Posted by jmaasing on Jul 24, 2014; 1:32pm
URL: https://forum.jogamp.org/Need-Help-Solving-OpenGL-GLSL-4-4-Issues-tp4032557p4032657.html

xghost wrote

I'm not exactly sure about the 'strict drivers'. I'm speculating a bit here, but when drivers implement the OpenGL/GLSL specs, the spec is responsible for stating what the result should be, unless we're talking about undefined behaviour. If the shader code is invalid, then it would not compile as shown below after intentionally messing up the vertex shader:

Not really, compliance means correct programs must compile. Compliance does not entail that incorrect programs must not compile. But basically nVidia, AMD and Apple drivers have very different behavior in dealing with incorrect programs.

There actually is a reference compiler but I haven't used it, on my ToDo-list :-)
http://www.khronos.org/opengles/sdk/tools/Reference-Compiler/

But you are right in that 1.0 is a float and should not compile differently than 1.0f. I was thinking of "1" as compared to "1.0", the first is interpreted as an int and not a float which is an error I often make :-)

As for why it works now and didn't work before I still don't know, please let us know if you figure it out :-) I'm betting it's a simple oversight or some simple thing that is done in a different order.