@delta-1 I had similar issue and was the IPAddress and Whitelist setup that was blocking me. Search for those terms, or add your IP explicitly in whitelist and set IPAddress to “0.0.0.0” to test it.

I don’t have access to file now as I built for daughter and it’s at hers.
Cheers

@jimbydude
Update
I used this guide to setup calibration:
https://www.thingsconnected.io/raspberry-pi-7-inch-touch-screen-calibration/

Also, despite the manufacturer saying only supports 1 point on Linux, I definitely see multi-point, tried with six fingers using this site:
https://paveldogreat.github.io/WebGL-Fluid-Simulation/

Update
I used this guide to setup calibration:
https://www.thingsconnected.io/raspberry-pi-7-inch-touch-screen-calibration/

Also, despite the manufacturer saying only supports 1 point on Linux, I definitely see multi-point, tried with six fingers using this site:
https://paveldogreat.github.io/WebGL-Fluid-Simulation/

@Henning I’m not at home for next couple days but i got the 43" IR frame from Shenzhen on ali (I’m pretty sure is same one).

I’m running mm on pi 4. The frame worked out of box, didn’t need to install anything, however finding calibration info was difficult. I can post more detailin that once home.

Having said that it was pretty close without calibrating.

I got 43 inch IR frame from aliexpress, it took a bit to track down calibration steps. It basically worked out of box just plugging into pi 4.

I’m powering directly from pi also, seemed ok.

I’m not at home but can add more detail and link to how to calibrate, once back in couple days.

@Parvez hi,

I took a fork of SMAI that uses recoDNN an doesn’t require any external python running (recoDNN does that), i also detail training.

https://github.com/jimbydude/MMM-AI-Face-Login

Cheers

Just tried this out and works like a charm!

Rihanna now sounds and looks as smooth as intended.

Thanks for the quick and super helpful reply!!

@sdetweil I’m actually running this on Pi 4, so looks like your script is still necessary for anyone looking to play video.

@sdetweil very nice and thank you. I’ll try that out asap!

Cheers for helping :)

@sdetweil Do you mean MM can be configured to use Chrome by default, or you mean just starting Chrome with URL (on the PI) using pm2 etc?

Thanks for the replies, looks like for out of box video niceness, need to wait until electron catches up.

I don’t think the PI itself is struggling. As experiment I played in full screen on PI using Chrome, and at same time had full screen on two separate monitors connected to my local PC and they all played totally fine. So sounds like Electron is needing some TLC to catch up.

I’ll mess with the OpenGL setting etc, and see how they compare.
Cheers

Not sure what Category this belongs since could depend on solution, anyhoo.

So this applies to any of the modules that have embedded playback but purposes of this discussion I’m using MMM-MyVideoPlayer as I like it’s simplicity.

When starting the mirror using the standard: DISPLAY=:0 npm start the mirror starts on the connected TV (via the micro HDMI). So I assume that the default Chromium for the mirror is being used. However videos selected are very choppy.

However, if I start as a server and, on the PI itself navigate to the mirror 0.0.0.0:8080 using Chrome and select same videos, they play perfectly. This is using Chrome browser v78.0.3904.108

Does the Mirror use it’s own specific Chrome Version and can this be updated?

I’ve also updated to latest mirror v2.120
Cheers
Jim