Different schools of thought regarding EAC, Secure vs Test & Copy Burst. Options

[EagleScout1998]

There appears to be two different schools of thought regarding the accuracy of using burst mode in conjunction with test and copy. It is without question that burst mode alone is unwise. However, it makes sense to me that if the test and copy CRCs are a match, than the rip is as accurate as it can get. In fact, I believe I have read somewhere that the author of EAC (Andre Wiethoff) uses this method all the time. In my mind, matching CRCs are a better way to gauge the accuracy of a rip than a mere message saying "no errors occured."

I do somewhat question using this method on discs in less than fair condition. I do know that it is possible that a read error will give the same value twice in a row (but even secure mode would be compromised in that situation).

I would be perfectly comfortable using this method as well if it were not for the opinions of those that firmly belief secure mode is the ONLY way to go. I would not be opposed to using secure mode, except I am re-ripping my CD collection into a lossless format. (I am kicking myself because I used lossy the first time). I do not want to spend an entire year doing this (I have over 1,000 CDs).

Is there a definitive answer?

[lextune]

It is true that Andre uses Burst+T&C. and only uses Secure+T&C if that produces errors.

I, for one use secure.

/me shrugs

[Martin H]

Burst mode + Test & Copy is just as secure as secure mode with no C2 and caching disabled(if present) and getting "No errors occured" + 100% track quality. Secure mode with C2 enabled has the added advantage of also being able to detect consistent errors, but sadly they won't be reported as EAC reverts to it's standard majority voting system when uncorectable C2 errors are flagged.

You can go ahead and use Burst mode + Test & Copy without a doubt as it is atleast as safe as EAC's Secure mode, and just ignore people that dosen't know what they are talking about...

[gaillard]

Or you can use Spoons new R12 ripper that addresses that issue and more.

[Firon]

I use burst T&C because I consider it to be just as secure as secure mode, provided the CRCs match, and a lot less stressing on the drive. Only if the CRCs don't match do I re-rip the tracks in secure mode.

[satorippoi]

All this (Burst+T&C) makes sense only if you rip into separate tracks...
If you make cue-sheets - then secure mode with no c2 is much more preferable...

[db1989]

Why is that?

Andre has not yet implemented T&C for images yet, but it is already simply enough to do and CRCs are still available.

[greynol]

It's all simply a means to an end. Matching CRCs does not mean a rip was accurate. Precise, yes, accurate no.

Let AccurateRip be your arbiter.

Ripping single-file images instead of separate tracks are fine, too. Test and Copy is still possible with this, and it's even possible to still use AccurateRip after jumping through a few hoops.

In the event you get an error while ripping an image, what do you do? Do you use a wave editor to patch the bad spot, or do you rip the whole thing over again?

[evereux]

In the event you get an error while ripping an image, what do you do? Do you use a wave editor to patch the bad spot, or do you rip the whole thing over again?

Or try a different drive?

[funkyblue]

It's all simply a means to an end. Matching CRCs does not mean a rip was accurate. Precise, yes, accurate no.

I would dispute that. The track would be accurate. I would say it's near IMPOSSIBLE at such high rip speeds to get the SAME CRC if there was ANY errors on the disc. For all intents and purposes Bust Mode T & C with Matching CRC's is ACCURATE!

[Eli]

Or you can use Spoons new R12 ripper that addresses that issue and more.

Agreed, I think when the new ripper goes wild, EAC will be a thing of the past. Its only in alpha and it is already amazing.

[funkyblue]

Yes, and it will also cost money!

[greynol]

I would dispute that. The track would be accurate. I would say it's near IMPOSSIBLE at such high rip speeds to get the SAME CRC if there was ANY errors on the disc. For all intents and purposes Bust Mode T & C with Matching CRC's is ACCURATE!

I'm guessing you've never encountered a consistent error before?

