disk update - building disks using host computer?

OK, biting the bullet and upgrading to solid state disks. My collection is now 140G and still growing, so 2 x 128G it is.

To minimize the downtime of the empeg, I am wondering if it is possible to build (partition and pre-populate) the disks on a Linux host computer instead of running the builder on the empeg?

You can, but then run the problem of the host having a (MUCH) newer Linux kernel, which likely turns on filesystem flags that the empeg doesn't understand.

Or something like that -- been a while since I've done that, and I *always* used my host PC for the task because the IDE is so much faster than on the empeg.

Oh.. ya know, the way to do this is to run the builder image on the empeg first, then pull the drives and take them to a PC for the time-consuming data copy phase. The empeg can then run on the old drives while this happens.

Yes, that thing would work. Make sure to use Mark Lord's builders with the drives in the player, remember to put the player software back onto the player and the disks while they are still in the player after the build (using Mark's installer), and *then* hook the disks up to the PC to copy the files.

Of course, you can't just "copy the files" as raw MP3s; you have to use a FID generator program on the PC and then copy the resulting FID files to the player's drives.

Or just hook up the old empeg drive and copy them to the new one.

Oh. Right. Yeah, if you've already got built drives full of FIDs, yeah, that'll work. You only need the FID generator if you're copying the original MP3s from the PC to the player's drives.

Well, yes, I am using the FID builder + rsync anyway for updating the music, so that part is easy. I had been thinking about the "build disks in empeg, move back to host to copy music" strategy, but was hoping to avoid the back-and-forth swapping of the disks. Anyway, thanks for the advice!

Yeah, though the back-and-forth is a hassle, the only way to reliably create correctly-built disks is with Mark's builder program, and that only runs on the empeg.

OK< better clean up more space on the lab desk then...