I'm using EAC to rip and encode my CDs as flac. I've set the command line argument for FLAC to set the cuesheet block based on the generated cuesheet file.

My problem is that one third of my CDs have invalid ISRC track numbers which causes the FLAC encoding to fail. I think despite the fact that the numbers on the original CD don't conform to the ISRC numbering standard, they should be able to be preserved regardless.

Does anyone know of a way to side-step flacs validation of ISRC numbers so they can be preserved in the cuesheet block?

(note, I know I can import a cuesheet as a tag, but I would be more satisfied if it were in the block also).

For reference, these are characters I've found in the ISRC numbers when FLAC encoding fails ;@[=?<. Also, some contain lowercase characters which FLAC does not like also.

The problem isn't that your CDs have invalid ISRC codes, the problem is that EAC cannot get them correctly from your drive. While I'm not sure whether Andre knows of the problem or plans on addressing it, I do know that people have been aware of the issue for several years.

Oh ok. I wondered that. Is there a work around? I don't believe there is a better ripper than EAC, but perhaps there's a tool to retrieve the correct ISRC and replace the values in the cuesheet?

The bad ISRCs you get are almost definitely not what is on the CDs. Some drives just cannot read ISRCs correctly.

I have two Samsung drives in my workstation, a plain DVD reader and a DVD writer. The writer reads ISRCs correctly while the reader does not due to a firmware bug.

So to fix your problem you can either get a drive that reads ISRCs correctly or disable ISRC reading in EAC (I suppose it has an option to disable it).

You may also want to try a different tool. IIRC, Alcohol 120% (for example) has been able to extract correct ISRC information in situations where EAC couldn't; at least with some drives.

Thanks for the info guys. My drive is an ASUS 2014L1T SATA with the latest firmware version (1.02).

I'll try Alcohol 120 but other than that I guess I'll head over to EAC forum to see what my options are.

For others reference, the cdrdao command line utility included with EAC is able to extract the correct ISRC number from your CD. Unfortunately, I have no idea how I'm going to automate this process

I never recieved any advice about this issue on the EAC forum. I've almost resolved myself to the fact that I have to skip on the ISRC numbers. Does this simply mean that I can't produce a clone of the CD in terms of data structure? In the big scheme of things, does ISRC matter?