I've seen plenty. The great thing about consistent errors is that you can easily reproduce your findings. Would you like to see some log files or links demonstrating this? No need to answer, here's one of the more troubling rips I've seen:

CODE

EAC extraction logfile from 9. September 2006, 0:07 for CD
Dire Straits / Alchemy Part One

Used drive : PLEXTOR DVDR PX-760A Adapter: 1 ID: 1
Read mode : Burst
Read offset correction : 30
Overread into Lead-In and Lead-Out : Yes

Used output format : C:\Program Files\ExactAudioCopy\Flac\flac.exe (User Defined Encoder)
320 kBit/s
Additional command line options : -T "artist=%a" -T "title=%t" -T "album=%g" -T "date=%y" -T "tracknumber=%n" -T "genre=%m" %s -8 --verify

Other options :
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks : No
Installed external ASPI interface


Track 5
Filename E:\Ripped CD's\Rerip\Dire Straits - 05 - Private Investigations.wav

Peak level 89.1 %
Test CRC 570D3692
Copy CRC 570D3692
Copy OK

No errors occured


End of status report


------------------------------------------------------------

EAC extraction logfile from 9. September 2006, 0:12 for CD
Dire Straits / Alchemy Part One

Used drive : PLEXTOR DVDR PX-760A Adapter: 1 ID: 1
Read mode : Secure with C2, accurate stream, disable cache
Read offset correction : 30
Overread into Lead-In and Lead-Out : Yes

Used output format : C:\Program Files\ExactAudioCopy\Flac\flac.exe (User Defined Encoder)
320 kBit/s
Additional command line options : -T "artist=%a" -T "title=%t" -T "album=%g" -T "date=%y" -T "tracknumber=%n" -T "genre=%m" %s -8 --verify

Other options :
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks : No
Installed external ASPI interface


Track 5
Filename E:\Ripped CD's\Rerip\Dire Straits - 05 - Private Investigations.wav

Peak level 89.1 %
Track quality 100.0 %
Test CRC 6982AF7B
Copy CRC 6982AF7B
Copy OK

No errors occured


End of status report

------------------------------------------------------------

EAC extraction logfile from 9. September 2006, 1:24 for CD
Dire Straits / Alchemy - Dire Straits Live Part One

Used drive : PLEXTOR CD-ROM PX-32TS Adapter: 0 ID: 3
Read mode : Burst
Read offset correction : 679
Overread into Lead-In and Lead-Out : Yes

Used output format : C:\Program Files\ExactAudioCopy095pb5\Flac\flac.exe (User Defined Encoder)
320 kBit/s
Additional command line options : -T "artist=%a" -T "title=%t" -T "album=%g" -T "date=%y" -T "tracknumber=%n" -T "genre=%m" %s -8 --verify

Other options :
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks : No
Installed external ASPI interface


Track 5
Filename G:\TEMP\alchemy\Dire Straits - 05 - Private Investigations.wav

Peak level 89.1 %
Test CRC 570D3692
Copy CRC 570D3692
Copy OK

No errors occured


End of status report

------------------------------------------------------------

EAC extraction logfile from 9. September 2006, 1:18 for CD
Dire Straits / Alchemy - Dire Straits Live Part One

Used drive : PLEXTOR CD-ROM PX-32TS Adapter: 0 ID: 3
Read mode : Secure with C2, accurate stream, NO disable cache
Read offset correction : 679
Overread into Lead-In and Lead-Out : Yes

Used output format : C:\Program Files\ExactAudioCopy095pb5\Flac\flac.exe (User Defined Encoder)
320 kBit/s
Additional command line options : -T "artist=%a" -T "title=%t" -T "album=%g" -T "date=%y" -T "tracknumber=%n" -T "genre=%m" %s -8 --verify

Other options :
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks : No
Installed external ASPI interface


Track 5
Filename G:\TEMP\alchemy\Dire Straits - 05 - Private Investigations.wav

Peak level 89.1 %
Track quality 100.0 %
Test CRC 56C511B5
Copy CRC 56C511B5
Copy OK

No errors occured


End of status report

------------------------------------------------------------

EAC extraction logfile from 9. September 2006, 2:00 for CD
Dire Straits / Alchemy Part One

Used drive : HL-DT-STDVD-ROM GDR8164B Adapter: 0 ID: 1
Read mode : Burst
Read offset correction : 102
Overread into Lead-In and Lead-Out : No

Used output format : C:\Program Files\ExactAudioCopy\Flac\flac.exe (User Defined Encoder)
320 kBit/s
Additional command line options : -T "artist=%a" -T "title=%t" -T "album=%g" -T "date=%y" -T "tracknumber=%n" -T "genre=%m" %s -8 --verify

Other options :
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks : No
Installed external ASPI interface


Track 5
Filename E:\Ripped CD's\Rerip\Dire Straits - 05 - Private Investigations.wav

Peak level 89.1 %
Test CRC F08A88E8
Copy CRC B35EA6F3
Copy OK

No errors occured


End of status report

------------------------------------------------------------

EAC extraction logfile from 9. September 2006, 0:32 for CD
Dire Straits / Alchemy Part One

