0
Corrigé

Incorrect number of arguments on Fresnel node

H92 il y a 9 ans mis à jour il y a 9 ans 7

Well...so I can't connect Fresnel multiplied with Color in Emission, and Fresnel in Opacity like in screen:


Image 490


I enclose also the full console error:


Shader error in 'Shader Forge/NewShader': incorrect number of arguments to numeric-type constructor at line 169 (on d3d11)



Compiling Vertex program with DIRECTIONAL SHADOWS_OFF LIGHTMAP_OFF DIRLIGHTMAP_OFF DYNAMICLIGHTMAP_OFF

Solution

+1
Solution
Corrigé

This has now been fixed in 1.25, thanks for the help in spotting the issue!

No chance of fix?

À l'étude

I'll have to look into it, not sure what could be wrong

I think I found the error:


1.19 (working):

////// Emissive:
float node_1237 = (1.0-max(0,dot(normalDirection, viewDirection)));
float3 emissive = (_node_1172.rgb*node_1237);
float3 finalColor = emissive;
fixed4 finalRGBA = fixed4(finalColor,node_1237);
UNITY_APPLY_FOG(i.fogCoord, finalRGBA);
return finalRGBA;
}



1.24 (not working):

////// Emissive:
float4 node_1237 = (1.0-max(0,dot(normalDirection, viewDirection)));
float3 emissive = (_node_1172.rgb*node_1237);
float3 finalColor = emissive;
fixed4 finalRGBA = fixed4(finalColor,node_1237);
UNITY_APPLY_FOG(i.fogCoord, finalRGBA);
return finalRGBA;
}

+1
Solution
Corrigé

This has now been fixed in 1.25, thanks for the help in spotting the issue!

Thanks. However I think it would be easy to avoid these types of problems, with a set of advanced shaders for recompiling before each new release. What do you think about it? Can be helpful?

+1

I generally do it with all the example shaders at the moment

Then it could be very useful the shader that I created to simulate an approximate liquid in a cruet (maybe you have seen it). Thanks to this I was able to identify this problem :)


https://mega.nz/#!NZknhBSA!GNRwKPC0RQRv80Ovx9kvK6zEZxyA3lkUhntuvKR7fzI