Ripping CDs/Sync error

[Dob]

I have a few CDs that play fine and appear to be scratch free, but when I try to rip the tracks in EAC I get lots of errors. Some are sync errors, others are "missing sample" errors, sometimes EAC just gives up. In cases like this, I've tried to use other modes besides "secure", which does enable me to rip the track, but in all cases the track is loaded with (sometimes) hundreds of digital clicks.

When zooming in using Cool Edit, I can see that it is just one sample that is out of place and is causing each click. Often the clicks are primarily confined to one channel.

Can anyone help?

[Xenu]

It could be your drive. What CDs are in question here?

[Dob]

Up until recently, the only CDs that gave me trouble are the SH mastered "Rock Of The 70's". However, I recently acquired a MFSL "Surfer Girl/Surfin' U.S.A." that gave me the same trouble.

With the '70's CDs, every EAC extraction would give me a different combination of digital clicks, almost as if they were somewhat random. I know there's hundreds of them, because I did an inverse compare of two different extractions and it sounded like I was making a batch of popcorn.
The clicks are fairly quiet; during loud passages they're very difficult to hear, but if there's enough of them it causes audible distortion.

I should add that some tracks are worse than others. On the MFSL, the worst tracks are in the middle, with the '70's CDs the worst ones seem to be towards the end. But some tracks on all of them seem to extract fine; however, they may have some clicks that I'm not hearing.

I have a Pacific Digital burner and a Western Digital hard drive.

[krabapple]

I've had this happen on a few discs. One thing to check for is a tiny scratch on the *label* side. But on some discs I haven't been able to find any reason for the errors. Luckily they are quite rare (one of them is disc one of Prince's 'Sign O' the Times', for some reason...a comparatively rather old CD n my collection). I rip these rare problem tracks using a different tool (Stomp) that muscles its way right through the errors, and I don't hear any digi-clicks, but then again, I don't go listening for them either ;>

[J_Partyka]

It could be your drive.

I second this ... Not long ago, I got a copy of the Eels' CD Shootenanny! that gave me this kind of trouble. I tried extracting using the CD-ROM drive in my 4-year-old iMac, and also my more modern external Yamaha CD-RW drive. No dice.

Then I brought the disc to work and tried using EAC on some Windows PCs. I tried three different drives before I found one with which EAC was able to give me a flawless extraction (it was the internal drive on a Dell PowerEdge server ... the extraction did take quite a long time, but it finally worked).

So I would suggest trying different drives, if possible.

I also had a copy of Wilco's Summerteeth that wouldn't extract the last track properly on any of these drives, despite multiple attempts. That disc did have some marks on the data side. I eventually gave up and bought another copy of the CD.

[Dob]

But on some discs I haven't been able to find any reason for the errors. Luckily they are quite rare...

Yes, luckily. I've also been lucky in that, for the 70's CDs, the tracks that I really wanted to extract (the better sounding ones) have been OK, except Grand Funk's We're An American Band. And even that one wasn't too bad...but I fixed it as best as I could, by hand. Which involved scrolling through the entire track, zoomed in closely enough so that I could see the individual samples. Took about 3 hours. I must've been feeling particularly stubborn that day!

I rip these rare problem tracks using a different tool (Stomp) that muscles its way right through the errors, and I don't hear any digi-clicks, but then again, I don't go listening for them either ;>

The burst mode in EAC will "muscle through" as well...but if you have to resort to that, the extraction will probably be crap anyway. It is useful, however, if you just want to get past one problem section.

Speaking of the MFSL Beach Boys, I was using it to make a "best sounding" version of those two albums (plus others). Some of my other sources included a needle drop of the MFSL LP, the Japan "Pastmasters" CP21 CDs (which SH says are flat transfers, and I believe it), the HDCD Linett CDs, and a Japan TOCP remaster (from 1997). For some of the mono tracks (from Surfin' Safari), I was surprised to see that the tracks from the Linetts were not in sync - the left channel was one sample ahead of the right channel. As soon as I saw this I knew it would drive me crazy, even IF it is inaudible (biased perceptions and such), so I copied one channel and used that to make a new, in sync, two channel mono file. I am curious as to why that happened...does it have something to do with extracting HDCD encoded tracks, perhaps?

