CatBus;367898 Wrote: > I have to say I agree with Phil on most but not all points. > > Basically what we're dealing with is (in almost all cases) either > inadequately or improperly tagged music. As such, the best overall > solution is to fix the tags. Once fixed, everything will work > perfectly.
that really isn't true. if you want things to properly sort in say both SC and winamp, or WMP, or whatever, you have to have TPE2 filled out and treated as album artist for comps. this means SC sees nothing as a comp. this has to be done to defeat the VA logic for things that are misidentified, but also for the other apps too. CatBus;367898 Wrote: > Where I disagree with Phil is about "guessing". I believe what > SqueezeCenter is doing is guessing. If there's no COMPILATION tag set, > then there's simply no way to be 100% certain whether an album is a > compilation or not. There simply isn't enough data to draw a > conclusion with absolute certainty. So it falls back to "mismatched > artists and album artists", which is a very good way to guess with (I'm > guessing) well over 99% chance of being correct, but it's still > guessing. > > I don't personally see the benefit in skipping this "guessing" phase, > esp. since this guessing phase is right so much more frequently than > it's wrong, and is only triggered by badly tagged music. For the odd > few where the guessing logic doesn't work out, fixing the tags seems > the obvious solution. i can tell you that in my case, half of what SC called a comp wasn't a comp. sure, i could just set album artist on those that were misidentified, as SC intends me to do, but both gracenote and winamp (and other apps) set album artist (via TPE2) on everything including comps to get comps to behave well in those apps. its a catch 22. as to guessing, clearly thats what SC is doing. the idea that it isn't makes no sense. the VA function may itself follow static rules of conduct, but thats not the point, no one disputes that, one artist mismatch results in SC seeing a comp, that isn't the criticism i have. the criticism i have is that that assumption is not very logical as a method to employ for comp detection. its simply guessing that the rule it follows will be right often enough. its nothing more than a guess, and while it is right at times, and maybe right for a lot of people a lot of the time, its not fullproof or anything close to it. it should be optional. -- MrSinatra www.LION-Radio.org Using: Squeezebox2 (primary) / SBR (secondary) / SBC - w/SC 7.3b - Win XP Pro SP3 - 3.2ghz / 2gig ram - D-Link DIR-655 ------------------------------------------------------------------------ MrSinatra's Profile: http://forums.slimdevices.com/member.php?userid=2336 View this thread: http://forums.slimdevices.com/showthread.php?t=56078 _______________________________________________ ripping mailing list [email protected] http://lists.slimdevices.com/lists/listinfo/ripping
