IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
EAC and C2 Pointers, How they work in drives
dgauze
post May 17 2009, 00:23
Post #1





Group: Members
Posts: 45
Joined: 13-March 09
Member No.: 67901



After reading some contradictory posts on this forum, I wanted to clarify some points. With regards to EAC, does a drive with bad C2 implementation :

a) Only fail to report errors when they exist.

b) Fail to report existing errors AND report errors when there aren't any.

All this using secure mode. If the first is true, then wouldn't using T&C With C2 in secure mode always be a safer bet that using Secure without c2?

BTW, I have been using C2 pointers with Test & Copy during ripping with my TSSTCorp TS-H653b (basically a samsung sh-s203b) with good results. Are these settings ideal for my drive?
Go to the top of the page
+Quote Post
greynol
post May 17 2009, 00:43
Post #2





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



The problem is with drives that fail to report all existing errors.

I have seen no evidence of a drive reporting errors where there weren't any. Even if there are drives that exhibit this behavior, this is a non-issue for EAC since EAC does not use C2 pointers when performing re-reads. I feel strange saying this since I feel that it is preferable that C2 pointers are used when performing re-reads. In the event that a ripper drops frames that are erroneously marked bad, I would hope it would still choose this "bad" data based on the fact that it is consistent.

Regarding whether it's safer bet to use C2 pointers w/T&C over secure rips without C2 pointers and presumably without T&C, I don't feel comfortable making that call. My interest in this subject primarily revolves around the exceptions to rules regarding "best practice". I have seen instances where one method has delivered errors that would otherwise go undetected when the other method doesn't and vice-versa.

The most thorough way to be sure your rips don't have hidden errors is to use a combination of techniques, including multiple ripping methods and multiple drives. Most importantly is judicious use of AccurateRip.

Does your TS-H653b cache audio data? If so, I probably would use Burst mode, at least with discs that are in good condition. EAC is not efficient when working with caching drives, even when combined with C2 pointers. If it doesn't cache audio data, then I'd use C2 pointers with T&C if ripping that way gives burst speeds (some drives do not do so well with EAC in secure mode with any combination of features), though I would also perform a burst pass or a pass with a different drive if I were feeling paranoid about my rips.


--------------------
Placebophiles: put up or shut up!
Go to the top of the page
+Quote Post
dgauze
post May 17 2009, 00:57
Post #3





Group: Members
Posts: 45
Joined: 13-March 09
Member No.: 67901



My drive is faster when using C2 pointers, but not as fast as burst mode. And no, it doesn't cache audio. I have "tested" my drive to some extent, and both w/ and w/o C2 have always produced the same results. But my decision to enable using C2 pointers has been somewhat due to the following (based on my understanding thus far, and also in addition to your advice on a previous post):

Given what you said, wouldn't Secure, C2 with T&C always be the same or better than Secure w/o C2 (no T&C)? Logically, since EAC will perform the same during rereads on both methods, and each method will read the tracks twice, the first method should always produce as good or better results, regardless of the C2 accuracy or reliability of the drive? (all this provided it doesn't report errors when there are none on the CD)

To clear up my thoughts, If one's goal is to merely detect errors with the highest possible degree of accuracy, then wouldn't the above point make sense?

This post has been edited by dgauze: May 17 2009, 01:03
Go to the top of the page
+Quote Post
greynol
post May 17 2009, 01:03
Post #4





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



Could you explain why you think Secure T&C w/C2 pointers has the potential of delivering better results than ripping w/o C2 pointers?


--------------------
Placebophiles: put up or shut up!
Go to the top of the page
+Quote Post
dgauze
post May 17 2009, 01:10
Post #5





Group: Members
Posts: 45
Joined: 13-March 09
Member No.: 67901



In the case of Secure w/ C2 T&C vs. Secure w/ C2, no T&C:

My understanding is, In each case the data is read twice and verified (either by CRCs in the first, or a check between much smaller blocks in the second), and when an error is found with either method, the same rereads are performed (not relying on C2 information). Is my understanding correct?

(Maybe not better, but AS good, regardless of the drives ability regarding C2 pointers??)

Another related, and possible more appropriate question:

Will any drive become Less accurate if C2 pointers were being utilized with t&C, as opposed to without?

This post has been edited by dgauze: May 17 2009, 01:14
Go to the top of the page
+Quote Post
greynol
post May 17 2009, 01:23
Post #6





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (dgauze @ May 16 2009, 17:10) *
Is my understanding correct?

Yep.

QUOTE (dgauze @ May 16 2009, 17:10) *
Will any drive become Less accurate if C2 pointers were being utilized with t&C, as opposed to without?

Well your drive shouldn't be any less accurate simply because it's being asked to do the same thing a second time around.

What I think you mean to ask is whether your rips are guaranteed the same level of accuracy when using C2 pointers without T&C as they are with T&C. The answer is no. If your drive fails to report an error to EAC and this error would show an inconsistency between test and copy passes, you wouldn't be able to detect it without a test CRC. Of course I'm assuming that AR isn't able to verify any tracks on the disc in this example. If you get AR verification then a test pass really isn't necessary.