Used drive : HL-DT-STDVD-ROM GDR8164B Adapter: 0 ID: 1
Read mode : Secure with C2, accurate stream, disable cache
Read offset correction : 102
Overread into Lead-In and Lead-Out : No

Used output format : C:\Program Files\ExactAudioCopy\Flac\flac.exe (User Defined Encoder)
320 kBit/s
Additional command line options : -T "artist=%a" -T "title=%t" -T "album=%g" -T "date=%y" -T "tracknumber=%n" -T "genre=%m" %s -8 --verify

Other options :
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks : No
Installed external ASPI interface


Track 5
Filename E:\Ripped CD's\Rerip\Dire Straits - 05 - Private Investigations.wav

Peak level 89.1 %
Track quality 100.0 %
Test CRC B35EA6F3
Copy CRC B35EA6F3
Copy OK

No errors occured

------------------------------------------------------------

End of status report

EAC extraction logfile from 9. September 2006, 0:36 for CD
Dire Straits / Alchemy - Dire Straits Live Part One

Used drive : _NEC DVD_RW ND-3540A Adapter: 1 ID: 0
Read mode : Burst
Read offset correction : 48
Overread into Lead-In and Lead-Out : No

Used output format : C:\Program Files\ExactAudioCopy\Lame\LAME.EXE (Monkey's Audio Lossless Encoder)
Fast Lossless Compression
Additional command line options : %l--alt-preset 128%l%h--alt-preset standard%h %s %d

Other options :
Fill up missing offset samples with silence : Yes
Delete leading and trailing silent blocks : No
Installed external ASPI interface


Track 5
Filename D:\MP3 nog uitzoeken\Posten\Dire Straits - 05 - Private Investigations.wav

Peak level 89.1 %
Test CRC 570D3692
Copy CRC 570D3692
Copy OK

No errors occured


End of status report

------------------------------------------------------------

EAC extraction logfile from 9. September 2006, 0:39 for CD
Dire Straits / Alchemy - Dire Straits Live Part One

Used drive : _NEC DVD_RW ND-3540A Adapter: 1 ID: 0
Read mode : Secure with C2, accurate stream, NO disable cache
Read offset correction : 48
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
Installed external ASPI interface


Track 5
Filename D:\MP3 nog uitzoeken\Posten\Dire Straits - 05 - Private Investigations.wav

Peak level 89.1 %
Track quality 100.0 %
Test CRC 4FFB6246
Copy CRC 4FFB6246
Copy OK

No errors occured


End of status report

------------------------------------------------------------

EAC extraction logfile from 9. September 2006, 2:02 for CD
Dire Straits / Alchemy - Dire Straits Live Part One

Used drive : HL-DT-STDVD-ROM GDR8161B Adapter: 2 ID: 0
Read mode : Burst
Read offset correction : 102
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
Installed external ASPI interface


Track 5
Filename G:\TEMP\alchemy\Dire Straits - 05 - Private Investigations.wav

Peak level 89.1 %
Test CRC B35EA6F3
Copy CRC B35EA6F3
Copy OK

No errors occured


End of status report

------------------------------------------------------------

EAC extraction logfile from 9. September 2006, 2:04 for CD
Dire Straits / Alchemy - Dire Straits Live Part One

Used drive : HL-DT-STDVD-ROM GDR8161B Adapter: 2 ID: 0
Read mode : Secure with C2, accurate stream, disable cache
Read offset correction : 102
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
Installed external ASPI interface


Track 5
Filename G:\TEMP\alchemy\Dire Straits - 05 - Private Investigations.wav

Peak level 89.1 %
Track quality 100.0 %
Test CRC B35EA6F3
Copy CRC B35EA6F3
Copy OK

No errors occured


End of status report

EDIT: BTW, according to AccurateRip (which has several submissions for this exact pressing of this disc) none of the above attempts was accurate.

This post has been edited by greynol: Oct 16 2006, 06:17

[gaillard]

I hope it won't. It would be nice if he earns his money off selling it to business and making it free for personal use. Business is where the money is anyways. Not a couple of audiophiles... =)

Maybe if we shower spoon with compliments...

[spoon]

I would say it's near IMPOSSIBLE at such high rip speeds to get the SAME CRC if there was ANY errors on the disc

Not at all, lets say for example (a frame which is made up of 2352 bytes) is ripped as (g is good, b is bad):

gggggggggbgggggggggg

it is a small part which is bad, the drive (depending on model) will self interpolate, that is it takes the last g sample before b and first g sample after b and draws a line between. If you were to rip twice on that drive you would think it was without error, when it does have an error.

