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: Yet another EAC question (Read 5772 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Yet another EAC question

I did a little experimenting just to verify that different methods of extracting audio and creating cue sheets in EAC would give me the same result in the end. All tests were done with the same original CD, and I tried different things like "copy image & create cue sheet uncompressed", "copy image & create cue sheet compressed", both as a single wav file and as multiple wav files. I then burnt a disc for all different methods I had used, and then ripped these copies.

I checked every single rip against every other, "using the compare wav" function i EAC. Every comparison failed, I got the "different samples" reported for somewhere between 50 and 100 positions in every case.

Even when I ripped the original cd twice using the exact same method, and having no errors reported, comparing the resulting wav files also resulted in "different samples". Now I got nervous, and just to make sure my system wasn't completely messed up, I took one wav file and copied it to another drive, and then compared the two copies. Thankfully, they compared ok...

When I first set up EAC, i believe I did everything correctly. I have a Maxtor 6Y160PO, "drive caches audio data" is unchecked, "use C2 error information for error correction" is also unchecked. (This is what EAC comes up with when I use "Detect Read Features".)

Why is this happening?

Sorry if I am not providing enough info about my setup etc, but I am hoping there might be a simple, general explanation to this. And of course, sorry if this has been dealt with before, but I didn't find anything when I searched the forum.

Thanks for any help!

Von

Yet another EAC question

Reply #1 –
EAC is not configured correctly.
If you're not sure, enable "drive caches audio". You'll only lose speed but not accuracy.


Anyway, or you have a very bad reader or the disc is a bit scratched.
Probably it's a quite old CD-ROM, isn't it? (if it's a modern writer, then maybe it's damaged)

Yet another EAC question

Reply #2 –
Quote
When I first set up EAC, i believe I did everything correctly. I have a Maxtor 6Y160PO, "drive caches audio data" is unchecked, "use C2 error information for error correction" is also unchecked. (This is what EAC comes up with when I use "Detect Read Features".)

It doesn't hurt to try with "drive caches audio data" enabled. This will tell EAC to circumvent it (if there is one).

Most drives DO cache, so my guess is that is the problem. Do you get sync or read errors on this disc? If so, the problem is that this disc is unrecoverable.

Yet another EAC question

Reply #3 –
Thank you both,
I am trying now with "drive caches audio data" checked.

I'll be back in a little while to tell you how it went.

Von

Yet another EAC question

Reply #4 –
You won't get same results if the drive's offset correcton isn't set (read+write or combined), and copy must be made from single wav file (if using EAC for recording). But if you can't get same results in ripping the same CD twice...? Well, I would open both files with CoolEdit and save same portions of the music to eliminate the offset issue. If they are equal now - you probably have drive with offset problem. But, if not - your drive could be damaged (assuming you have configured EAC correctly and your drive can work with it) or incapable of audio extraction. Need a gun?

Yet another EAC question

Reply #5 –
Quote
You won't get same results if the drive's offset correcton isn't set (read+write or combined), and copy must be made from single wav file (if using EAC for recording). But if you can't get same results in ripping the same CD twice...? Well, I would open both files with CoolEdit and save same portions of the music to eliminate the offset issue. If they are equal now - you probably have drive with offset problem. But, if not - your drive could be damaged (assuming you have configured EAC correctly and your drive can work with it) or incapable of audio extraction. Need a gun?

You'll still get the same results each time without setting the offset, it just won't necessarily be the same file you would get from a different drive (with a different offset).

Yet another EAC question

Reply #6 –
By Jebus:
Quote
You'll still get the same results each time without setting the offset, it just won't necessarily be the same file you would get from a different drive (with a different offset).

Sorry, my bad! Didn't think EAC can detect sample difference. And I knew it uses FC for comparing (?!) >>> I'll crawl back into my hole... 

Yet another EAC question

Reply #7 –
Hmm.

Well, I did check the box for "drive caches audio data". It seems like it helped, but when I compare the extracted wav from the original cd with the wav extracted from a copy, I still get "different samples" for many positions.

I am saying that it seems to have helped, because "different samples" are now reported for fewer positions than before, there were about five times as many errors before.

This is starting to bother me now...      Any ideas on further action?

The read and write offsets should be correctly set. Besides, wouldn't the wrong settings here make the files compare even worse, more like two totally different wav files?

Perhaps I should just be a bit more pragmatic about the whole thing. No! I want bit perfect copies, please help me...!

Yet another EAC question

Reply #8 –
Quote
Besides, wouldn't the wrong settings here make the files compare even worse, more like two totally different wav files?


Not necessarily.
The software can't change what the drive reads.
"Drive caches audio" setting is to correctly report errors. (Slower speed in this mode might produce less reading errors).


It would be very nice to say what drives are you using to rip.
It can be helpful because maybe someone has the same drive as you, and knows how to solve its problems, and it's interesting for the rest of us in order to know which drives perform better DAE.

