Naive checksum questions

[HamLet]

Suppose that i do an accurate rip and then I burn a copy CD. It seems that if I look at the copy CD, the CD is recognized in the data base, but if I re-rip the copy I see completely different checksums. Does this mean there is something wrong in the process?

I understand how checksums work in general, irreducible polynomials, etc -- but I am not sure how the audio checksums are done.

Is there a process (burn software + parameters) that would preserve the checksums? Alternatively, is there a way to compute the checksums in a different way to get them back if the rip/burn were error-free? (for example if there is a need to shift by X samples, or something like this.)


Thanks! -- Ham

[LtData]

The offset of the CD you burn is affected by both the offset when it was ripped and the offset of when it was burned. The best way, I believe, to get checksums as close as possible is to use a cue/bin method that describes the exact spacing of the tracks to where the other tracks should have close checksums. The only way to get nearly exact checksums is to use cue/bin and have a CD burner that compensates for offsets, but I am not away of any burning software that does this.

[EliC]

I believe EAC's burner can do this.