But there are of course downsides as well.īaking is a process that we need to go through. We have already looked at some positives. What are the downsides of texture baking? We change the data from being a very dense mesh that can take a lot of time to process into a texture and lower resolution object that combined is much less resource intense. When we bake from a high to a low poly object, we compress geometry into a normal map. When we bake from a shader, we compress the shader into a single texture set, making CPU and Memory savings that would otherwise be used up by the shader. This allows us to fake detail by controlling the direction of light rays based on texture information baked into the normal map rather than the original geometry that we had in the high poly object.īoth texture baking types will save memory and calculations during render time, but in different ways. The normal map texture contains angle data and tells the render engine what outgoing direction an incoming light ray should have. We then bake the difference in geometry between these objects to a normal map texture. One with a dense mesh called the high poly and another object with fewer polygons but with a similar shape that we call the low poly. The other texture baking is when we use two objects. This is the baking we will focus on in this article. One is when we bake a new texture from a material. There are two kinds of texture baking we can do. The data is not only calculated beforehand but also compressed into only the result that we need for the rendering process. You can say that we save both CPU time but also RAM during rendering by making the calculations ahead of time. Related content: The complete beginners guide to Blender nodes, Eevee, Cycles and PBR Only one diffuse, one roughness and one normal map if we are using a PBR workflow. This way we redistribute the load from the CPU to memory.įor texture baking, we also save RAM since we can bundle a shader that may contain 10 or more image textures into a single texture set. Instead, we can bake this information beforehand so that the result of the textures, simulations and animation is readily available, and the render engine can just go grab the information.Īnother way to put it is that we trade CPU cycles into data that we can store on disk. We could calculate all this information at render time, but it would take a very long time and it could be error prone. There could also be simulations or animations that move geometry around in the scene between frames. We may also have hundreds of different materials that each needs to be calculated before we can determine the effects, they have on the light coming into the scene. The baking process moves that calculation away from the render process in order to save time during rendering.įor instance, we may have a material that contains many shaders, image textures, procedural textures, and calculations that the render engine needs to calculate a result for. The need we have is that we have some kind of calculation that needs to be done. This can be simulation baking, animation baking, light baking or texture baking to name the most common ones. Related content: Blender add-on review: Baketoolsīaking is when we calculate some data and store the result. We will cover baking as it is with no add-ons, but if you want an even better experience, I suggest you get Bake tools. In this article I will start by covering the background of baking and some original use cases, then continue with an example and discuss some limitations and more concepts around Blenders baking implementation.
0 Comments
Leave a Reply. |