDex came second & EAC's secure mode last

http://www.hydrogenaudio.org/forums/index....ype=post&id=974

[span style='font-size:14pt;line-height:100%']Conclusion[/span]

EAC’s secure mode had the least amount of errors making it best at recovering errors but difference wasn't large between EAC's secure mode & CDex Paranoia, Full. I believe the differences in scores are due to the amount of read-retries the program commands the drive to take.

CDex 1.51 Paranoia, Full did very well & the test has shown that when using a supported drive (a drive that doesn't cache audio or jitter) CDex does in fact report not corrected errors & is a good compromise between speed/quality for recovering errors.

EAC's old secure mode (Paranoid) was the fastest & may be helpful for highly damaged CD's because of it's speed but should be avoided if you want accuracy.

EAC vs. CDex

Reply #1 –
Thank you for the test.

Two more things could have been included : burst mode, to see if it performs better or worse... Different programs used in burst mode (for example EAC vs Feurio) have already been reported to perform very differently.
And a test for eror detection. But for this, you'd need to find a CD with errors that are all correctable.

However, I find the error count very similar for the three tests. I bet that you could have found the same difference if you had given one of your programs a second try, without changing any option.

EAC vs. CDex

Reply #2 –
Firat of, congratulations on a test with the most popular rippers!

As i am a Linux user, i use CDParanoia but as stated, this is useless unless the drive doesn't cache audio data! My Pioneer 107 DVD recorder doesn't, according to EAC which is great, and it does report errors on CD's using CDParanoia! I remember using my old plextor 12/10/32?? which used to cache after a certain firmware version, but not before, which was very confusing!

I know the AcurateRip database of drive contains read offsets for cerain drives, the SatCP? EAC database contains caching, read/write offsets etc. for drives for using EAC. Are there any other databases that are A, not out of date, or B cover a more diverse range of details about specific drives? If there are not, is there any way a specific database could be concieved? If this has been discussed before, then please stop me where i am.

If such a scheme hasn't been carried out before then perhaps we could develop a new database that everone here, and elsewhere can contribute to and verify?

The test parameters for any drive could be constrained to a specific piece of test software, for example i know EAC can be run under linux, and so this could be used to test drive features, or perhaps a new test suite could be developed, that can verify the capabilities of specific drives and firmware revisions?

Cheers,

Kristian

EAC vs. CDex

Reply #3 –
Quote
Thank you for the test.

Two more things could have been included : burst mode, to see if it performs better or worse... Different programs used in burst mode (for example EAC vs Feurio) have already been reported to perform very differently.


Yes I agree it would have been interesting to test burst mode against the error recovery modes. I also would have liked to have tested EAC's experimental extended C2 correction but my Pioneer didn't support C2.

Quote
And a test for eror detection. But for this, you'd need to find a CD with errors that are all correctable.

Well this is a hard thing to do because how can you find such a CD? the only practical way I can think of is using EAC with test & copy until you have a CRC mismatch.

I once run into such a situation.

Quote
However, I find the error count very similar for the three tests. I bet that you could have found the same difference if you had given one of your programs a second try, without changing any option.


Well you can only assume that. To know for sure you would have to test each program/mode several times & see the results but I don't think they would be very different compared to the current test results especially considering how consistent the extraction time was with the different samples (or errors); the faster program/mode the more different samples.

EAC vs. CDex

Reply #4 –
Quote
Well this is a hard thing to do because how can you find such a CD? the only practical way I can think of is using EAC with test & copy until you have a CRC mismatch.


I've tried to burn a CD and scratch it lightly, but without success, because I always pass from no errors at all straight to unrecoverable errors everywhere.

However, if you have old CDRs that are becoming unreadable, it is easy. Take one with some tracks OK, and some tracks dead (usually the last ones).
Then, since the error rate changes progressively, you can always find a range where it is high enough to produce C2 errors, and low enough for them to be correctable.