View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0004952 | MMW v4 | Conversion/Leveling | public | 2008-11-20 21:51 | 2009-04-04 02:27 |
| Reporter | rusty | Assigned To | |||
| Priority | urgent | Severity | major | Reproducibility | always |
| Status | closed | Resolution | won't fix | ||
| Product Version | 3.1 | ||||
| Summary | 0004952: Album Gain: Values calculated incorrectly if only a subset of Album is analyzed | ||||
| Description | When volume analysis is conducted in the background, MM may analyze the volume of a subset of the tracks on an album (if Track/Album gain values exist for other tracks on the album). In such cases, MM generates incorrect Album Gain values, since they are calculated based on a subset of the album's tracks. What should happen if MM detects that a subset of an Album's tracks must be analysed is that the Album Gain figure for ALL of the tracks on an album should be recalculated so that they're all identical (I believe that this doesn't require re-analysis, but rather just incorporating the Track gain values from the previously analyzed tracks into the calculation of the Album Gain value for the entire album). | ||||
| Tags | No tags attached. | ||||
| Fixed in build | |||||
|
|
Is it really so? My understanding is that MM really properly re-analyzes whole album in this case. |
|
|
All album tracks are analyzed even if only one of them is unanalyzed. |
|
|
Upon further testing with Petr, it seems that the reason many Albums in my collection have unmatching Album Gain figures is that they were processed using Tools > analysis volume (i.e. Manual volume analysis). Both Petr and I have many albums in our Libraries with non-matching AG figures. From testing, we can see that MM currently assigns an Album Gain figure for tracks even after only analysing a selected subset of tracks on the Album. e.g. if the user selects 2 tracks from an album, MM will assign an AG figure for those 2 tracks even if MM 'knows' that there are other tracks on that same album. I'm guessing that the order in which Tracks are analyzed might also be messing with the accuracy of AG figures. e.g. Select 2 albums and sort by Track#, and then analyse them. Is MM smart enough to analyse the tracks an Album at a time despite the fact that they're not sorted by Album. I'm not sure. |
|
|
Clarification of manual calculation of volume leveling: 1. Order of tracks - it doesn't matter, MM internally sorts tracks according to Disc # and Track #. 2. Incomplete albums - that's right, unlike automatic (background) volume leveling, manual calculation accepts even incomplete albums. I designed it this way, because I believe that the other way would cause more bad than good. For example, if I have album 'One' in Library and then try to analyze leveling for this album on external disc that isn't in Library, Album leveling wouldn't be calculated. So, I believe that this is the best approach. |
|
|
Resolving--this isn't necessarily a bug. |
