[Suggestion] Improve `sync.monitor` color test

[mbsck]

I think an additional test like this colorwheel should be implemented for the sync.monitor.

After having implemented the X11 support, I ran the colorwheel test and noticed that sometimes the colors are on the wrong sides (all individual colors are correct).
To check this, I looked at dxcam where the output is RGB, and the output of the mss is BGRA where I removed A and reversed the order

screen = np.array(screen)[..., [2, 1, 0]]

However, in the video some of the colors seems to be on opposite sites. With the full screen color test this was not the case.

Maybe there is some reversion of happening somewhere? Or is the RGB format the expected output for the PIL.Image.Image?

[Minlor]

Are you talking about the location of the colors not aligning with the led location on the monitor and on the screen? ex. Something red being on the upper left corner of the screen but it doesn't align with the location on the led strip? 🤔 if that's the case I believe that it's just due to the led strip and that something would need to be implemented where the user will be able to actually configure which part of the screen will reflect to which part of the led strip since maybe the orientation of it is different to ex. what I have tested it and tried it out, etc.. I just hope that I understood you correctly

[mbsck]

Basically, when the full screen was, for instance red, then it showed all red on the LEDs, but if there was, let's say green in the top right, and blue in the left bottom it would show it the other way round (green bottom left, blue top right).

It still might be a problem of the configuration and it might be the case that my LED lightstrip is improperly configured…

However, as you said it might be nice to implement some functionality in the long term to be able to configure that. I think that is probably part of the Windows software, right?.

I saw you just recently create an issue for that. Awesome.

Also on a side note. I will probably be a bit short on time until the end of this month, but will at latest then invest more time to finish the remaining Unix implementation (and if I have time in between work on this then).
Afterwards, I could also take on other issues :)

[Minlor]

Basically, when the full screen was, for instance red, then it showed all red on the LEDs, but if there was, let's say green in the top right, and blue in the left bottom it would show it the other way round (green bottom left, blue top right).

It still might be a problem of the configuration and it might be the case that my LED lightstrip is improperly configured…

However, as you said it might be nice to implement some functionality in the long term to be able to configure that. I think that is probably part of the Windows software, right?.

I saw you just recently create an issue for that. Awesome.

Also on a side note. I will probably be a bit short on time until the end of this month, but will at latest then invest more time to finish the remaining Unix implementation (and if I have time in between work on this then). Afterwards, I could also take on other issues :)

Yup! So in this case it would just be because of the way you pretty much put up your led strip, my one is setup counter clockwise and that's kinda how I made everything work so a implementation for configuring this would be needed I'll see if I get some free time to implement this myself but I'm unsure at the current moment I'm also thinking about possibly just doing like a web view or something for the GUI since it would make it way easier to just work on it so maybe there would also be a nice way to implement a configuration for these