#238102 - 18/10/2004 18:18
End Of Playlist
|
journeyman
Registered: 24/04/2002
Posts: 78
|
Every time i repower the empeg with "all" as the playlist (ie: 3 down button presses), i get "End of playlist". How can i fix this problem?
|
|
Top
|
|
|
|
|
#238103 - 18/10/2004 18:28
Re: End Of Playlist
[Re: woops]
|
carpal tunnel
Registered: 20/12/1999
Posts: 31597
Loc: Seattle, WA
|
Do you have a lot of tracks? By a lot, I mean more than a few thousand. Or, do your track tags contain unusually large amounts of information that might make the database very big? For instance, is there a lot of information in the Comment tag on all your tracks? What I'm getting at here is that it might be the database size limitation problem.
|
|
Top
|
|
|
|
|
#238104 - 18/10/2004 18:31
Re: End Of Playlist
[Re: tfabris]
|
journeyman
Registered: 24/04/2002
Posts: 78
|
Database size limitation. I've deleted all comments. Well, its always something!
Thanks Tony
|
|
Top
|
|
|
|
|
#238105 - 18/10/2004 19:09
Re: End Of Playlist
[Re: woops]
|
carpal tunnel
Registered: 20/12/1999
Posts: 31597
Loc: Seattle, WA
|
Not clear on what your answer means... Do you mean that deleting all comment tags solved the problem?
|
|
Top
|
|
|
|
|
#238106 - 18/10/2004 19:11
Re: End Of Playlist
[Re: tfabris]
|
carpal tunnel
Registered: 20/12/1999
Posts: 31597
Loc: Seattle, WA
|
Another thing that might cause it... Tagging all your top-level playlists "Ignore as child". You wouldn't happen to have done that would you?
|
|
Top
|
|
|
|
|
#238107 - 18/10/2004 22:37
Re: End Of Playlist
[Re: tfabris]
|
journeyman
Registered: 24/04/2002
Posts: 78
|
Tony,
I meant to say that i am having this problem only because of database size limitation. I don't have any comments and didn't do the ignore as child.
Thanks.
|
|
Top
|
|
|
|
|
#238108 - 18/10/2004 23:44
Re: End Of Playlist
[Re: woops]
|
carpal tunnel
Registered: 05/01/2001
Posts: 4903
Loc: Detroit, MI USA
|
I feel your pain.
_________________________
Brad B.
|
|
Top
|
|
|
|
|
#238109 - 19/10/2004 08:33
Re: End Of Playlist
[Re: SE_Sport_Driver]
|
carpal tunnel
Registered: 15/08/2000
Posts: 4859
Loc: New Jersey, USA
|
As do I. But I do not do the <down> <down> <down> playlist very often... Current FID count with my latest upload = 0x84DC... I bet a few folks never dreamed in their wildest nightmares that they would need to account for 100GB drives or databases this huge. Or memory upgrades. Or illuminated buttons. ... Oh, strike that - they did think of and provide soldering pads for illuminated buttons...  This is why the empeg still beats anything on the market today!!! Years ahead of its time...
_________________________
Paul Grzelak
200GB with 48MB RAM, Illuminated Buttons and Digital Outputs
|
|
Top
|
|
|
|
|
#238110 - 19/10/2004 09:54
Why the Empeg player still beats anything on the market today
[Re: pgrzelak]
|
enthusiast
Registered: 06/03/2003
Posts: 269
Loc: Wellingborough, UK
|
Quote:
This is why the empeg still beats anything on the market today!!!
<monty_python>
...And because the team that gave us the Empeg actually cared about the product.
And because that team relished that they were getting paid to pursue their hobby or pet project.
But apart from some future proofing, being years ahead of its time, the team caring about the product and being paid to work on a pet project, there is no rationale for why the Empeg still beats anything on the market today!
</monty_python>
|
|
Top
|
|
|
|
|
#238111 - 19/10/2004 10:00
Re: End Of Playlist
[Re: pgrzelak]
|
carpal tunnel
Registered: 05/01/2001
Posts: 4903
Loc: Detroit, MI USA
|
I'd use "down, down, down" more often, but having to restart it is always a bummer. It's too bad we can't make a work around.
_________________________
Brad B.
|
|
Top
|
|
|
|
|
#238112 - 19/10/2004 10:26
Re: End Of Playlist
[Re: SE_Sport_Driver]
|
enthusiast
Registered: 06/03/2003
Posts: 269
Loc: Wellingborough, UK
|
Quote:
I'd use "down, down, down" more often, but having to restart it is always a bummer. It's too bad we can't make a work around.
I havn't been following this issue - what happens if some of the playlists are tagged "ignore as child"? - If you are correct in your assertion that the database limit is the cause of your problem, then if down-down-down creates a playlist with less than <limit> items, does the player remember where it got to?
