SetPixels and GetPixels of Texture2D behaving unexpectedly

Hi everyone! I am trying to manually set the pixels of a texture2D. Here is how I'm doing it (I've altered the code a bit just to create an example that demonstrates my problem):

Code (CSharp):

1.             for (int i = 0; i < count; i++)
2.             {
3.                 Color color = new Color(randomX, randomY, randomZ, 0f);
4.                 texture.SetPixel(i, 0, color);
5.                 texture.Apply();
6.                 // Print out the first iteration of the loop.
7.                 if (i == 0) {
8.                     Debug.Log("Color: " + color);
9.                     Debug.Log("Texture 0: " + texture.GetPixels()[0]);
10.                 }
11.             }

Strangely, the print results of this code are:

"Color: 2183, -1059, 1152, 0"
"Texture 0: 2183, 1, 1, 1"

It appears that only the first value of the color (its X value) is making it into texture. For the life of me, I can't figure out why.

Thanks for the help!

 

Not sure why you'd get that specific result, but you should note that the rgba values of a Color are supposed to be in the range of 0-1, and an alpha value of 0 means fully transparent, so the color you were trying to set was not a meaningful value.

As a side note, you really should only be calling Apply() once, at the very end of the loop, not inside the loop on every iteration. The whole reason it's a separate call is that it's expensive.

And if you only want to get one pixel, you could use GetPixel(x, y).

 

Thanks! Yes, I'm aware Apply() is very costly. I was just running it here as an example to get the log output I wanted to show you guys.

My texture uses the TextureFormat.RFloat, which takes RGBA values as floats, not 0-1 ranges.

 

SetPixel documentation says: "This function works only on RGBA32, ARGB32, RGB24 and Alpha8 texture formats. For other formats SetPixel is ignored."

Additionally, I haven't played around with SetPixel personally, but my expectation would be that any argument of type Color is going to interpret its data according to the rules of the Color struct, no matter what it is ultimately being used for. That is, Color.red should still mean "red", and if red has to be represented by different numbers, the function receiving that argument should do the conversion internally.

Also, I'm not sure what RFloat does, but colors expressed using 0-1 range for each channel are using floats.

 

Full disclosure, I am using Texture2D to pass positional information into the Visual Effects Graph - not using it for actual colors. Apparently this is a common practice for moving data into the Visual Effects Graph from script - since you only have a constrained set of data types you can expose.

In essence I need to move a set of positional coordinates into VFX - My goal is to create a texture, which has an array of colors. Each color has an RGBA float value - these i had hoped to use as coordinate values.

 

Well, that's pretty thoroughly outside of my personal expertise, so I suppose I shall just wish you good luck.