So, i go through these phases where i rip all kinds of CDs. Some weeks go buy, and i don't rip anything. But when i start ripping CDs, i usually end up ripping 20+.

The end product that i'd like to build is a 4-10+ drive, headless, automated ripping machine. Walk up to a drive, eject the tray, insert disc. From there, i'd like to use Linux scripts to rip the CD to WAV, encode it as a single cue / flac, and also generate a 320Kbps MP3 for my ipod in a seperate directory.

The only thing i can't figure out is the second step: Ripping the CD, securely, from the command line. Currently, i rip CDs in EAC, but the process of RDPing into my windows box, launching two instances of EAC, pressing ALT-G to get track titles, and then selecting Action->Copy image and create cue sheet->compressed has really started to get old, especially with all of my linux scripting knowledge.

So i figured i'd ask the most knowledgable and helpful ripping forum i've used, to see if any of you guys can recommend the command line ripping portion. I want this to be just as secure as my EAC rips.

One last thing: Right now, i'm using some Plextor PX-760As in my main ripping machine. If the hardware makes a difference in your software for the linux side, just keep in mind that if you're going to recommend drives, i have to be able to get them new easily.

Looking forward to the flames that will ensue. :D

I would use cdparanoia + lame for this.

I'm not too sure about looking up the track details on Freedb, but there are scripts which have such things built in (some are listed here but all the appropriate links seem to be down).

From everything I've read from reputable people, cdparanoia does not properly handle caching drives out of the box and the Plextor PX-760A is definitely a caching drive.

Whether this can actually be fixed with a patch is questionable.

Clearly there must be a solution for this problem that EAC and others have implemented. Why is this issue such a hangup for cdparanoia? It seems like the authoritative conclusion is that cdparanoia is a sub-par ripping program, yet no-one feels the need to address the issue. Is there a lack of community support for fixing the program, some technical detail, or merely a lack of resources?

That being said, if there is a viable solution out there, I would be willing to spend some time on it.

It's not a subpar ripping program in my opinion. I have used it on multiple CD's even ones that were scratched and it did a pretty good interpolating the erros (EAC on the other hand is a nightmare to configure and run with I don't have time for that shit no offense). Some folks in the EAC community don't like it due to the fact that it does not have C2 error correction and won't work on drives that cache audio. There is still some FUD about it flying around out there, because the internals of it have yet to be well documented, but like you I agree that whatever the problem is it can be fixed.

Does cdda2wav also suffer from the caching drives problem?

If the box doesn't have to be linux-based, why not look at the dbpoweramp R13 Batch Ripper, which can control as many drives in parallel as you can hook up to a windows box?

-brendan

Wow, i didn't know there was such a beast. I'll google it today. Is this as secure as EAC for ripping music with a caching drive?



I'm not married to the PX-760A. If i could get 10 NON-caching drives, new, for less than $100 a piece, that would be okay, too.

Yeah, different from EAC's method but secure nevertheless. It's shareware tho, the product is called "dBpoweramp Reference R13" and it costs 36$ ( price of 2-3 CDs ), after buying that you can download the batch ripper.

http://reg.dbpoweramp.com/store/product.as...71&P_ID=129

Yes, at least as secure as EAC. With drives that support C2, such as the the 760, it may be more secure. With the 760 you can also use the FUA command to stop the drive from caching.

Have you tested a 10x drive system? most people go for 6x drives maximum on one system, if they are used simultaneously.

Where do you think the limitations lie? Transcoding CPU use? IO port (IDE/SATA/Firewire/USB) congestion? Hard disk IO? Memory constraints?

I ask only because I'm thinking of moving to an eight-drive setup on one of my robots...

-brendan

Depends on the design, the easiest way to put 10x drives on a system is USB, but the transfer limit of USB can be reached on 5 drives (I must admit I have never tried seperate USB cards).

Looking at the maths, if you were lucky (?unlucky in this design) and had a drive which could rip x40 on the outside then transfer is:

176KB x 40 = 7MB

x10 drives and the possibility is there for 70 MB / second transfer

If I was forced to create a 10x drive system it would be quad core using SATA, ripping to something like lossless then 40MB a second, so a modern drive should be able to handle that.

It is often easier and less expensive to have 2x systems with 5 drives each.

You really should stop saying things that aren't true.

If I present you with evidence contradicting this will you finally stop it?