Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Recent Posts
71
MP3 - General / Re: Resurrecting/Preserving the Helix MP3 encoder
Last post by maikmerten -
Thanks, yeah, seems you're correct. It appears most sources consider the correct bias to be 128. Fixed!

edit: Actually, thinking about this, I may have managed to confuse myself. If 0x80 (128) is the midpoint ("0.0"), and 0xFF (255) is "1.0", then shouldn't have -127 be correct in the first place? Either way, given that the range isn't perfectly symmetric, shouldn't there be a DC bias either way? So isn't "-127.5" technically even more correct?

Also, perhaps I should multiply with 256 instead of 255 to scale it properly?
73
3rd Party Plugins - (fb2k) / Re: JScript Panel
Last post by marc2k3 -
By default, the first value of multi-values tags is already used...

Code: [Select]
$meta(artist,0)

You can override this by adding this inside the panel as the first line of code.

Code: [Select]
DEFAULT_ARTIST = '$if($stricmp(%album artist%,Various Artists),$meta(artist,0),%album artist%)'
75
Support - (fb2k) / Re: Unexpected Behaviour with Multi-Value Performer Tag in M4A Files
Last post by marc2k3 -
According to this...

https://picard-docs.musicbrainz.org/en/appendices/tag_mapping.html#performer

Picard doesn't even write performer to m4a files. I've tested this in the past on windows but not checked recently. I guess the page could be outdated???

I know for sure foobar2000 can't read the TCML/IPLS id3 frames Picard writes to mp3 files. If you ask fb2k to write performer, it uses TXXX frames.

edit: I tested the windows version of picard and it seems to match the behaviour I linked to above - they're not written. Testing an mp3 copy of the same album does get written so the performer entries are definitely there for the release I'm testing.
76
3rd Party Plugins - (fb2k) / Re: JScript Panel
Last post by KthugS -
Hello everyone!
Among the samples, “Last.fm Bio + Images” was used.
There are a lot of artist folders in the "\js_data\artists" path.
Due to the musicbrainz artist field format, folders are created for each participating artist, such as "feat.", "with", "&", etc., and images are downloaded.
Most of the downloaded images are album artist images, ignoring the participating artists, and most of them are the same images only with different folders.
Over the course of a few days, the amount of duplicate images alone approaches 1GB.

Is it possible to make this create folders based on album artist?
1. Use the track artist field only in the case of Various Artists.
2. Only use the first artist in the “%ARTISTS%” field
3. Filter only the first artist in the “%ARTIST%” field?

Anyone experiencing a similar problem?

Except for this situation, I really like musicbrainz's method.  :D
77
3rd Party Plugins - (fb2k) / Re: Georgia-ReBORN - A Clean foobar2000 Theme
Last post by regor -
I am not sure what are you trying to test with that, it's not doing what you think it does. New requests should also be directly skipped on shutdown.

Anyway here is a test file which crashes if you delete "window.IsUnload = true;" line; but works fine with it. Note the script will try to access a list of urls and close foobar immediately. So upon foobar restart, it will do the same and effectively make it unusable unless the last line of code closing foobar is removed.

Also note the interval ms affects how it's handled. 30 ms seems to work fine, but higher settings may crash too.  For sure 3000 ms will crash, there is a small window of time between the shutdown request and the actual shutdown and you have to use that.

That can be avoided with a direct abort routine at the callback and a stack of request similar to your approach, but the stack should be managed at the callback (not saying all this is needed, just that it would be the best approach, but the other one may be good enough).

Alternatively, wget or similar may be used to retrieve data from web without all these problems... it may be wise to apply this fix to xmlhttp requests too (and beware of similar behaviors on any activeX object).
79
MP3 / Re: Low bitrate MP3 (+ unsupported bitrates)
Last post by itisljar -
Oh sweet Je... that's a blast into the past.
I've tried 1.02 version now, and I didn't manage to encode a song without white noise at the end.
Be aware of bugs in that piece of software.
80
MP3 - General / Re: Resurrecting/Preserving the Helix MP3 encoder
Last post by maikmerten -
Okay, I now threw out the 8-bit resampling code - now all sample formats are handled by the same 32-bit float resampling/conversion steps. https://github.com/maikmerten/hmp3/commit/a07d35c3dbe5acb22e1301c27df2dfe92dccb908

I also adjusted the Makefile and VisualStudio project files (by hand). Seems the automated builds on AppVeyor still work ( https://ci.appveyor.com/project/maikmerten/hmp3/builds/49846828/artifacts ), so I assume the VisualStudio project file surgery went okay.