Traded an OpenRGB issue with a ColorHoster issue...

[3941]

Was using an ORGB QMK firmware for my keyboard before switching to ColorHoster.

I disconnect/reconnect my keyboard a lot, since it's also wireless.

With OpenRGB I always had to rescan devices to make my RGB solution (with Artemis as client) work again.

ColorHoster doesn't seem to require this, however, OpenRGB was apparently doing something to make Artemis reconnect.
Maybe it resets its server, I have no idea, tbh. But after rescan, Artemis would send the RGB signals.

Now it will not. Now I have to go into Artemis and toggle the OpenRGB plugin off and on again.
Which is arguably even more annoying, because the plugin is a lot deeper inside the application, than the ORGB rescan button was.

Any ideas, or maybe an extra function you could add to ColorHoster to automate the whole thing?

[Azarattum]

I think I've seen this behavior. Wrote it off as I though this was an Artemis issue. Didn't know that OpenRGB's rescan button fixed it for Artemis. I might need to look at what exactly this button does. If it just restarts the server, I don't think this would be a good solution. In that case Artemis should handle device list update notifications and act appropriately. Would be good to also cross-reference if there are any related issues in the Artemis repo.

[3941]

Thinking about it...
Maybe it's because OpenRGB never de-registers the device and Artemis just keeps sending.
I wasn't rescanning devices on disconnecting the keyboard, only on reconnecting.

[Azarattum]

Unfortunately I was able to reproduce "rescan works with Artemis" behavior. I don't have devices that connect through OpenRGB natively. For some reason "Serve All Devices" option in 1.0rc2 OpenRGB is broken right now, so I couldn't just proxy ColorHoster devices either. So, for now I don't have any idea how to work around this problem. Maybe it is worth trying to raise an issue on Artemis or RGB.NET repo (not sure which one is actually responsible).

[3941]

Aurora RGB seems to deal with this perfectly, so it's probably not RGB.NET.. I'll go and poke Artemis then, I suppose. Thank you for your time!

[Azarattum]

These might be relevant:

• Feature Discussion - Device Addition and Removal DarthAffe/RGB.NET#360
• Openrgb device updated implementation DarthAffe/RGB.NET#338

[3941]

Hmmm... but if it is RGB.NET, why does it work perfectly in Aurora?
Those are 2 year old issues, too... Maybe just not marked as fixed.

[DarthAffe]

RGB.NET does not support any form of hotplugging currently.
I assume you are right with the assumption that rescanning openrgb just causes the device to reappear with the same ids and continue working as before (with artemis never stopping to send updates) but I don't know anything about openrgb internals.

As far as I know aurora uses a custom provider for openrgb and handles disconnects/reconnects explicitely to work there. Artemis doesn't do this.