0

So I have a simple OpenGL viewer where you can draw any number of boxes that the user wants. Ive also added the ability to take a PNG or JPG image and texture map it to a primitive.

I want to be able to have the user specify any of the cubes on the screen and apply different textures to them. Im fairly new to OpenGL. Right now I can easily map an image to a single primitive, but Im wondering whats the best way to map 2 seperate images (which may be different sizes) to 2 separate primitives.

Ive done a fair amount of reading up on 2D Texture arrays and it would seem this would be the way I wanna go since I can store multiple textures in one texture unit, but I'm not sure if this is possible considering what I mentioned above. If the images are both different dimensions then I dont think I can do this (at least I dont think so). I know I can just store each image into separate texture units but doing it in an array seemed like the cleaner way to do it.

What would be the best way to do this? Can you in fact store different size images into a 2d texture array? And if so how? Or am I better off just storing them on separate texture units?

user3827379
  • 43
  • 7

1 Answers1

0

Texture arrays are mainly meant if you want to draw a single primitive (or a whole mesh) with the shader being able to select between images without exhausting the number of available texture sampling units. You can use them in the way you thought, but I doubt it will benefit you. Another approach (which is similiar to texture arrays) is using a texture atlas, i.e. creating a patchwork of images that constitutes a single texture and using appropriate texture coordinates to select the subimage.

In your case, I suggest simply load each picture into a separate texture and bind the appropriate texture before drawing the cube.

datenwolf
  • 159,371
  • 13
  • 185
  • 298