I guess that it might be possible to put some hooks in Hijack to squirrel away the current position on power loss and restore on power resumption - but it seems like a lot of work for a problem that the Empeg guys are working hard to fix properly.
[edit: improve clarity]
Edited by mdavey (19/10/2004 10:32)
|
|
Top
|
|
|
|
|
#238113 - 19/10/2004 10:44
Re: End Of Playlist
[Re: mdavey]
|
carpal tunnel
Registered: 13/07/2000
Posts: 4180
Loc: Cambridge, England
|
Quote:
If you are correct in your assertion that the database limit is the cause of your problem, then if down-down-down creates a playlist with less than <limit> items, does the player remember where it got to?
Yes. It's only the size of the current running order (the 5000 in the "[36/5000]" display) that's the problem. The actual database size isn't important: if you've only got one track on the player and you "append" it thousands of times, you'll hit this problem; if you've got 40,000 songs on your player but only play them an album at a time, you won't hit this problem.
Peter
|
|
Top
|
|
|
|
|
#238114 - 19/10/2004 13:01
Re: End Of Playlist
[Re: peter]
|
carpal tunnel
Registered: 05/01/2001
Posts: 4903
Loc: Detroit, MI USA
|
This is due to the scratch partition size correct?
_________________________
Brad B.
|
|
Top
|
|
|
|
|
#238115 - 19/10/2004 13:21
Re: End Of Playlist
[Re: SE_Sport_Driver]
|
carpal tunnel
Registered: 13/07/2000
Posts: 4180
Loc: Cambridge, England
|
Quote:
This is due to the scratch partition size correct?
Yup. Same reason you can't bookmark really long running-orders.
Peter
|
|
Top
|
|
|
|
|
#238116 - 19/10/2004 13:25
Re: End Of Playlist
[Re: peter]
|
carpal tunnel
Registered: 15/08/2000
Posts: 4859
Loc: New Jersey, USA
|
The weird thing is that metadata for the individual FIDs still seems to work. I remember this was based on specific sectors (or was it blocks...) based on the FID number. So far, I have not blown anything up yet. And that surprises me slightly.
I was going to experiment when building the new drives to increase the size of the scratch partition, but after reading on the board that the upgrader depends on specific size partitions that it might not be a good idea after all... Even then, there is no certainty that the extra space would be used / recognized...
_________________________
Paul Grzelak
200GB with 48MB RAM, Illuminated Buttons and Digital Outputs
|
|
Top
|
|
|
|
|
#238117 - 19/10/2004 13:32
Re: End Of Playlist
[Re: pgrzelak]
|
carpal tunnel
Registered: 13/07/2000
Posts: 4180
Loc: Cambridge, England
|
Quote:
The weird thing is that metadata for the individual FIDs still seems to work. I remember this was based on specific sectors (or was it blocks...) based on the FID number.
That's only true for so-called "dynamic" metadata (play count, time last played, track profile). If your dynamic data is working for FIDs above 0x7000 I'll be impressed, not to say worried. Fortunately dynamic data failures are benign, unless you're a big user of shuffle modes that depend on play-count or time-last-played.
Peter
|
|
Top
|
|
|
|
|
#238118 - 19/10/2004 13:38
Re: End Of Playlist
[Re: peter]
|
carpal tunnel
Registered: 15/08/2000
Posts: 4859
Loc: New Jersey, USA
|
In which case it may be failing, but I haven't noticed. I do not rely on any of those values. A good thing.
_________________________
Paul Grzelak
200GB with 48MB RAM, Illuminated Buttons and Digital Outputs
|
|
Top
|
|
|
|
|
#238119 - 19/10/2004 13:43
Re: End Of Playlist
[Re: peter]
|
carpal tunnel
Registered: 27/06/1999
Posts: 7058
Loc: Pittsburgh, PA
|
Quote:
unless you're a big user of shuffle modes that depend on play-count or time-last-played.
Or silly 3rd party applications that write song ratings to the skips count field... Hmm. I wonder what would happen if someone has a lot of FIDs and tries to use emphatic to write to a non-existent sector of hda3. What's after per-FID dynamic data? Running orders? Is 0x7000 the exact number that I should use to avoid writing past the end of the per-FID data?
|
|
Top
|
|
|
|
|
#238120 - 19/10/2004 13:49
Re: End Of Playlist
[Re: tonyc]
|
carpal tunnel
Registered: 13/07/2000
Posts: 4180
Loc: Cambridge, England
|
Quote:
Or silly 3rd party applications that write song ratings to the skips count field... Hmm. I wonder what would happen if someone has a lot of FIDs and tries to use emphatic to write to a non-existent sector of hda3. What's after per-FID dynamic data? Running orders? Is 0x7000 the exact number that I should use to avoid writing past the end of the per-FID data?
