Jump to content

where change transcode buffer setting


riku

Recommended Posts

Heya,

Want to thanks this nice project. i love it.

 

However i need to change ffmpeg settings on linux, where i can find those? i look /usr/lib/emby-server and there is binary files only, is there a way without recompile server?

Problem is the ffmpeg is not buffering enough. i have a lagging stream it stops every time for couple second and then it start transcode bit by bit. it is issue not enough of cache. I dont know if this a bug or what, but everything works fine, when it reach end of buffer and it need transcode the next .ts file then it take few seconds. i want server to keep doing transcode few minutes ahead, not on demand. so i want change some settings around.

 

im streaming from cloud so this brings couple more second need to buffer. this is the reason why it is not working how it suppose.

 

streaming directly from cloud have no problem, if i use vlc and put buffer big enough and stream direcly the file it plays without problems. so there is no speed issues. there is no internet issue, no cpu issue. ffmpeg only takes 1-3sec to transcode each .ts.

 

 

Link to comment
Share on other sites

Hi, we don't currently have a setting for this, but it's a possibility for the future. Thanks !

Link to comment
Share on other sites

Animosity022

What are you using for your cloud mount? I've played big movies and tv shows from the cloud without much a problem.

Link to comment
Share on other sites

Nice to have someone understand what i meant.

im using atm combination of amazon cloud, rclone and s3ql. as s3ql do the drive chunks there cannot be use max read ahead as mount option. only way to make this work is have the ffmpeg read more the file than is needed because on extra delay.

 

 

i wait abit and try research more info, and have hopes of devs emby might come a solution for this, but if is this not possible, i might try do a rclone direct fs with encryption and try via that, but that means few Terabytes need to copy new drive lol.

Edited by riku
Link to comment
Share on other sites

Animosity022

I have roughly 20TB using ACD on a rclone encypted mount point. It works pretty well and have no issues with buffering currently.

 

My mount looks like:

rclone mount \
--allow-other \
--default-permissions \
--uid 1000 \
--gid 1000 \
--umask 002 \
--acd-templink-threshold 0 \
--buffer-size 1G \
--max-read-ahead 200M \
--syslog \
--stats 1m \
-v \
media: /media &

I keep everything to syslog and the only times I see problems are when ACD seems to have an issue.

 

You'll see some low level retries which sometimes the buffer handles it and sometimes not.

 

 ReadFileHandle.Read error: low level retry 1/10: unexpected EOF

Link to comment
Share on other sites

thanks for info, i just might try it some point. even its a pain to move alot files around.

im still reading https://forum.rclone.org/t/best-mount-settings-for-streaming-plex/ its very good one.

 

after doing some tests i see logs some of chunks come 60Mbit/s and some chunks come just barely 300kb/s... it seems the servers are slow, i was also seeing with netstat it changes ip all the time of connecting endpoint. bottleneck is this atm + s3ql performance.

now im trying find out what account i have, i guess its a US one, and my server is located on EU.. i might be need to switch the amazon location if that will not work. and that is even bigger work

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...