Ripping audio is a tricky issue and some years ago, there were a lot of drives that couldn't rip audio correctly even on perfect discs: different data at each read operation on the same sector.
There were other drives that read correct data most of the time but failed sometimes easily (usually not audible).

Here you have a list of drives and the column "Correct" shows if the drive can read audio "correctly":
http://www.feurio.com/English/cd_roms/fram...e_list_ide.html


Another explanation might be that something is causing data corruption in your computer, although unlikely. (What's you mainboard/chipset, IDE controllers...?)


Have you tried to compare results with 2 different drives?

please, tell us your hardware.

Yet another EAC question

Reply #9 –
When you rip the copy, do the red lights of error correction light up ? And how often ?

Yet another EAC question

Reply #10 –
Minix; I have a Dell Dimension 4500 computer, less than a year old, with a Pentium 4 2.4 GHz CPU. I'm afraid I don't know straight away how to dig up info about what motherboard, IDE controllers etc I have...  Please tell me how to check that, and I will.

My CD-ROM drive is a Samsung SC-148C. (Sorry about my previous post saying it was a Maxtor, obviously that is one of my hard drives.)

I could try ripping with my Philips DVD+RW-D28, but I remember reading somewhere that it's generally not recommended using a burner for ripping, as its heavier head tends to suffer more from all the movement. (?)

Pio2001; I just tried again to rip one of my copies twice. The error correction lights did not light up a single time, and EAC reported no errors. I then compared the two wav files, and EAC reported different samples in 18 locations. Each location is only one sample, as opposed to an extended period of time.

I'm curious about all this. On many occasions, with scratched discs, EAC reports errors. Sometimes these are audible, sometimes I'm perfectly happy with how it sounds. Anyway, I have always believed that when EAC says no errors occured, the rip was perfect. Why is it that I am getting errors that are NOT reported?

Thanks for your help!

Von

Yet another EAC question

Reply #11 –
Quote
I could try ripping with my Philips DVD+RW-D28, but I remember reading somewhere that it's generally not recommended using a burner for ripping, as its heavier head tends to suffer more from all the movement. (?)



You'll have to do it.
That's the way to discard general data corruption in your PC.


Are the discs you're ripping scratched?
Use a perfect disc to rip with both drives and compare the WAVs.

If the Samsung shows errors while the Philips not, then you know where's the problem.
Your drive is not in the list I said, but you can see there are several Samsung CD-ROMs similar to yours that are quite poor at ripping audio.


Well, I wouldn't use a DVD+RW to rip audio normally because of the price, but there's no problem to rip with it sometimes.


why are there not reported errors?
I've got no idea. Maybe you have to disable "Accurate Stream" ? (terrible reader?). The rest of the settings seem correct: C2 disabled and drive caches enabled. Maybe the drive is returning the same wrong data twice? Is that possible?

Feurio has a reading test. Feurio CD-Manager -> Program Parameters -> Test Device and select the "read test" only. It will compare ten readings of the same zone, and it will show if it has evident problems with head positioning (jitter).


The manual of your computer should say what your chipset you have (probably a Intel i845E or similar), but I don't think that's the problem...

Yet another EAC question

Reply #12 –
OK, I will try that tomorrow (almost bedtime in Norway), and tell you how it goes.

Thanks.

Yet another EAC question

Reply #13 –
OK folks,
I have tested and come to the conclusion that my Samsung CD drive is not working properly. Repeated rips with my Philips DVD+RW-D28 as well as rips of copies compared perfectly well in EAC.

Now that I know this, I will probably be using (after careful testing) another CD or DVD drive on another computer I have.

The bad news is I have no idea if every EAC rip I have ever done is corrupted, or if it's only the last 10-20, or... You get the picture. How do you guys make sure that your rips are perfect? Do you burn and rip a copy and do a compare every time? I guess not. Every 10 times, perhaps? I think I will go for something like that.

Maybe my CD drive has always been bad, maybe it's gone bad from the stress involved in ripping copy protected discs. I'll never know. But at least I know now that "No errors occured" doesn't necessarily mean the rip was perfect.

coy-itys; you are saying that in order for the rip of a copy to be identical to the rip of the original, one must use single wav file with cue sheet? Is this because of offset issues?

I was considering starting to do multiple wav files with cue sheet, but in that case, I'll stick with single wav files.

Von

Yet another EAC question

Reply #14 –
With new drives, I always run Feurio's tests. If they are OK, I check my copies against original. I compared wavs for the 10 - 20 first copies I made, then gave up and trusted my drives.
I also checked several burned CDs with Nero CD Speed. I found C2 errors on HiSpace Black, so I switched to HiSpace gold.

