Yes, currently the black screen is shown instead of a screensaver over roScreen. I don't think anyone is talking about private screensavers here. What Komag is proposing is to DISABLE the screensaver, by periodically sending ECP commands, and then have his main app behave like a screensaver when appropriate.
Oh i see. Seems i was mislead when further TheEndless, squirreltown and RokuJoel talked about "own screensaver" and then about testing pvt.SS w/timeout (new subject apparently).
So a private SS (implementing RunScreenSaver()) cannot interfere with a "black death" event. I.e. cannot say have runScreenSaver() that just does a `return` when invoked or ECP sends a key to wake up and return to the core app. Because it won't get invoked at all when roScreening.
Hmmm... radical question here!
Aren't periodic calls to ifScreen.swapBuffers
() and .Finish
() enough of a reason to ward-off the Black Screen?
I mean how the SS watchdog decides if it's time to start (imaginary)saving the screen - it checks for recent events like remote I/O and video component playing. It can do the same based on whether roScreen graphic activity is happening or not. Now that is a heuristic since one might be drawing and never SwapBuffers() - or swap them while keeping the same image. But same can be said for roVideoPlayer - it may loop a video of the same still frame for many hours and ostensibly burn-into a cathode tube or plasma display after left for days