This post has been edited by greynol: May 17 2009, 01:24


--------------------
Placebophiles: put up or shut up!
Go to the top of the page
+Quote Post
dgauze
post May 17 2009, 01:37
Post #7





Group: Members
Posts: 45
Joined: 13-March 09
Member No.: 67901



I apologize for not being very clear. I was kind of in a hurry.

The REAL question smile.gif

Would someone who has their drive (non caching, capable of reporting C2 errors according to EAC) set up for Secure ripping, no caching, not utilizing C2 pointers and not using a test pass lose any "secure-ness" or accuracy or error-reporting if they were to change their ripping process and set-up to Secure, with C2, and start using T&C? And why?

In other words, how are these two different, and why?

a) Secure mode, C2 NOT checked, not using Test & Copy

b) Secure mode, C2 checked, using T&C

The above points I made seem to suggest that a drive, regardless of its accuracy in reporting C2 data, should provide the same level of "secure" ripping using either of these two methods.

This post has been edited by dgauze: May 17 2009, 01:48
Go to the top of the page
+Quote Post
greynol
post May 17 2009, 02:09
Post #8





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



a) EAC asks the drive to read 27 sectors at a time in a burst of 2MB then repeats. For the next 2MB it overlaps two frames to make sure the data is synchronized. If it finds a discrepancy it performs re-read sets over 27 sectors including that which contained the discrepancy.

b) EAC asks the drive to read 24 sectors at a time in a burst of 2MB. For the next 2MB it overlaps two frames to make sure the data is synchronized. For sectors where the drive reports an error it performs re-read sets over 27 sectors including that which contained the error. This procedure is done twice, once for the test pass generating a CRC of the data that was ripped and once where the data is written to your hard drive with a CRC generated for this data as well.

Neither method is any more secure than the other.


--------------------
Placebophiles: put up or shut up!
Go to the top of the page
+Quote Post
dgauze
post May 17 2009, 02:43
Post #9





Group: Members
Posts: 45
Joined: 13-March 09
Member No.: 67901



Thanks for clearing that up, Greynol. I have recently switched from set-up (a) to (b), due to a suggestion by you on another thread, advising me to give C2 pointers a chance. Since, then I've wondered what the drawbacks would be if in fact my drive isn't very reliable in reporting C2 errors. Now I know there are none!

Is it possible, given a consistent error on a disc, that set-up (b) could be more effective in reporting that error if in fact the drive in question did reliably report C2 errors?
Go to the top of the page
+Quote Post
greynol
post May 17 2009, 02:49
Post #10





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



It will report the error, sure, but when it comes time to perform re-reads, EAC will think the error is good data if it is consistent enough to pass EAC's criteria (at least 8 matches within a set of 16 re-reads). This is where using C2 pointers during re-reads can make a big difference.

This post has been edited by greynol: May 17 2009, 02:50


--------------------
Placebophiles: put up or shut up!
Go to the top of the page
+Quote Post
dgauze
post May 17 2009, 06:54
Post #11





Group: Members
Posts: 45
Joined: 13-March 09
Member No.: 67901



So in it's current implementation, are you saying it serves no purpose to use C2 pointers when ripping CDs with EAC, unless if you were watching the screen while the ripping was taking place, to note whether C2 errors were reported by your drive (but not by EAC)?

How is utilizing C2 pointers useful at all in this circumstance, even if the drive was 100% reliable in reporting them? I am assuming no AR verification, of course.

It seems that the C2 feature in EAC was meant to be used in place of re-reads, and not in conjunction with them. Otherwise, how are you notified if your drive reports a C2 error?


EDIT: In the case that a drive is 100% reliable in reporting C2 errors, the ability to utilize secure mode (and its synchronization check) without rereads would be one useful advantage. But it is my understanding that no drive is this accurate when it comes to C2 pointers.

This post has been edited by dgauze: May 17 2009, 07:14
Go to the top of the page
+Quote Post
greynol
post May 17 2009, 17:41
Post #12





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (dgauze @ May 16 2009, 22:54) *
So in it's current implementation, are you saying it serves no purpose to use C2 pointers when ripping CDs with EAC, unless if you were watching the screen while the ripping was taking place, to note whether C2 errors were reported by your drive (but not by EAC)?

I have no idea what caused you to arrive at this conclusion.

QUOTE (dgauze @ May 16 2009, 22:54) *
It seems that the C2 feature in EAC was meant to be used in place of re-reads, and not in conjunction with them.

That's correct.

QUOTE (dgauze @ May 16 2009, 22:54) *
But it is my understanding that no drive is this accurate when it comes to C2 pointers.

What makes you say this? It certainly isn't true.


--------------------
Placebophiles: put up or shut up!
Go to the top of the page
+Quote Post
dgauze
post May 17 2009, 17:55
Post #13





Group: Members
Posts: 45
Joined: 13-March 09
Member No.: 67901



Understood. Thanks again for clearing this up for me, greynol. One last question, do you know if the author has any plans to implement using C2 pointers on rereads any time in the future?
Go to the top of the page
+Quote Post
greynol
post May 17 2009, 18:07
Post #14





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



EAC used to be able to use C2 pointers during re-reads, but Andre felt that they didn't work well and removed that feature.

http://www.digital-inn.de/exact-audio-copy...html#post122541


--------------------
Placebophiles: put up or shut up!
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 30th July 2014 - 14:45