AA not working on PowerVR6250

[MichaelJPCT]

Hi Egon,
i tested AAConfigChooser.
with my mali450 device, my app works correctly, either using AAConfigChooser or the 'force 4x MSAA' option in OS.
but with my PowerVR6250 device, enabling AAConfigChooser makes the gfx to be very corrupted (no force 4x MSAA option on the device).
is it normal with 6250 hardware?

[EgonOlsen]

Let's put it this way: jPCT-AE's GL-Renderer has a method called isCrappyPowerVR(), which detect PowerVR chips >=6000 either via their code name or their number and in turn enables all kinds of workarounds to make these suckers work at least half decent.

How does it look with AA enabled? I assume as if some wireframe is drawn over everything? If not, can you post a screenshot?

[MichaelJPCT]

the 2 screenshots show the correct result and wrong result.


problem seems to be that, some part of the screen is not updated, but reverse to older buffered image. when i rotate the camera, part of image is jumping back and forth. the size of unupdated part is not constant.

my app has the machanism that the main scene is rendered into texture then blit the texture on screen. if this is on. AA works correctly. if the scene is rendered directly on screen. AA causes problem.
it is also related to my cloud effect. if the clouds are not drawn, no problem. the clouds are drawn with several passes at different time points and there are some gl state changes, it's quite complicated.

[EgonOlsen]

Yep, that's the exact same issue that I had with these PowerVR chips as well, regardless of using AA. That's why jPCT-AE now defaults to not use VBOs on these chips if detected. That was the only solution that I found that fixed it. Maybe you are enabling VBOs somehow again? Or maybe you are using an older version that doesn't include this fix...are you seeing some PowerVR related setting in Config?

[MichaelJPCT]

my version is 1.32 .
there is usePowerVRTweaks in Config, i don't change it. if it's false, my device crashes.
i don't change vbo setting.

[EgonOlsen]

Then it's the same problem that surfaces in some other circumstances (using AA in this case) as well. I don't see what you do about this. These chips suck. The funny thing is, that it actually renders the graphics somehow (I can't remember how I found out about it though), it just fails to display it.