I am in the process of archiving my cd collection to FLAC, I think my process is solid (outlined below). I have a question about Replay-Gain though. From what I have read, it appears that FLAC only uses the Meta-tag implementation of Reply-Gain rather than applying the settings at the Audio data level. Is this correct? If so, does it make sense to start encoding my files with the Replay-Gain meta-data? My understanding is that this will not affect the original audio data so it won't do any harm.

Also, does it make sense to embed the cuesheet if I'm making FLAC files of individual songs rather than entire cd images?

My Method

EAC V0.99 prebeta 5
Drive: Plextor PX-230A
Read mode: Secure
Utilize accurate stream: Yes
Defeat audio cache: Yes
Make use of C2 pointers: No

I'm using Accurate Rip with my drive, using test & copy, detecting gaps, appending to the logfile and creating a cuesheet.

Moderation: Removed unnecessary codebox.

1. Yes.
2. Yes, if you have a compatible player, and would like to use Replay Gain.
3. In my opinion, no; it would confuse players.

With regard to your EAC set-up, my only recommendation would be to confirm your options with our wiki (or wait for greynol to post).

According to spoon, this drive provides reliable C2 error information, so that setting can safely be enabled, other than that the settings are fine. Because it's a caching drive, I'd probably use burst mode and generate a test pass for tracks that cannot be verified by AccurateRip; at least with discs that appear to be in good condition. I could go on more about the use of test and copy in conjunction with secure mode and/or AccurateRip, but seeing that this was posted to the flac forum, I won't.

I fail to see any sense in embedding a cue sheet in each individual track.

thanks guys this is exactly what i was looking for.

if you could provide a few words about test and copy i'd appreciate it (but understand your point about being off-topic). Not needed with Secure Mode?

Test and copy basically IS a secure mode. Instead of reading a little bit then rereading and comparing, over and over, it reads an entire track, then reads it a second time and compares. It is much easier on the drive and accomplishes the same thing.

If you're using burst mode and can't get positive AccurateRip verification then a test pass is necessary to ensure that your rip was secure.

If you've told EAC to use C2 error information with a drive that is not known to provide reliable pointers and can't get positive AccurateRip verification then a test pass is necessary to ensure that your rip was secure.

If you're using secure mode without the use of C2 error information and get 100% track quality or re-reads aren't occurring until the very very end of the track then a test pass isn't necessary.

Of course there will always be exceptions such as this recent post might be demonstrating, but there are also exceptions where you can have errors despite having matching T&C CRCs.

Having AccurateRip verify a track is the most reliable method of ensuring that a track has been ripped accurately.