[EricZX505]

I have a collection of about 20,000-30,000 opus files and this takes the Tangara 4-5 hours to attempt to put them in a database and often leads to boot looping in which I need to keep reupdating the database when it happens. If there is a way to compile the database on my computer it would make using the tangara much easier and convenient. I have already stripped the opus files of their image tag data.

[ChriChri]

If there is a way to compile the database on my computer it would make using the tangara much easier and convenient. I have already stripped the opus files of their image tag data.

This would be a nice feature request for tangara-companion ...

[redshift]

In fact, there is already a request for it: https://github.com/haileys/tangara-companion/issues/19

[tbcollins1]

I have a collection of about 20,000-30,000 opus files and this takes the Tangara 4-5 hours to attempt to put them in a database and often leads to boot looping in which I need to keep reupdating the database when it happens. If there is a way to compile the database on my computer it would make using the tangara much easier and convenient. I have already stripped the opus files of their image tag data.

You might also want to follow this thread over on the code repo re: big libraries -
https://codeberg.org/cool-tech-zone/tang...issues/413

In general, the firmware in its current state doesn't really handle large libraries well (or at all really) and I don't think I've seen one person yet report a successful scan of a big library without either bootloops and/or the scan ending prematurely. 

I have an iPod classic 5th gen with Rockbox installed that is able to initialize my database of approximately 32k files (mostly MP3, with some FLACs and M4As thrown in) in about 15-20 minutes. I'm hoping either through firmware updates or through a desktop app the Tangara can get close to this someday.

[EricZX505]

This may be a weird coincidence but having my SD card formatted to fat32 and running fatsort on the card before putting it in my Tangara has given me much more successful and consistent results. It still takes about 2.5 to 3 hours but it gets a lot more of my collection than it otherwise would. It’s still not perfect and doesn’t grab everything but it’s much better.

[ailurux]

This may be a weird coincidence but having my SD card formatted to fat32 and running fatsort on the card before putting it in my Tangara has given me much more successful and consistent results. It still takes about 2.5 to 3 hours but it gets a lot more of my collection than it otherwise would. It’s still not perfect and doesn’t grab everything but it’s much better.

That would potentially make sense with the way LevelDB stores keys as the difference between the key and the previous one. I've been doing some debugging to try figure out what's been causing a lot of issues people have been having with the database, and at least one crash I've seen happened during the step in which LevelDB compacts the database. 

One of the things I've wanted to try to help figure out these issues is breakout the database indexing into it's own application to run on my computer, as that would also help a lot with being able to create databases for testing. If I find the time to do that and it works well it would also give people a faster way to index their libraries. Especially those with large music collections.