I am a new user and i try to understand how I can to write a perfect copy cd using EAC.
I have a cd that was ripping by another driver and I want to write this on a blank cd (include: .cue + EAC log = Used drive: HL-DT-STCD-RW GCE-8527B, Read offset correction: -6).
Can I make a perfect copy using the program EAC?
Is there a manual?
My driver is NEC DVD RW ND-4550A (Read offset correction: 48)
Can someone point me in the right direction?
Thanks for advance.

I don't use EAC, but these guides from the HA wiki should point you to the correct settings. Generally: use "Accurate Rip" to verify the integrity of your rips and you can be very sure to have a bit-perfect rip.

Your situation:
• You want to burn a perfect copy
• You have already ripped with proper offset correction, presumably securely (either using secure mode, or better still using ANY mode with Accurate Rip Confidence 2 or higher). If you need to check, CUETools can recheck against the AccurateRip database and report any offset discrepancy (which ought to be zero).
• You need to search the EAC guides (possibly including those at digital-inn.de) for combined read write offset correction, which will tell you how to set the write offset in EAC used in conjunction with something like burrn. Hopefully there's usually enough silent lead-in and lead-out for you to get away without over-reading and over-writing abilities in your optical drive. You could try first on a blank CD-RW and re-rip in EAC and compare the logs and their CRCs checksums against the original. You can then erase and retry until you get it right. There are also some write-offset calibration test images you could burn, IIRC. I've never done this stuff myself though as I'm happy to use digital files rather than red-book CD-Rs.

What's with the confidence of 2? I read a previous post of yours suggesting that if it's only one then you need a test CRC. In the case of there being a confidence of only one the only issue is if it was your submission, in which case a test CRC doesn't really add anything. Furthermore, secure mode is not necessary for a secure rip. Without AR verification one can use burst mode with test and copy and with matching CRCs, the rip will still be secure.

Now there have been cases where consistent errors presumably tied to a manufacturing defect combined with drives using a common chipset have populated the AR database. When taking this into consideration, I don't think that simply bumping the confidence level by only one is going to provide any significant level of warm fuzziness to those who are concerned about this inconvenient truth.

Back to the original point, simply burning with EAC and a properly configured write offset is really the way to go.

Ok. Thank you very much for your help guys....