For those interested in DAE quality

Andre Wiethoff, the author of EAC, has started a new project on DAE quality. With the use of a selfmade test-cd and some analyse software you can test your own drives for its DAE capabilities.

I have tested 3 drives: Plextor Px-40ts, LiteOn Ltd-163 and LiteOn Ltr-40125s.

The results and some discussion can be found here on the EAC forum.

Feel free to ask questions, discuss the results or display your own results.

As I already said elsewhere this project has
been quite badly devised, so one should not
consider its results as gospel and certainly not
choose a drive based on them.

@spath

As I am still seriously interested in this subject and as I am also willing to learn, I will PM you about this subject. Hope you can clear up things for me!

Not only is this project badly devised, but the
thread you started (http://www.digital-inn.de/showthread.php?threadid=15838)
shows how confused Andre is about audio
extraction : don’t trust his stuff or his
explanations.

I’m struggling with his picture…what exactly does he mean with "one block? Does he mean an F2 Frame?

http://www.chipchapin.com/CDMedia/cdda7.php3
http://www.chipchapin.com/CDMedia/cdrom3.php3
http://www.ee.washington.edu/conselec/CE/kuhn/cdaudio2/95x7.htm

From Chip Chapin’s page:

“A Sector is the smallest addressable part of the information area that can be accessed independently” [ECMA-130 section 14]. A sector contains 98 frames. The frames themselves are the same as was described in the CD-DA section – each frame of 588 channel bits contains a “payload” of 192 data bits, or 24 data bytes. Thus there are 24*98 = 2352 data bytes in a sector.

In the drawing Pio2001 refers to a block which is the same as a frame.

The Plextor was always considered as the king of DAE… This turns out to be a myth

Having read your post there I’m now :confused: what to think about it. What’s your final opinion about it after all ?
I mean I too have both Plextor PX-W4012A and LTR-48125W,
so I’m really concened in finding out which one of these 2 is actually the best in DAEquality ?

Thanks in advance ! :smiley:

It seems there are many factors which might influence the graphs and outcome of these tests so my preliminary results might not be correct…

Yeah, I kinda understand that. But anyway, I’d really appreciate it if you could tell me
which one of the drivers is in your view a better audio reader/ DA extracter.

Thanks again !

I really can’t say, they are both excellent drives. I still have faith in Plextor so I would take the Plextor…

I don’t understand the post by BobHere in that thread:
If there are too many C2 errors in one frame, then it’s not possible to say which bytes are good or bad. The whole frame should then be considered bad.

I assume he is talking about E12 and E22 errors.

Here is an interesting

(http://www.digital-inn.de/showthread.php?s=&threadid=16313#post52038) from the EAC forum user BobHere

[QUOTE]That is a point of some discussion. There is no doubt that a correctly operating C2 error detection scheme will detect errors far more reliably than a double read system like EAC’s non C2 secure mode. The issue is how good is the C2 reporting of the drives.

Almost all testing I have seen on C2 reporting has been under extreme circumstances (long black mark tests). The performance of C2 reporting revealed by these extreme tests may be affected by a number of factors that may have nothing to do with its own accuracy and more to do with how the drive responds to such extreme conditions. Most everyone would abandon a rip that had this level of damage.

The LTR40125s gives very good (almost perfect) C2 reporting even with these extreme tests. I have previously done some extensive testing of C2 and I am satisfied that it works perfectly on all three of my drives under normal conditions. In the absence of definitive test results the final decision is up to you.

Regards,
Bob

Here is another

(http://www.digital-inn.de/showthread.php?s=&postid=52095#post52075).

[QUOTE]The behaviour around a large black mark is difficult to describe without knowing how the chipset has been designed, these things are not laid down in stone in the Red Book standards. It will probably deviate from what is predicted by CIRC and it may be this that is causing some drives to test poorly to C2 reporting using Andre’s test disc.

These are some questions that need answers to get a 100% understanding and be the God of DAE!

Does the drive mute?

If it mutes what is the trigger point to start and end a mute?

If it mutes, what algorithm does it use to prevent a click at the mute and demute points?

Does it have advanced interpolation and how does this work?

How does the drive resynch after a loss of synch? How many frame headers does it check, does it ‘backtrack’ after meeting its resych criteria (may relate to your question)?

Does the drive flag bad EFM symbols and mark them as erasures for the C1 decoder or does it pick the most likely answer and just hope.

What does the drive do with the data stream when it gets a HF optical output dropout (caused by the loss of return signal), does it act on this information to mark the data as erasures or pass it along as zeros or pass along bad data (at least at the transistion point)? At the transistion point does it mark individual symbols as erasures or does it mark the whole frame ( this may relate to a question you have asked about the beginning of the black mark). This dropout can also be detected in CIRC as both the audio data and parity data would be zeros which is not allowed in normal operation of the code (the parity symbols are inverted), I don’t know whether this behaviour is specificed in Red Book.

There are probably many more.

So many possible combinations.

This all sound very plausible to me. A lot of factors are mentioned which might confuse the outcome of the DAE quality test…

For more info on technical issues concerning audio cds search for the posts of BobHere or Pio2001

> For more info on technical issues concerning
> audio cds search for the posts of BobHere or
> Pio2001

This is hardly “info”, more opinions and
guesses. Generally speaking, if you’re looking
for technical accuracy, EAC forum is not a good
place. And for serious DAE quality tests stay
around, we’re working on it.

Hi,

Excuse me, is there any news regarding those DAE quality tests?

Cheers.

Originally posted by Leolo

Excuse me, is there any news regarding those DAE quality tests?

Good point ! :smiley:

> Excuse me, is there any news regarding those DAE quality tests?

ask G@M3FR3@K :slight_smile:

Hi spath,

I’m afraid that the DAE quality tests are time-consuming and G@M3FR3@K hasn’t had time to complete them.

But, could you please tell us some details about the tests?

Could we make them on our own using common software (like CDSpeed or DAEquality package by Andre) or is it necessary to have another utility?

Thanks a lot.
Regards.

These were tests with ABEX discs and CDSpeed
on various drives to measure C2 reporting
precision and audio extraction quality in general.
Unfortunately they never got finished, and only
someone with these ABEX discs could help.