And why does it take as long as 10 minutes to detect gaps on a few CDs, while most take just 5-10 seconds? Are the slow CDs written differently than the others? These CDs are unscratched and end up ripping at pretty much the same speed as those with very fast gap detection.

What method are you using? A, B or C?

Accurate or Secure?

What drive?

A
Secure
Plextor PX-712A

Has anyone tested the retrieval methods (A, B, C) and accuracy (Inaccurate, Accurate, Secure) against one another? Regardless of how much you care about gaps, I suspect the differences are insignificant in practical terms.

I'm not sure how every method works (IIRC one of them retrieves pre-gaps information from TOC, another extracts audio and finds digital silence, etc.) but from practical point of view the only difference between them is that some of them will work on particular drives while some won't, and in speed (A being the fastest but supported on least drives, and C being slowest but supported on most drives IIRC), actual results should be the same for all methods which work on your drive.
As for accuracy, the difference between accuracy levels lays in number of detections - in inaccurate mode every pre-gap is detected once, and in accurate/secure mode every pre-gap is detected several times (and the lenght which was detected most times is taken as actual one probably).

I'm just done ripping my collection of 400 CDs, and yes, it seems like certain discs (about every fifth or so) are of some kind of different "type", that requires much slower pregap detection when using mode A with the PX-712A.

Out of curiosity I compared some of them (20-30?) with mode B and/or C, and they were equal except for two cases. The differences were only by one and two single frames. Both discs were in excellent shape (At least no visual damage).

I then tested the two discs once more, and with exactly the same settings as the first time (Secure). Mode C now gave me a third result on the second disc. I also did a test with the CD/DVD-info in Plextools (PTP 2.24b).

---

The Beatles / "Revolver"

Mode A:
TRACK 14 AUDIO
INDEX 00 31:59:68

Mode C:
TRACK 14 AUDIO
INDEX 00 31:59:69

Plextools: same as mode C

---

Stereolab / "Mars Audiac Quintet"

Mode A:
TRACK 13 AUDIO
INDEX 00 53:02:15

Mode C:
TRACK 13 AUDIO
INDEX 00 53:02:17

Mode C, 2nd time:
TRACK 13 AUDIO
INDEX 00 53:02:16

Plextools: same as mode A

---

Mode B and C acted normally until reaching the certain "problem" tracks. Mode C then slowly detected the gap, while mode B seemed to hang completely and I decided to cancel.

According to Andre Wiethoff mode A might be a little more accurate in some cases:

http://www.digital-inn.de/showpost.php?p=6062&postcount=5