Hi Mustaf, On Wed, Mar 10, 2010 at 12:59 PM, Mustafizur Rahaman <[email protected]> wrote: > The other GLES library which is returning you [00,00,18] on tapping Red > actor, are you able to detect events properly i.e. if I am tapping on a red > actor is _clutter_do_pick() returning you the correct actor id? >
Yes that worked for me. > Binary representation of 0x18 is 1,1000. When I convert ff to RGB565 format, > it becomes 1111,1000 =>f8 & then the least 3 bits gets ignored.But for you, > it seems the most significant 3 bits got ignored and it became > 0001,1000=>0x18 > > Therefore, to me i have a strong feeling that it is related to endianness > where it is messing up between the most significant 3 bits & least > significant 3 bits. > Thanks I will check that. regards, Amit Pundir > Regs, > Mustaf > > On Wed, Mar 10, 2010 at 12:41 PM, Amit Pundir <[email protected]> wrote: >> >> Hi team, >> >> I'm facing an issue in which _clutter_do_pick() throws me the >> following infamous error when an incorrect pick color is read by >> glReadPixels. >> >> --------------------------------------------------- >> Error: >> "(./.libs/test-events:1635): Clutter-CRITICAL **: >> clutter_id_pool_lookup: assertion `id < id_pool->array->len' failed" >> --------------------------------------------------- >> >> I've been going through couple of bug reports being filed on this >> issue and got a decent overview of how Clutter pick actor colors based >> on glReadPixels() call. >> >> In my case glReadPixels() is returning me the actual actor colors i.e >> for a Red actor i'm reading [f8,00,00] which is Red in RGB565 format. >> Whereas the same working Clutter event example, with a different GLES >> librarry, gives me glReadPixels value as [00,00,18] on touching Red >> actor. >> >> What am I doing/interpreting wrong here? >> It would be really helpful if someone can guide me on correct >> behavior/implementation of glReadPixles() as well. >> >> Thanks for your concern. >> regards, >> Amit Pundir >> -- >> To unsubscribe send a mail to [email protected] >> > > -- To unsubscribe send a mail to [email protected]
