Compiling Vertex program with UNITY_PASS_FORWARDADD SPOT _COLORIFYMASKED
Platform defines: UNITY_ENABLE_REFLECTION_BUFFERS UNITY_USE_DITHER_MASK_FOR_ALPHABLENDED_SHADOWS UNITY_PBS_USE_BRDF1 UNITY_SPECCUBE_BOX_PROJECTION UNITY_SPECCUBE_BLENDING UNITY_ENABLE_DETAIL_NORMALMAP SHADER_API_DESKTOP UNITY_COLORSPACE_GAMMA UNITY_LIGHT_PROBE_PROXY_VOLUME UNITY_LIGHTMAP_RGBM_ENCODING```
Odd because the line that's pointing to:
```UnityLight light = AdditiveLight (s.normalWorld, IN_LIGHTDIR_FWDADD(i), LIGHT_ATTENUATION(i));```
Is calling the function with three parameters.... I hate the shader compiler so very much.
Hi @CatsPawGames , I’d like to get this package but according to the Asset Store you haven’t updated the asset since last November. I guess this means no support for 2018.X (and 2017.X) yet? Can you let us know when you have updated?
PS: I personally don’t need support for the new, scriptable render pipelines…
Hello @CatsPawGames I got an error when I used it. This error caused the entire shader to be unusable.
Is there any way to solve it?
I tested three different computers, two computers went wrong, only one computer can be used, I am not sure what the reason is.
I hope this can be solved. Grateful
@CatsPawGames I succeeded! It’s version 2018.1.0f2 problem!
There is a problem with the 2018.1.0f2 version, so please don’t use it. This version will cause the standard shader of colorify to be unavailable.
Update to 2018.2.3f1 to solve the problem!
Hi. I need to add a roughness map to my models. Will that be possible with the shaders you have provided ? Currently I am using the Standard Roughness Setup shader.
We’re trying to really trim the build size of our scene bundles down for WebGL. The Colorify shader portion seems to be pretty large in the build. Do you have any tips for how we can trim slim down the size this takes up in our bundles and main build? Thank you!
Hello. One possibility: take note of what shaders from Colorify you’re using, then delete all unused ones. I think unused ones won’t be included in the build by default - still worth checking.
If you’re using Standard Colorify shader, it can be a problem for build size too since it compiles into multiple variants. There’s not straightforward solution for that unless you’re willing to dive into shader source code.
Given a white sprite with black dots (anti-aliased), is it possible to colour shift the background and the dots separately without running into edge/halo artefacts at the boundaries between the dots and the background colour?
I have tried writing my own shader to do this but always seem to end up with some edge artefacts that look OK at certain colour combinations but look bad with others.
I don’t think recoloring would cause any artifacts, especially if you’d create recoloring map in image editor instead of using auto-mapping feature (there’s description of recoloring map format in manual, but basically R = 1st recoloring are, G = 2nd recoloring area). There’s 2 available algorithms for recoloring - based on linear RGB difference and on greyscale multiplication.
That said, using Colorify for this task is a bit overkill. Then again, performance impact is negligible so if it’ll work for you it’s fine.
I’m looking into using Colorify for a project, but I’ll likely end up needing the ability to change 3 colors rather than just 2. How difficult would it be for someone who has never touched shader code to add a third color to one of them? If it’s not particularly simple, how likely would it be for you to add third-color versions to the asset? Thanks.
I’m more or less in the same boat. As much as I love the asset, It’s a bit surprising that there isn’t a three or four color version already given that there are four color channels.
We would like to use it to color our low poly ships. Best with 4 colors, but it also works with 2. We would like to use ID Maps to tell the program exactly which areas need to be recolored in which color. Is this possible? So far we haven’t found a shader that makes this possible. The Recolor Shader lets parts of the ship disappear completely. Are we doing something wrong?
