The Philips hdd120 recording audio jukebox is an mp3 (and wma) player with a USB connection.
The software (DMM) that comes with the player is MS only, not even the MAC is supported. According to the manual, only mp3 files copied using the DMM will be played.
This document details the set-up I use to, as far as possible, work under Linux.
Packages and configuration
The hdd120 handles each track individually and without the id3 tags all tracks would appear in one big list. With the inclusion of id3 tags, you can peruse the list by artist, album and genre. Therefore, the process of ripping of tracks should include the addition of id3 tags.
For ripping, the latest grip package is used as this integrates the cddb lookups and the inclusion of all the required id3 tags seamlessly. This requires an Internet link to a suitable cddb server or the actual datafiles downloaded from freecddb. I then store these on my laptop (Linux partition).
mp3 transfer
To transfer the mp3s to the player, I copy the mp3s from Linux to a vfat partition and reboot into XP and use the provided software (DMM). I am working on reducing the reliance upon DMM …
Progress on moving to linux
Connections and transfers simple:-
- Re-build kernel with scsi and USB mass storage support.
- Connect and manually power on HDD.
- Check if device found 'cat /proc/scsi/scsi' and determine scsi dev name (e.g. /dev/sda1)
- mount the device with 'mount -t vfat /dev/sda1 /mnt/mp3'
So far I have determined that there are no alterations to the mp3 files themselves. They are identical (HDD and PC). The files are renamed to Therefore, the whole problem comes down to determing what the 'Philips HDD naming convention' is.
I am surmising, but I suspect the naming convention is the key as it looks suspiciously like a hash or checksum value. Thus the DMM software calculates an hash value from the mp3 file and names it that value when saved on the HDD. The HDD would then only play those mp3 files whose file name matches the hash value calculated by the HDD itself.
Therefore, I 'just' need to determine the hashing algorithm used by the HDD and DMM … It could also seed the hash value with a unique number (id) from the HDD to tie the mp3 track to that specific HDD device … (To check this I would need access to another HDD or someone to create a small mp3, use DMM to install it on their HDD, then use linux to get the installed file and send it to me …)I found a site with details of the 'Philips HDD naming convention' where it is used to create a windows utility which bypasses the DMM. So all I have to do is code it up for Linux ! …