INTERACT FORUM
More => Old Versions => JRiver Media Center 20 for Windows => Topic started by: daek on April 18, 2015, 03:01:30 pm
-
Hi folks!
Is it possible to configure the skip count behavior?
I'm wondering why the skip count is incremented when I stop playback.
-
The logic is relatively simple:
[Skip Count] is incremented if: (a) playback of a particular file is started, and (b) playback is stopped (via any means, next track, stop, playing something else, etc) before the playhead reaches 50% in the file (for Audio files) and ~90% (for video files). The video file "trigger percentage" was recently changed (it used to match the Audio file one) and I don't remember exactly where it landed, but it is somewhere around 90%.
[Number Plays] is incremented if: (a) playback of a particular file is started, and (b) playback continues through the point where it will no longer count as a skip.
You don't have to actually play through the files to see this effect. You can manually drag the playhead and watch the [Number Plays] and [Skip Count] fields change, depending on where the playhead is when you stop playback.
You can't directly alter it, but you can make up your own rules if you want:
http://yabb.jriver.com/interact/index.php?topic=85974.0
-
glynor, I'm not stalking you this weekend... honestly....!
I don't use skip counts for anything. Since the day they were introduced I absolutely failed to get my head around how me deciding to stop playback by pressing Stop could ever be considered a skip?
The skip count figure, to me, is completely meaningless.
They've been around for a long time now, so, out of curiosity, is anyone using this data for anything useful?
-
They've been around for a long time now, so, out of curiosity, is anyone using this data for anything useful?
I would use skip counts if there was a way to configure the behavior :)
I would prefer for the skip count to stay the same if:
- playback was stopped
- MC was closed
- playback was switched to the previous track
- I clicked currently playing track 2 times to start from the beginning
The way it is right now skip counts are not useful for me since they increase like mad
-
Thank you glynor.
I am using play/skip count for audio in smartlists. Songs that have (play - skip < 0) are excluded for 2 months since last skipped date.
I think there are some usefull scenarios...
But I don't like that stop is equal to skip.
-
I made a page here to document this better:
http://wiki.jriver.com/index.php/Play_Statistics
-
Since the day they were introduced I absolutely failed to get my head around how me deciding to stop playback by pressing Stop could ever be considered a skip?
The skip count figure, to me, is completely meaningless.
They've been around for a long time now, so, out of curiosity, is anyone using this data for anything useful?
I actually use it a bit, and I understand why Stop is considered a skip (both technically and philosophically). MC doesn't distinguish between a stopping and advancing to the next track (or skipping to the previous track). When you advance to the next track, as far as the engine is concerned, this is three "discreet" actions: stop playback of the current file, advance Playing Now to the next file, and start playback of this file.
But, I also get why it might consider a stop a skip. Because, in my house, quite often if "something bad" comes on, my reaction is to hit Stop (especially if one or two other tracks in the list have been bad as well). Sometimes, yes, I'll hit next to skip past it, but I'd say almost as often, I hit stop, especially if the file is particularly bad. I do this (as a gut instinct sometimes, but) so I can browse around and pick something new, without the obnoxious sound of the "terrible" in my ears (and sometimes so I can delete the offender).
So, if you tracked only those times when I skip to the next track soon after starting playback of a file, you'd miss a bunch of these instances.
It does not care how playback was stopped. It only cares where the "bookmark" in the file was (where the Playhead was) when you stopped it. This is, for me, a fairly accurate representation of how likely I am to dislike a file. There are exceptions, and you can't just blindly trust the statistic, but I think this would apply to other methods of collecting the statistic equally. You'll get "extras" no matter what. This way, at least you don't ever miss anything.
I have a Smartlist that filter out songs I've rated highly, those with a similarly high [Number Plays] count (if it is played a lot, it'll also be skipped a lot, so I do some math trickery comparing the two stats), and a few other things I use to "protect" files, and then it sort files with large [Skip Count] to the top.
Every so often, I go though this list, and I usually find I want to delete or Rate 1 Star a bunch of the files in the first 20-40 entries in the list.