[greynol]

What's even more discouraging is that in my above example the differences between the different versions of the same track had nothing to do with interpolation.

[spoon]

It was odd how the CRCs matched then the rip later the next 2 matched, the quality is always 100%, perhaps c2 would find the error.

It would be interesting to see what the correct CRC for that cd is, have you ever had track 5 rip accurately according to accuraterip? (what about in r12 with c2 on and a high number of re-rips, say 700).

[Eli]

Yes, and it will also cost money!

well worth it for the features, which include AMG tagging w/ album art, not to mention a ripper that is outperforming EAC in all my test cases.

[Be Positive]

So Test&Copy in Burst mode is not recommentable anyway, but it's safe enough to copy without testing in secure mode without C2?

I'm switching around, when I have a mint CD to rip and it's not in accuraterip database, I rip in T&C Burst mode, both CRCs are always equal, only when D is scratched I use Secure Mode. Thought this way was OK...

[JeanLuc]

What's even more discouraging is that in my above example the differences between the different versions of the same track had nothing to do with interpolation.

Now I know why I mostly rip with two different drives ... one does the test run, the other does the copy run.

[greynol]

I'm switching around, when I have a mint CD to rip and it's not in accuraterip database, I rip in T&C Burst mode, both CRCs are always equal, only when D is scratched I use Secure Mode. Thought this way was OK...

It usually is.

With my Plextor PX-716A I use T&C in secure mode, but only because -usefua works with the drive and the C2 performance is pretty reliable. Otherwise I'd be using burst mode like I do with all my other drives unless I've been given reason to believe that re-reading will be necessary in order to get an accurate rip as you've mentioned.

I thought it would be helpful to get the word out so that people might not be misguided into thinking that matching CRCs mean that rips are accurate.

Also, looks can be deceiving. I've seen brand new discs (that aren't copy protected) or discs that appear to be in mint condition that can't be ripped accurately. Chalk it up to a manufacturing defect, I guess.

Now I know why I mostly rip with two different drives ... one does the test run, the other does the copy run.

You'll notice that various drives were used to rip the track I've given as an example and different drives got the same CRC. None of them got the rip right. Plextools wasn't able to get it right using the PX-760A either.

In general, I think this is a good idea to use different drives especially when using drives designed around different chipsets. IOW, using a PX-712 and a PX-708 probably won't give you as much diversity as a PX-712 and PX-230. I could have thown in a drive made by NEC instead of the PX-230, but I like hammering home that the PX-230 isn't a real Plextor. . Speaking of NEC, using one with a Pioneer probably won't do much good.

EDIT: BTW, Some people have suggested that using speed reduction in burst mode can produce more accurate results. How well this works will depend on the drive and the disc being ripped.

This post has been edited by greynol: Oct 16 2006, 19:52

[EagleScout1998]

but I like hammering home that the PX-230 isn't a real Plextor. .

It is a shame Plextors are difficult to come by where I live. I had to order mine from the Internet. Only after I received it did I realize it was the PX-230A.

I do not want to ship it back, and I do not want to pay the $150 for the model you mentioned (PX-716A). I assume that matching CRCs, in conjunction with AccurateRip, will be sufficient in any mode. I just wish more of my discs were in its database.

Correction: I knew the drive was the PX-230. What I did not know, until after it arrived, was that this drive was the one dubbed "not a 'real' Plextor."

This post has been edited by EagleScout1998: Oct 17 2006, 01:00

[gaillard]

you can find 708a's pretty cheap...

and grey how do you know the drives weren't interpolating in those examples?

[greynol]

I do not want to ship it back, and I do not want to pay the $150 for the model you mentioned (PX-716A). I assume that matching CRCs, in conjunction with AccurateRip, will be sufficient in any mode.

Don't get me wrong, it's supposed to be a very good drive. Some have suggested that it is more accurate than newer "real" Plextors like the Premiums or the PX-708.

My point was that when doing what JeanLuc suggested, you're better off with a PX-230 and a PX-712 than you are with a PX-708 and PX-712.

I just wish more of my discs were in its database.

Me too!

and grey how do you know the drives weren't interpolating in those examples?

By looking at the the files. What I found were versions where 12 samples got repeated, each in slightly different areas.