Yet another EAC question

Reply #15 –
Quote
I have tested and come to the conclusion that my Samsung CD drive is not working properly.

All amsung products are crap!!! Their CD-Rs are the worst I've ever seen. Just because you buy a branded PC like Dell, it doesn't mean they use the best components. Many people who buy branded computers are non tech-savvy people. Companies like Dell know this fact and they make good use of it. Try building your own PC the next time. It's easy. I know this well because I worked for Dell a long time ago...

Buy a good drive like Plextor. If Plextor is not available, try an Asus. Asus makes pretty good drives. Since Plextors are not available in my country, I've been using an Asus all these years. I am very happy with Asus.

Have a nice day.

Yet another EAC question

Reply #16 –
What it will possibly be is the drive is not accurate stream, and is returning large jitters that EAC is not able to match with the previous frame.

Yet another EAC question

Reply #17 –
I'm guessing it's something more serious that whether the first and last track lead-out and lead-in have to be padded with zeroes (if the drive can't over-read into them). Spoon's jitter suggestion (no accurate stream) could well be right.

Yet another EAC question

Reply #18 –
But just to get one thing straight once and for all: Can I safely say that my rip is perfect if I burn a copy, rip the copy, and compare the two waves with EAC and get a match? (Of course I'm assuming that no errors occured during the ripping, or else the exact same error occuring twice could theoretically result in a perfect match.)

Please don't tell me there is even more to it than this...! (Although I have a feeling there is...) 

Yet another EAC question

Reply #19 –
Quote
Can I safely say that my rip is perfect if I burn a copy, rip the copy, and compare the two waves with EAC and get a match?


Yes.
You can even say that your rip is perfect without burning, just comparing 2 ripped WAVs.
OK, maybe the drive reports the same wrong data twice, but I think it's not common.


The question is why isn't EAC reporting errors?

Could you try to disable "Accurate Stream"? The drive supports Accurate Stream almost surely, but maybe the implementation isn't very good?

Is it possible that flushing cache doesn't work for all drives?


Anyway, if you were ripping perfect discs, that drive is really bad... there shouldn't be any errors with good discs and good drives.


Quote
How do you guys make sure that your rips are perfect?

Using good hardware   
Buy a drive that reports C2 errors. You'll have more possibilities configuring EAC or even using Feurio to report the C2 errors.
There's a very recent thread about good drives:
http://www.hydrogenaudio.org/forums/index....showtopic=11189
LiteOn or LG burners are very cheap and good at ripping difficult discs.

Yet another EAC question

Reply #20 –
IIRC, there was also a Samsung drive that mutes audio for a while near the end of every track. This has been quite a mess to point out in EAC's forum.

Yet another EAC question

Reply #21 –
OK, thanks all of you! I'll start using another drive, and check to make sure things are working properly.

Von

Yet another EAC question

Reply #22 –
By Von:
Quote
coy-itys; you are saying that in order for the rip of a copy to be identical to the rip of the original, one must use single wav file with cue sheet? Is this because of offset issues?

No, it's like this:

- due to the structure of the CD directory, the start of a track always has to be a sector start. A CD-Audio sector contains 588 samples. Well, it's rather unlikely that your wavs are all multiple of 588 samples. So, while recording, EAC will insert exactly as many null samples at the end of each wave file as are required to get a multiple of the CD sector size. And that's why you wouldn't get a perfect copy.

>>> Feurio! has an option "Don't insert pauses between tracks - round track markers" which handles that. You could use this option with "combined read/write sample offset correction" set in EAC in order to get a same copy. Or, you can do multiple wav files, but with cue sheet for single wav file - when you wish to burn a CD, just combine wavs together.

Hope that helped!   


(Parts taken from: Feurio! CD-Writer online help  © by Fangmeier Systemprogrammierung)

Yet another EAC question

Reply #23 –
Quote
due to the structure of the CD directory, the start of a track always has to be a sector start. A CD-Audio sector contains 588 samples. Well, it's rather unlikely that your wavs are all multiple of 588 samples. So, while recording, EAC will insert exactly as many null samples at the end of each wave file as are required to get a multiple of the CD sector size. And that's why you wouldn't get a perfect copy.


No, coy-itys.
All WAVs you rip from a CD with any program will be multiple of 588 samples.
If that wasn't the case, then most burning programs wouldn't be able to burn those WAVs without small pauses between tracks.
(This is only an issue with MP3 files or normal WAVs not ripped from CD, or edited in length).

The only problem are offsets (and reading errors). These WAVs may vary in the point where they start: they can be "displaced".
http://www.feurio.com/English/faq/faq_exactcopy.shtml

It's the same if you make a copy from multiple tracks or only one track.
The audio data will be the same. Offsets will affect the same in both cases.

Yet another EAC question

Reply #24 –
Enable "sync between tracks" in order to ensure this for non-accurate streamed drives too.