- October 22, 2009 at 7:17 pm #29954
I’ve beed using mbx for few months and i have to port my applications to SDX now 🙂 having a very short time to make some demos i’d like to know if exist any library wrapping OpenGLES1.x on top of OpenGLES2.x (i know, i know), really just to make demos then to write down my applications with OpenGLES 2.x
Thank you in advance.
October 23, 2009 at 10:15 am #33463
Depending on which SGX hardware platform your targetting, it’s most likely that OpenGL ES 1.1 is implemented so there should be no need to port at all. What’s your target platform?
I believe someone was implementing a direct wrapper to ES 2.0, but I recall it being very much a hobby project and I can’t remember the details clearly enough to give you link.October 23, 2009 at 7:35 pm #33464
Thank you for the reply.
I would like to run my application on SGX535 + annother nvidia chip based boards.
Thank.November 20, 2009 at 4:20 pm #33465November 20, 2009 at 4:31 pm #33466
Thanks – that’s the one. It seems to have progresse since I last looked at it. As I say above, on SGX platforms a wrapper is probably not necessary as GL ES 1.1 should be supported directly (and better performance will probably come from utilising this). It’s an interesting project though.
I suppose it may solve the problem of other vendors’ implementation’s missing this functionality for the OP.November 20, 2009 at 4:56 pm #33467
The gles2-bc library is at least one option to use if one wants to leverage old gles 1.1 code but add some gles2 specific visual effects…
Or would it be possible somehow to still run gles 1.1 and 2.0 code simultaneously on a SGX platform which supports both?November 20, 2009 at 6:54 pm #33468
I don’t believe this is possible so you’re right – something like this library might be the only way to do what you say. If you were making an app with a few ES 2.0 enhancements and wanted it to be very easy to deploy on ES 1.1 platforms as well then it might be very handy, esp considering the performance advantage the newer platforms with SGX generally have over the old. I enjoyed reading the thesis; it’s a nice investigation.
However, it’s just not possible to achieve the same kind of efficiency as a driver-level implementation can from a higher level library like this. Also, a large advantage of ES 2.0 is the optimisations that can be made _beyond_ what even a native ES 1.1 implementation can achieve. In addition, replacing the renderer entirely might be less work than it first appears and the expertise gained will be more valuable in the future.