I've read the "truth about offests," here. According to the accurate rip datatbase, my drive, the Pioneer dvr-110d has a sample offset of 48. I've verified this with a CD from my collection that was stamped at the same time as a "known key." The write offset of my drive is 18 according to this thread, which I have also verified independantly (via EAC).

So... my question boils down to this:
In EAC, is my "Read sample offset correction vaule" "+48"/ or "-48?"
and
In EAC, is my Write sample offest "+18" or "-18?"

My head is spinning from reading tutorials and I figured out most everything else except for the damn +/- sign, which is for the moment the bane of my existence.

The write sample offset would be +18. If accurate rip contains the drive offsets, then the offset correction in your case would be -48 - if the DB contains offset corrections, then you can simply take that value and use it in EAC (+48).

Even though it says "sample offset", the AccurateRip database actually lists the read sample offset correction value. Your drives read offset is negative, so you would enter the positive value, as listed in the database, into EAC.

This site does the same thing.

[edit] +18 would seem to be the write offset correction
[edit.2] offset, not offset correction, as Sebastian points out below

Well, the database at accurate rip says that their list contains submitted drive read offsets. So that would mean that my Read Sample Offset Correction Value is -48. Can anyone confirm this to be true for the pioneer drive in question?

Shouldda previewed cosmo... This is the most confusing thing ever. The answer is always hidden in the damn semantics, which are totally incosistent at best.

And to be completely clear, the vaule I'm now entering into the little box "Read Sample Offset Correction Value" is +48 and +18 for the write offset correction.

What was wrong in what I told you?

And by the way, the write offset is +18. The write offset correction would be -18 but EAC asks for the write offset and not for the write offset correction. The combined read/write offset is +66.

You're correct, I forgot about that difference in EAC.

No you weren't wrong... my fault. You were exactly right, the problem seems to be in the wording of "what's in the database" vs "what accurate rip says is in the database." In which case, the accurate rip site is wrong (or very poorly worded).

The read offset for your drive is -48. Your drive reads 48 samples too soon.

The read offset correction for your drive is +48. EAC will give you an output that is delayed by 48 samples.

When we say samples we really mean stereo (L+R) sample pairs. <-- EDIT: nevermind this.

Your write samples offset is +18. This means that your drive writes 18 samples late.

Got it guys, thanks for the help and sorry for the confusion. I am no longer confused. BTW, this means that I did it right the first time. I was quite worried that I'd have to re-rip the first half of my collection...

Sorry, for butting in.
Am I right in saying that EAC asks for the read offset correction, but only the write offset value ?

Yes.

Question for greynol and Cosmo: are you 100% sure that values for Sample Offset in Accurate Rip database are actually values for Offset Correction?

Second, if my drive (ASUS CRW-5224A, with read offset -6, according to what Cosmo and graynol said for Accurate Rip database) can only overread into Lead-In, should I confirm option Overread into Lead-In and Lead-Out in EAC?

Yes, I'm 100% sure. And just to make sure I am clear, the actual offset of the drive will have the sign reversed. Your ASUS drive, according to the AccurateRip offset database requires an offset correction of +6 samples. This means that the drive has an offset of -6 samples which, IOW means that it extracts data 6 samples too soon and requires a delay by 6 samples in order to conform to the standard.

No. Overreading for this drive would require that it can read into the Lead-Out.

Thanks!

oops i have posted to the wrong thread