Hmmm, I did some totally unscientific, but interesting to me, tests with EAC and AG once. I used new (ie no scratches at all) CDs as that is what I typically rip - along with the occasional CDs from the library (that frequently are quite scratched). With 3 different new CDs, AG was in average almost 5 times faster than EAC and the extracted wavs were identical, bit by bit. On a couple of scratched CDs the time difference was even larger, but in the AG rips there were some clicks and pops that didn't happen in the EAC rips...

So I use EAC for scratched disks and AG for clean/new CDs.

/Michael