Refraction Issue

[rpwh]

Sorry for posting again but I was wondering if anyone has any idea what's going on with my dielectric material/refraction. Here is a picture of what's happening:

I'm not sure why the area inside the dielectric sphere's aren't being inverted horizontally and why the reflection from the metal sphere is appearing in the left sphere.

https://github.com/rpwh/ray-tracing

[NicolasGuillaumeSoulie]

There's a mistake in your dot product implementation.

// It should be
pub fn dot(&self, other: &Vec3) -> f32 {
    self.x * other.x + self.y * other.y + self.z * other.z
}
// But you wrote
pub fn dot(&self, other: &Vec3) -> f32 {
    self.x * other.x + self.y * other.y + self.z * self.z
} __________________________________________________^^^

Once fixed the refraction seems almost fine, but you also forgot to return black in fn ray_color() if rec.mat.scatter() return 0. Because of that your metal shine on the edges.

dot product fixed	dot product + ray_color fixed

[NicolasGuillaumeSoulie]

And I just realised that my implementation (that was supposed to be completed) also lacks that black return yet I never had that glow...

[rpwh]

This is awesome! Thanks for taking the time to look through my code and find the fix!

[NicolasGuillaumeSoulie]

No problem.
Your bug just made me curious (although I suspected it was vector based) and since I'm also doing it in Rust I thought it could be a learning experience.