[Dob]

Then I brought the disc to work and tried using EAC on some Windows PCs. I tried three different drives before I found one with which EAC was able to give me a flawless extraction (it was the internal drive on a Dell PowerEdge server ... the extraction did take quite a long time, but it finally worked).

So I would suggest trying different drives, if possible.

Thanks, Jeff. I'll keep that in mind, even though at present it is difficult for me to try that.

But, doggone it, I hate hearing stuff like that. Because just when I think I sorta understand how things work and I think I'm standing on solid ground, I've been knocked on my ass again. Why on earth would there be such a difference between drives - specifically, the laser assembly (which I imagine would be the culprit)? Aren't they all made by the same three (or whatever) Asian factories?

Oh well, I guess I should be used to this kind of stuff by now...

[lukpac]

Random replies:

hard to read discs/drives - I've had CD-Rs that won't even show up in one drive, but extract flawlessly in another. No clue why. Same thing with CD players.

EAC burst mode - I have had times when secure mode is really slow and gives lots of errors (and clicks), but burst mode (I think) gives a rip that sounds fine. Go figure.

channels one sample off - I've seen this a few times, notably Buddy Holly FTOMT (*some* pressings) and Quadrophenia (original mix - don't know which is correct, but the Polydor and MCA have a one sample offset in one channel). As an aside, I used to have a drive (TEAC) that would reverse the channels, and probably create a one sample offset. Something about reading the wrong sample first. A firmware update fixed that.

[krabapple]

The burst mode in EAC will "muscle through" as well...but if you have to resort to that, the extraction will probably be crap anyway. It is useful, however, if you just want to get past one problem section.

Less than theoretically perfect, for sure..Crap, well, IMO that's too extreme a dismissal, like something I'd see on SHtv. I certainly wouldn't bet the farm that I could always tell a perfect rip from the 'muscle through' rip, or either from the source, on audible evidence alone.

[Dob]

Less than theoretically perfect, for sure..Crap, well, IMO that's too extreme a dismissal, like something I'd see on SHtv. I certainly wouldn't bet the farm that I could always tell a perfect rip from the 'muscle through' rip, or either from the source, on audible evidence alone.

Sorry, I should have been more clear. For these particular CDs it didn't give me a better copy...in fact, it sounded like it had even more clicks (i.e., crap). However, I was able to extract an otherwise OK track that the secure mode gave up on, because it hit a particularly bad spot that the burst mode went through.

I didn't mean to imply that I can hear the difference between a secure extraction and a burst extraction, much less that one method is fine and the other is "crap".

...like something I'd see on SHtv.

That was a low blow, Krab...

[Dob]

channels one sample off - I've seen this a few times, notably Buddy Holly FTOMT (*some* pressings) and Quadrophenia (original mix - don't know which is correct, but the Polydor and MCA have a one sample offset in one channel).

How can I spot this one sample offset if the tracks are in stereo? And do you think this offset is audible?

[lukpac]

Don't know, and I doubt it.

[Xenu]

"Muscle through" rips can easily give you an extract extraction. EAC's "secure mode" can often be about peace of mind and little else.

For example, my college "player" copies of the Stones' London catalogue was made two days before I had to leave, and thus consists of buffered-burst Audiocatalyst audio. I recently did a digital comparison to audio ripped in secure mode from the original source disc (this on a far better drive), and they null-tested. So in that particular case, my drive was bit-accurate.

Actually, my CD drive in my old computer did non-secure ripping fairly well. When it would encounter an error, however, it would tend to skip wildly, so I'd always have to check to see whether the song "ended" in the correct place. If it didn't, I'd go back and do a syncronized rip on A/C.

[Dob]

"Muscle through" rips can easily give you an extract extraction. EAC's "secure mode" can often be about peace of mind and little else.

My (brief) experience with the different EAC extraction modes would lead me to agree.

[Dob]

...I doubt it.

Even if it was audible, I know of no way to fix the offset for stereo.