Home/ Questions/Q 8866379
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-06-14T16:46:13+00:00

When one calls: glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, texWidth, texHeight, 0, GL_RGBA, GL_UNSIGNED_BYTE, rawData); it’s easy

  • 0

When one calls:

glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, texWidth, texHeight, 0, GL_RGBA, GL_UNSIGNED_BYTE, rawData);

it’s easy to know how to format the data: pass in an array of bytes, in the range 0-255, where 0 is lowest and 255 is brightest for each color component.

When you do the same with GL_FLOAT, you pass in an array of floats. My question is, what is the range of each color component value? 0 to 1.0? 0 to the maximum value of a float?

And you would pass in an array of floats, 4 floats per pixel when GL_RGBA is used, correct?

Also, to confirm: GL_HALF_FLOAT_OES is not a parameter for glTexImage2D, but instead is only for verticex, correct?

My iPad 3 reports “GL_OES_texture_float” and “GL_OES_texture_half_float_linear” as being supported. This means I can indeed use GL_FLOAT as a parameter to glTexImage2D, right? Anyone know if the iPad 2 supports this?

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    The range of floating-point values is… the range of floating-point values. That’s kinda the point of using them. When you fetch from a floating-point texture in a shader, you get the floating-point value stored there. There is no clamping or anything applied.

    If you make a half-float texture, then it’s values have the range of a 16-bit floating-point value. If you make a single-precision float texture, then it’s values have the range of a 32-bit floating-point value.

    • 0