[QUOTE=aerotight31;1270774]I tried that without it working. I have, however, got it working somehow. QUOTE]
aerotight31 kindly redirected me from another thread to here (thank you!!), so I'm going to copy/paste my problem statement for future readers of this thread, and then tell you about my attempts to fix it.
Quoting my problem statement:
------
I'm on a recent (Dec08-era) RoadRunner and DigitalFX4, with Winamp 5.541. I can load .m3u and play playlists fine (via the AudioLoadPlaylist.skin UI), including navigating around, shuffle, etc.
However, once I round-trip through a normal RoadRunner Exit or a system hibernate, playlist functionality is busted in weird ways.
- After a graceful RoadRunner Exit & restart (via the exit menu), the playlist window will show songs from Playlist A, with the correct song# and position selected (eg, #5 in the playlist, 30s into the song) - however the song that is actually playing (audio + what's showing from the ID3 tag display) is #5 from Playlist B - a Playlist I had loaded earlier, but I haven't figured out what's special about PlaylistB vs Playlists C, D, & E which were also playing..... Even if I shuffle around several other playlists and then restart, PlaylistB's content is consistently chosen to override what's actually showing in the playlist window. (Suspect something related to winamp.m3u, but I've tried clobbering that whole folder in AppData...and it still persists in choosing PlaylistB! Must be cached somewhere else too...?).
- After I return from hibernate, the playlist playback is completely broken - the playlist window shows it is skipping around among songs, spending about a half-second on each before moving on to a new song. No audio is heard at all. It'll do this indefinitely. The system hibernate is invoked when I turn my car off (and the Carnetix handles the graceful shutdown) - not via the DigitalFX Exit menu.
If I am instead playing a batch of songs selected via the file browser, I am able to resume playback of that same batch of songs (in the middle of the song that happened to be playing at the time) just perfectly from either a restart or a hibernate/resume.
I've tried several different tools/editors for generating the .m3u files, and get the same behavior with each of them. (Incidentally, I've also tried making .WPL playlists work, and get to a similar point - most everything works perfectly until I return from hibernation, at which point *Other* Evil Behavior kicks in...)
I've been tinkering on and off with RoadRunner and DigitalFX for months now and these symptoms have been consistent regardless of software versions. I've even completely flattened the system and reinstalled everything at one point (including changing from VistaSP1 to XPSP3), and the symptom still persists! I've been unable to turn up any reports of similar issues.
I've poked around the skins enough to know about the ReloadPlaylist and the fact that there's some communication that happens there...but I'd assume everything is supposed to work correctly Out Of The Box, right? I can't figure out what I'm doing 'nonstandard' that gives me broken behavior whilst the rest of the world is happily resuming in the middle of their playlist playback!
Help? Any ideas would be MOST appreciated! Thanks,
----- //End Problem Statement
My experience applying the fixes suggested in this thread:
I tried hibernatereload=false, and I believe that did help fix the issues where playlists were getting mixed together and where my non-playlist "play all files in this folder" were not restoring correctly. Unless you're actually doing HORM, avoid setting this!

I had no idea what HORM was until recently, and I must have switched it on during previous troubleshooting attempts - "reload my playlist after hibernate" sounded like just the thing to fix my problem! :/
Tried 2.95 and 5.13 versions of winamp, and the both showed the same behavior. The main problem still persists, in that when I reload, the playlist just skips from song to song without actually playing anything.
I've narrowed the core cause of this a bit. I use relative paths in my playlists, as I port them around between devices. In order to find a file with a relative path, you need a root path. When I first select a playlist through RR, it correctly supplies Winamp with the "Playlist Folder" as my root path (or, perhaps it just passes along the actual location of the playlist in my filesystem, those are equivalent for me). However, after a restart, Winamp is using "C:\Program Files\Winamp" as the root path (you can see this by minimizing RR, going to the Winamp playlist window, right-clicking on a file and selecting File Properties), and attempting to find song files using paths relative to that location. Of course, that path is garbage and points to a file that doesn't exist, so it skips around trying to load every file on the playlist unsuccessfully.
I also observed that if I shutdown RR, launch winamp directly, start playing a playlist, exit winamp, and relaunch winamp, I get the playlist reloaded correctly and the correct song selected, however the song is not playing at all. I can press 'play' and it'll start playing from the beginning. This indicates that the root folder is getting restored correctly, but the song-in-progress is not actually restarted (and of course the position within the song is also not restored).
That's as far as I've gotten so far...Off to do more research to see if I can resolve the two remaining issues:
A) Get RoadRunner to supply Winamp with the correct Root Path after a restore (or stop interfering with Winamp's ability to remember this itself)
B) Get Winamp to remember the location of the song-in-progress after a successful restart so the song automatically continues playing, rather than having to manually restart from the beginning.
Again, any tips or expert wisdom is very much appreciated!
-D
UPDATE: correction, latest experiments indicate the "Conflating playlists" issue DOES appear to be caused by non-magic versions of Winamp, rather than the HORM switch.
UPDATE2: Winamp 2.95 doesn't have WMA support. I have some WMAs scattered among my collection, surely causing an increase in apparently random failures during my testing...