Hi,I am aware that "mediump" in fragment shader of Mali-400 MP2 is FP16.Is it bit exact with IEEE 754 half precision format https://en.wikipedia.org/wiki/Half-precision_floating-point_format or it uses different FP16 format?Please confirm.Regards,Rakesh
Hi Rakesh,
It's not full IEEE754 - for example, it is not required to generate infinities or NaN results in the OpenGL ES specification.
Kind regards, Pete