gewbert 2 Posted March 8, 2021 Share Posted March 8, 2021 Based on this post about using GVT-g to pass through an Intel iGPU to a VM, I started to wonder how my performance was in my set up, which was docker in a VM with GVT-g but I've since migrated to docker running directly on the host. I tested converting the same movie from 1080p to mobile/4mbps. And when I'm running Emby in Docker the performance is....awful. It takes about 40 minutes to convert the movie. I then installed Emby via the deb direct on the host and converting the same movie takes just 12 minutes. Is this normal behavior or am I missing something? I know that hardware acceleration is working because I've tested it with it off and can see the CPU utilization difference (~90% without, ~10-15% with) so I'm fairly certain it's not a docker configuration issue. I can provide logs depending on which one would be most useful and doing what actions. Link to comment Share on other sites More sharing options...
Luke 37251 Posted March 8, 2021 Share Posted March 8, 2021 Hi there, let's look at an example. Please attach the information requested in how to report a problem. Thanks ! Link to comment Share on other sites More sharing options...
gewbert 2 Posted March 9, 2021 Author Share Posted March 9, 2021 Sure, here you go. Let me know if you need anything else. embyserver.txt hardware_detection-63750824176.txt ffmpeg-transcode-a7a05733-f0f8-4fb6-8171-8fbc1af30b66_1.txt Link to comment Share on other sites More sharing options...
Luke 37251 Posted March 9, 2021 Share Posted March 9, 2021 Can you provide an equivalent set of logs from the .deb for comparison purposes? Thanks ! Link to comment Share on other sites More sharing options...
gewbert 2 Posted March 9, 2021 Author Share Posted March 9, 2021 Sure thing. embyserver.txt hardware_detection-63750836658.txt ffmpeg-transcode-69c18c56-1389-4e73-9c8e-28efd27ebc2e_1.txt Link to comment Share on other sites More sharing options...
softworkz 3349 Posted March 9, 2021 Share Posted March 9, 2021 Thanks for the logs. Very good comparison, you made sure that everything is identical: ...well - except performance. I don't think that's normal, but I can't say for sure. I'm not familiar with Docker. What I can say though, is that it's very unlikely a problem at the side of Emby. Link to comment Share on other sites More sharing options...
gewbert 2 Posted March 9, 2021 Author Share Posted March 9, 2021 Yeah looks like i'm seeing about an 8x drop in performance. It's not noticeable when I'm watching something as the performance is generally "good enough" and most things don't transcode. But holy cow it's noticeable when I try to convert something. What else could it be though but an Emby problem? I mean, it could be a docker problem but since I'm using the official container doesn't that make it a problem with Emby's implementation? Link to comment Share on other sites More sharing options...
gewbert 2 Posted March 11, 2021 Author Share Posted March 11, 2021 Sooooo....should I just deal with this performance or what? No suggestions? Link to comment Share on other sites More sharing options...
Luke 37251 Posted March 25, 2021 Share Posted March 25, 2021 Hi, we're not sure yet what the reason is for this, but the upcoming 4.6 server release will have a number of transcoding related changes so you could try again with that once available. Thanks. Link to comment Share on other sites More sharing options...
Luke 37251 Posted June 4, 2021 Share Posted June 4, 2021 @gewbert can you please try again with Emby Server 4.6? Thanks ! Link to comment Share on other sites More sharing options...
Recommended Posts
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 accountSign in
Already have an account? Sign in here.
Sign In Now