Shuffle mode needs to be refined

Bug #20032 reported by Trouilliez vincent
66
This bug affects 8 people
Affects Status Importance Assigned to Milestone
Rhythmbox
Fix Released
Wishlist
rhythmbox (Ubuntu)
Fix Released
Wishlist
Unassigned

Bug Description

I use RB to play an eclectic collection of 184 MPS's, with a total duration of
over 13 hours.
I use the 'shuffle' mode to play these. The problem is that it regularly play
the same song twice in a row, or 3 times in just 20 minutes.

I used XMMS to play this songs in the past (before using RB ie), and IIRC, XMMS,
in shuffle mode, will wiat for the entire playlist to have been played, before
allowing a song to be played again. So it's better, but not perfect, because if
the playlist is very long, it could take days before a song plays twice. Also,
IIRC, once it has played all the songs, it will not 'shuffle' them anymore : it
will just play them in the same order that they got played the first time :o(

I think a better solution would be for example to wait some time before playing
a given song again. Say, wait at least 30 or 45 minutes before playing twice the
same song, or make it configurable with some Gconf key or something. This way
you don't get the same songs twice in a row, nor do you wait 3 days to hear a
song twice. Best of both worlds... :o)

Revision history for this message
Sebastien Bacher (seb128) wrote :

There is some bugs upstream about that,
http://bugzilla.gnome.org/show_bug.cgi?id=165863 by example.

Revision history for this message
Sebastien Bacher (seb128) wrote :

*** Bug 23047 has been marked as a duplicate of this bug. ***

Changed in rhythmbox:
status: Unconfirmed → Confirmed
Revision history for this message
Charlie Kravetz (cjkgeek) wrote :

***Bug 16862 was rejected on 2005-05-15.

Re: inotify oops Posted by ChuckShort at 2005-05-15 15:27:42 MDT

2.6.11 is in universe and not supported***

Can this be confirmed as still valid?

Revision history for this message
Charlie Kravetz (cjkgeek) wrote :

Need to insure this is still a valid bug, Bug #16862 which was referenced has been rejected as being in universe package

Changed in rhythmbox:
status: Unconfirmed → Needs Info
Revision history for this message
Trouilliez vincent (vincent-trouilliez-modulonet) wrote :

> Can this be confirmed as still valid?

Yes, confirmed.

Revision history for this message
Sebastien Bacher (seb128) wrote :

do you still have that problem?

Revision history for this message
Trouilliez vincent (vincent-trouilliez-modulonet) wrote :

Yes.

Revision history for this message
Sebastien Bacher (seb128) wrote :

marking confirmed and opening an upstream task so it doesn't stay on the bug triager list

Changed in rhythmbox:
assignee: nobody → desktop-bugs
status: Needs Info → Confirmed
Changed in rhythmbox:
status: Unknown → Unconfirmed
Revision history for this message
Charlie Kravetz (cjkgeek) wrote : Re: [Bug 20032] Re: Shuffle mode needs to be refined

On Wed, 2007-01-31 at 15:11 +0000, Sebastien Bacher wrote:
> do you still have that problem?
>

Sorry, I didn't have that problem. All I did was what you are doing and
have others confirm it.

--
Charlie Kravetz
<email address hidden>
Linux User #425914
ref link http://counter.li.org/
Helping spread awareness and education about Multiple Sclerosis

Revision history for this message
StuartYeates (stuart-yeates) wrote : Workaround

There is a workaround for this problem described at:

http://www.last.fm/group/Rhythmbox/forum/8096/_/288791

cheers
stuart

Revision history for this message
Endolith (endolith) wrote :

Seems like "random by age" might be a better choice for Ubuntu's default setting?

Revision history for this message
Charlie Kravetz (cjkgeek) wrote : Re: [Bug 20032] Re: Shuffle mode needs to be refined

On Wed, 2007-10-24 at 23:38 +0000, Endolith wrote:
> Seems like "random by age" might be a better choice for Ubuntu's default
> setting?
>
I couldn´t tell you. I just verified that the bug report was still
valid. Perhaps Vincent-trouilliez, who submitted the original report and
has verified it each time will be able to answer this also.

