configuration files
txt_taglist.mte
and
html_standard.mte
$filename(txt,utf-8)$loop(%_path%)%artist% - %album% - %title% - %track% - %year% - %genre% - %comment%
$loopend()
$filename(html,utf-8)<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="de">
<head>
<meta name="GENERATOR" content="Mp3tag - http://www.mp3tag.de" />
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<title>Mp3tag - File Overview</title>
<style type="text/css"><!--
h1 { font-family: Verdana, Tahoma, Arial, sans-serif; font-size: 14pt; color: #000 }
body { font-family: Verdana, Tahoma, Arial, sans-serif; font-size: 8pt; background: #fff }
table, tr, td { font-family: Verdana, Tahoma, Arial, sans-serif; font-size: 9pt; color: #000; }
a:link, a:visited, a:active { text-decoration: none; color: #03f; }
a:hover { text-decoration: none; color: #f03; }
.title { font-weight: bold; padding: 7px; margin: 0px; color: #000; }
.odd { background-color: #ccc; }
--></style>
</head>
<body>
<h1>Mp3tag - File Overview</h1>
<p>%_date%</p>
<table>
<tr>
<th class="title">Title</th>
<th class="title">Artist</th>
<th class="title">Album</th>
<th class="title">Track</th>
<th class="title">Year</th>
<th class="title">Genre</th>
<th class="title">Filename</th>
</tr>
$loop(%_filename_ext%)
<tr>
<td$if($odd(%_counter%), class="odd",)>$replace(%title%,&,&)</td>
<td$if($odd(%_counter%), class="odd",)>$replace(%artist%,&,&)</td>
<td$if($odd(%_counter%), class="odd",)>$replace(%album%,&,&)</td>
<td$if($odd(%_counter%), class="odd",)>$replace(%track%,&,&)</td>
<td$if($odd(%_counter%), class="odd",)>$replace(%year%,&,&)</td>
<td$if($odd(%_counter%), class="odd",)>$replace(%genre%,&,&)</td>
<td$if($odd(%_counter%), class="odd",)>$replace(%_filename_ext%,&,&)</td>
</tr>
$loopend()
</table>
<p>built on %_date% with %_app% - the universal Tag Editor <a href="http://www.mp3tag.de/en/" title="Mp3tag Website">www.mp3tag.de</a></p>
</body>
</html>
@Deleted_Editor_2071313 We have implemented the changes to Picard’s matching algorithm that should improve the situation with too many files getting assigned to standalone recordings.
Can you check the daily portable build from GitHub - phw/picard-daily: Inofficial daily development builds for MusicBrainz Picard and see if this gives you better results on your files?
Specifically these things have been changed:
- AcoustID recording matches without metadata are no longer highly prioritized based on the submission count
- Recordings without releases no longer get an automatically higher matching score, instead the score gets adjusted based on the “Other” preferred release types setting
- Recordings not marked as “video” no longer get a lower score automatically
- Releases without any release type no longer get an extremely low score, instead the score gets adjusted based on the “Other” preferred release types setting
All of the above are only noticeable for certain kind of edge cases (with the first two having gotten more noticeable due to the AcoustID merged recording issue). So thanks a lot for your report and detailed inquiry and testing over the last weeks 
2 Likes
@outsidecontext I have a catalog with 312 standalone recordings.
I added this to Picard.
This is the result.
Should I now clear the tags, load the catalog, click Scan?
I did not change any options in the portable version.
Yes please. The fixes where specifically made after your original report, where you noticed a regression when doing clear tags and scan over your previous runs.
The expectation is that with the changes you get far more matches with releases again.
Please note that for many of the recording-only matches the root cause is really the AcoustID merge issue. So while the changes are mitigating this it cannot fully be solved by Picard. That means you might still get more then the 11 standalone-recording matches you reported to have had before. But we can look at some of those cases and maybe come up with even further improvements.
Yes, the portable uses a local location for the config files. If you want to replicate the same setup you could use the config backup and restore in Options > Advanced > Maintenance to export the config from your regular install and import it into the portable install (you can also just copy the picard.ini file from the indicated locations). You might also need to install required plugins (this currently is not handled by the export / import).
1 Like
Much better results. 
Only 8 standalone in this katalog.
Later I will upload the entire catalog with all the MP3s along with their named album catalogs.
1 Like
1 Like
By the way.
Why didn’t Picard update the files himself?
After all, instead of clearing tags, Picard was supposed to update mp3?
When I first load this directory?
Back to the main point.
After my changes to the options, there are a lot of standalone recordings 245 again.
I don’t know which option broke it.
I won’t have time today, but tomorrow I will reset the options in Picard, enable my options one by one and see which one breaks it.
Unless you have some steps for me to take.
@outsidecontext I already found it.
This preferred option spoils the result.
Other options together and separately do not spoil.
The official version of Picard with default settings has 102 standalone recordings.
MusicBrainz-Picard-daily.exe with default settings has 8 standalone recordings.
Both cases happened on a group of 322 files.
Thanks a lot for testing, this looks very good.
It’s likely because of the setting for Compilation, which has been set to 0. Setting any preferred release type setting to zero essentially excludes releases of this type completely. If the recording is only found on compilations but there is also one of those merged recording IDs without release data it could still happen that the recording entries without release data are used.
But something changed for the worse.
This used to work. I only had 11 standalone recordings.
There will be much more here anyway.
What should I set now, because I have a lot of compilations and I want Picard to find the original, first album of the song.
Match the remaining ones manually. 8 doesn’t seem to be much.
1 Like