Don't mind at all.

The new Android 2.0 phones would log EGL_BAD_SURFACE and show nothing
on the screen.

Problem was with my GL initialisation (which I think I copied from a
Google Android 1.1 Open GL example):

int[] configSpec = {
        EGL10.EGL_RED_SIZE,      5,
        EGL10.EGL_GREEN_SIZE,    6,
        EGL10.EGL_BLUE_SIZE,     5,
        EGL10.EGL_DEPTH_SIZE,   16,
        EGL10.EGL_NONE
};
EGLConfig[] configs = new EGLConfig[1];
int[] num_config = new int[1];
egl.eglChooseConfig(dpy, configSpec, configs, 1, num_config);


And here is their solution e-mail (I hope they don't mind me posting
it):

Craig,
After doing some testing and speaking with some of the other Android
engineers, I believe I found the problem.

In your EGL configSpec, you've specified the following:

               EGL10.EGL_RED_SIZE,      5,
               EGL10.EGL_GREEN_SIZE,    6,
               EGL10.EGL_BLUE_SIZE,     5,

With those three parameters, the sample app I was running displayed a
black screen. If I removed them, it worked fine.

When you specify the EGL_*_SIZE, it means you want AT LEAST that value
for the component. Also, EGL specifies that the EGLConfig are sorted
such that the highest total number of bits are first. Hence, the 8888
config will come first when you request (at least) 5 6 5.

The EGL spec also says that if EGL_*_SIZE is not specified or is
EGL_DONT_CARE, then the configs are sorted in order of the LOWEST
EGL_BUFFER_SIZE, so 5 6 5 will come first (unless there is 555 or
444).

We had a bug in previous versions of android where we were not sorting
the configs properly.

It's better to use GLSurfaceView to make sure things work properly. If
you insist on wanting to use eglChooseConfig yourself, you MUST go
through the list of returned configs and pick the one that matches
5,6,5 (or whatever the window format you selected).

You can also use EGL_DONT_CARE, which has a greater chance of
returning 565.

Let me know if this helps.


On Oct 28, 8:50 pm, String <[email protected]> wrote:
> On Oct 28, 7:20 am, CraigsRace <[email protected]> wrote:
>
> > I should also mention that Google were very helpful in tracking down
> > the reason that Android 2.0 broke my apps.  Thanks Google!  :)
>
> Lucky you. Mind sharing what the problem was? I've filed a report on
> one of the regression bugs that broke my app, but am still nailing
> down the second one.
>
> Thanks,
>
> String
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google
Groups "Android Developers" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/android-developers?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to