Thank you! I got it working. I’m still getting choppy playback but it’s not from chrome now. I’m on an Apple TV. So it might be the issue that was being noted above.
1 Like
For FruitDeepLinks sports on PrismCast, I've been playing with PrismCast and the /play API endpoint. It was really straightforward to stream directly to PrismCast without having to build new channels within the PrismCast web interface. Similar to CC4C in use. AND HBO Max links click the play button to start those sporting events which CC4C never did 
Would you consider an FruitDeepLinks "multisource-lanes-PCast.m3u" output file in the Prismcast /play format for ease of use? It would need the PCast url and port similar to the Chrome Capture M3U.
The CDVR custom channel settings source text I'm using as an example:
#EXTM3U
#EXTINF:-1 tvg-id="lane.1" tvg-name="Fruit Lane 1" tvg-chno="25001" group-title="Sports Lanes",Fruit Lane 1
http://192.168.50.18:5589/play?url=http://192.168.50.18:6655/api/lane/1/launch?deeplink_format=http
#EXTINF:-1 tvg-id="lane.2" tvg-name="Fruit Lane 2" tvg-chno="25002" group-title="Sports Lanes",Fruit Lane 2
http://192.168.50.18:5589/play?url=http://192.168.50.18:6655/api/lane/2/launch?deeplink_format=http
2 Likes
After playing around with it more, I was wrong the error is still on the sever side. It will skip in chrome which makes the playback think the fps changed which then creates a larger skip or buffer in the playback. This happens on my shield, chromcast and Apple TV.
Besides being a "10th gen", what are the full specs on your server? Also, can you post what you used to spin this container up?
Thanks for your help. i9 10850k, 32 gigs of ram. Intel arc, igpu and nvidia card. I’m not sure what you mean about how I installed it. I pulled it from the conversation above - I’m still learning.
I think there might be an issue with hardware acceleration in the docker. I’m trying to give the docker either the igpu or the arc but it’s not seeing them - both are idle when playing.
1 Like
Did you use the Docker Compose I posted a few back, or something else?
It's a known issue with the Linux version of Chrome, not Docker.
I used docker run -d … /prismcast:latest
Could you post the full command you used please?
Thanks again for your help with this.
I tried a few:
My last run was an attempt to give it the igpu and avoid the nvidia card so I did:
docker stop prismcast && docker rm prismcast
docker run -d
--name='prismcast'
--shm-size=2g
--device /dev/dri/renderD128:/dev/dri/renderD128
-e "RENDER_DEVICE=/dev/dri/renderD128"
-e "NVIDIA_VISIBLE_DEVICES=none"
-p 5589:5589
-p 5905:5900
-p 6085:6080
-v '/mnt/user/appdata/prismcast':'/root/.prismcast':'rw'
--restart unless-stopped
'Package prismcast · GitHub'
My first few attempts were more like what the GitHub show as didn't give the docker the igpu directly or change the shm size
docker run -d
--name prismcast
--shm-size=1g
-p 5589:5589
-p 5900:5900
-p 6080:6080
-p 5004:5004
-v prismcast-data:/root/.prismcast
How do I completely delete my previous prismcast setup and re-install? I sucessfully updated using Portainer, but the channels seem to be a combination of my previous and the default lineup - I tried deleting the container and deleting the image and re-deploying, but that doesn't work
1 Like
If your PrismCast data is stored in the Docker Volume prismcast-data, stop the stack and then delete that volume in Portainer-Volumes. Restart the stack, and everything will be as new.
The only potential issue I see is the above. What is the path to the left of the colon mounted to?
You mean like where physically? It's in the cache.
That should be fine -- I just wanted to make sure it wasn't an NFS or CIFS share.
I'm running out of ideas. There was another user (@mackid1993), that reported some strangeness with Docker on Unraid -- and as I recall, stuttering was described. Though in his case it was with the CDVR Docker container that was giving these results.
You could try creating a Windows VM, and installing PrismCast directly, especially if your NVIDIA card is available to passthrough to that VM, so that you'd get a GPU in on the fun.
I was having issues with Docker, Unraid, and SHIELD. The Shield would send RST storms and Windows fixed it.
I ended up giving up switching to Apple TV.
1 Like
Thanks. I'll keep trying too. Chrome shows it's hardware decoding but then all the gpus (including the igpu) are inactive, so something is confused. Thanks again for your help.
I dont think OP has implemented any GPU support the docker version of prismcast so...more likely than not, you may have more success in native and in windows
Agreed - there are just so many toggles in chrome at :flags and :gpu that I thought I would try to and force it. But yea, it does not look like it's supported. Not sure why the 12 gen is doing well while the 10th is not. I am sure someone smart than me knows.
I'm running PrismCast on a Mac, with the original install method (npm install). Recordings have been great, but a general question about Chrome running all the time. When I use the computer for other things, Chrome constantly jumps to the foreground even when I'm not recording anything. Is this a requirement of the implementation, or is there a workaround for this so I can avoid being interrupted?
I'm running PrismCast on a Mac, with the original install method (npm install). Recordings have been great, but a general question about Chrome running all the time. When I use the computer for other things, Chrome constantly jumps to the foreground even when I'm not recording anything. Is this a requirement of the implementation, or is there a workaround for this so I can avoid being interrupted?
It’s a function of how Puppeteer works. There isn’t really a way around it.
More broadly, I wouldn’t recommend actively using the Mac you’re running PrismCast on—treat it like a server and leave it alone. I understand that may not be possible in your setup, but it’s worth calling out.