--
Charlie Kravetz
Linux Registered User Number 425914
Never let anyone steal your DREAM.
Plumber, bicyclist, retired with too much to do...

Revision history for this message
Trouilliez vincent (vincent-trouilliez-modulonet) wrote :

Here I am ;-)

Well I am not dreaming of anything fancy really. Basically to make me happy, all I would want is:

1) keep the random generator as it is, that is, "truly" random (as far as random makes sense in a computer ie ;-)
2) THEN, check the song number, and if that song has already been played, say, in the last 10 tracks, well then just generate another random number, and that's is ! :o)

Revision history for this message
Endolith (endolith) wrote :

It shouldn't be the last 10 tracks, though. It should be more like "never play the same song twice during one session, unless every song has already been played". It's how you define "session" that's the problem.

Revision history for this message
Trouilliez vincent (vincent-trouilliez-modulonet) wrote :

> It should be more like "never play the same song twice during one session, unless every song has already been played".

Wow, please don't do that ! That would be horrible, at least for me ! :-(
I mean, I don't like all my songs equally... so there are songs that I don't even want to hear, and others that I like a lot, which I really don't mind if I hear them several times per evening. I just don't want to hear the same song litterally twice in a row, or 3 times in 10 songs, because that really is "too much". But I have nearly 600 songs in my library, and if I have to go through 600 songs before I have a chance to hear again a song I like, I will suicide ! ;-/

That's why I proposed to wait only a few songs, be it 10 or 20 or whatever, this way it garantees that I won't hear the same song twice in a row, yet I won't have to wait 3 months before I can enjoy it again. There is a balance to be found somewhere between 0 and 600 ;-)
Maybe RB could be adaptative, and determine this threshold automatically, to find an appropriate value based on the number of songs.

Changed in rhythmbox:
status: New → Fix Released
Revision history for this message
Pedro Villavicencio (pedro) wrote :

Fixed upstream, thanks.

Changed in rhythmbox:
status: Confirmed → Fix Committed
Revision history for this message
Sebastien Bacher (seb128) wrote :

there is a new version in hardy now which fixes the issue

Changed in rhythmbox:
status: Fix Committed → Fix Released
Revision history for this message
Gary Bickford (garyb-fxt) wrote :

No, it isn't fixed. I am really, really tired of hearing the same ten songs at least once or twice every day, out of 1800.

Revision history for this message
Pedro Villavicencio (pedro) wrote :

Gary, please comment on the upstream bug located at: http://bugzilla.gnome.org/show_bug.cgi?id=165863 ; thanks you.

Revision history for this message
Vern Wall (vernwall) wrote :

I had so much trouble with playlists that I don't use them any more. I navigate to a directory and use CTRL-A to select everything, then click ADD. It seems to work well with none of the problems that playlists have.

Revision history for this message
Juan Pablo (juanps90) wrote :

Not fixed, at least not on Karmic. Haven't tested it on 10.04 though.

Changed in rhythmbox (Ubuntu):
status: Fix Released → Confirmed
Revision history for this message
Charlie Kravetz (cjkgeek) wrote :

Please comment on the upstream bug located at: http://bugzilla.gnome.org/show_bug.cgi?id=165863 ; thanks you. Since Gnome considers this fixed, if they do not hear from you, it will not change. Thanks for helping.

Changed in rhythmbox:
importance: Unknown → Wishlist
Revision history for this message
Pedro Villavicencio (pedro) wrote :

according to the upstream folks this was fixed, if you don't think its please file another bug in the usptream bts. thanks.

Changed in rhythmbox (Ubuntu):
assignee: Ubuntu Desktop Bugs (desktop-bugs) → nobody
status: Confirmed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.