This enables some advanced background functionality for P96, and seems to have no performance drawbacks so far.
It is however possible that this may cause some problems with applications using unusual color formats for their bitmaps.
There is apparently no consistency whatsoever to the value reported when reading this register, and thus this can only be allowed to be enabled manually.
To test this, install AHI (probably preferably 4.18) and copy pi-ahi.audio to DEVS:AHI and PI-AHI to DEVS:AudioModes.
The IRQ trigger magically doesn't work, so this all runs of vertical blank... meaning that it stops working pretty much the instant you try to use it in an RTG mode.
Not very useful at the moment, but I need to take a break from working on this to do a bunch of work.
* Do not explicitly set INDISPLAYCHAIN flag, let P96 handle this through the tooltypes.
* Remove 030 card file, was identical to 020 version.
* Remove 020i card file, currently not working.
* Update build script, remove build of 030 card file and disable building 020i card file.
When the Amiga reboots with DPMS enabled but not asleep the HDMI is
reset which breaks the Raylib window.
This patch only resets the HDMI if the thread has been stopped first.
Also fix thread leak when DPMS terminates the RTG thread.
The first version can cause the RTG to reinitialize badly, causing
PiStorm to die. This version will gracefully shutdown the RTG thread if
DPMS is enabled and start the thread again when it wakes up.
This uses the Pi monitor control to turn the monitor off and on. Turning
on requires a full reinit of RTG including creating a new InitWindow so
the reinit functionality has been changed slightly.
It is enabled by setting the variable "rtg-dpms" in the config file.
This temporarily breaks the PiStorm temperature display, since I had to adjust the register offset for it.
Scale filtering currently doesn't work properly, probably due to unexpected use of shaders, will look into it later. Stick to point filtering for now.
Write to VPOSW before reading Agnus id seems to fix cold boot Agnus id read failure
Add "force-move-slow-to-chip" config option to force-disable Agnus check in case issue persists