The final 14Mb of the 16Mb scratch partition is allocated to per-FID dynamic data, one FID per 512-byte sector. So the 0x7000th FID (i.e. FID 0x70000) is the first one which can't be written. What happens when you try and write a sector beyond there depends on how you're doing it: if you've opened /dev/hda3 and are writing to it, then you'll just get end-of-file errors. If you've opened /dev/hda, or if you're using raw sector writes, you will corrupt the initial sectors of the next partition -- namely, the superblock of the music partition. That would be bad.
Peter
|
|
Top
|
|
|
|
|
#238121 - 19/10/2004 13:53
Re: End Of Playlist
[Re: peter]
|
carpal tunnel
Registered: 27/06/1999
Posts: 7058
Loc: Pittsburgh, PA
|
Quote:
That would be bad.
Yeah, that's what I was thinking. But fortunately I am working with hda3 and not hda. Still, to avoid the corrolaries of Murphy's Law, I'm going to put in a check for 0x7000 in the next version.
|
|
Top
|
|
|
|
|
#238122 - 19/10/2004 14:04
Re: End Of Playlist
[Re: pgrzelak]
|
journeyman
Registered: 24/04/2002
Posts: 78
|
Paul,
0x84DC?! That's obscene! (he jealously screamed....)
|
|
Top
|
|
|
|
|
#238123 - 19/10/2004 14:06
Re: End Of Playlist
[Re: woops]
|
carpal tunnel
Registered: 15/08/2000
Posts: 4859
Loc: New Jersey, USA
|
Actually, it is. The last thing I added was the (relatively new) Lenny Bruce boxed set...
Edit: added link...
_________________________
Paul Grzelak
200GB with 48MB RAM, Illuminated Buttons and Digital Outputs
|
|
Top
|
|
|
|
|
#238124 - 19/10/2004 14:29
Re: End Of Playlist
[Re: tonyc]
|
carpal tunnel
Registered: 29/08/2000
Posts: 14493
Loc: Canada
|
So long as you're opening /dev/hda3, my existing IDE safeguards will catch and block out-of-bounds accesses. No need for an extra check, and it could even get in the way if someday we get access to more FID space..
Cheers
|
|
Top
|
|
|
|
|
#238125 - 19/10/2004 14:31
Re: End Of Playlist
[Re: mlord]
|
carpal tunnel
Registered: 27/06/1999
Posts: 7058
Loc: Pittsburgh, PA
|
Good point. Belt it is then, no suspenders necessary.
|
|
Top
|
|
|
|
|
#238126 - 19/10/2004 14:35
Re: End Of Playlist
[Re: peter]
|
carpal tunnel
Registered: 18/01/2000
Posts: 5683
Loc: London, UK
|
Quote:
The final 14Mb of the 16Mb scratch partition is allocated to per-FID dynamic data, one FID per 512-byte sector. So the 0x7000th FID (i.e. FID 0x70000) is the first one which can't be written.
Assuming that your dynamic data partition is exactly 16Mb in size. That's only the minimum size. It could have been rounded up to a higher cylinder number.
Obviously, it might not be a good idea to rely on this...?
_________________________
-- roger
|
|
Top
|
|
|
|
|
#238127 - 19/10/2004 14:38
Re: End Of Playlist
[Re: Roger]
|
carpal tunnel
Registered: 25/12/2000
Posts: 16706
Loc: Raleigh, NC US
|
Does that mean we could construct a larger scratch partition by hand and expect the player to utilize it?
_________________________
Bitt Faulk
|
|
Top
|
|
|
|
|
#238128 - 19/10/2004 14:44
Re: End Of Playlist
[Re: Roger]
|
carpal tunnel
Registered: 13/07/2000
Posts: 4180
Loc: Cambridge, England
|
Quote:
Quote:
the 0x7000th FID (i.e. FID 0x70000) is the first one which can't be written.
Assuming that your dynamic data partition is exactly 16Mb in size. That's only the minimum size. It could have been rounded up to a higher cylinder number.
Well, yes, it could, but the player code assumes exactly 16Mb and will not read or write dynamic data in sectors beyond there.
Peter
|
|
Top
|
|
|
|
|
#238129 - 04/03/2005 20:23
Re: End Of Playlist
[Re: tonyc]
|
carpal tunnel
Registered: 29/08/2000
Posts: 14493
Loc: Canada
|
Quote:
Quote:
That would be bad.
Yeah, that's what I was thinking. But fortunately I am working with hda3 and not hda. Still, to avoid the corrolaries of Murphy's Law, I'm going to put in a check for 0x7000 in the next version.
Time to remove that check. The kernel already protects against overrun here anyway, and we now have players that bypass the old 0x7000 limit.
|
|
Top
|
|
|
|
|
|
Previous Topic
Index