Book 1, chapter 9.6; metal fuzzy reflection

[VxDxK]

class metal : public material {
    public:
        metal(const color& a, double f) : albedo(a), fuzz(f < 1 ? f : 1) {}

        virtual bool scatter(
            const ray& r_in, const hit_record& rec, color& attenuation, ray& scattered) const override {
            ...
        }

    public:
        color albedo;
        double fuzz;
};

Why we can create metal with negative fuzz? I think we need something like that:
metal(const color& a, double f) : albedo(a), fuzz(abs(f) < 1 ? abs(f) : 1) {}

[hollasch]

The code already basically does this. If fuzz is negative, it just flips the X and Y axes when creating a random distribution in a circle. Since there's no particular orientation to the random set of points, flipping in any direction doesn't really matter.

That said, I'm not sure why we bother trying to clamp values > 1. Given that we allow negative values, you could specify fuzz=-20 as a way to get the same effective result as fuzz=+20. +20 gets clamped, -20 does not.

I'm inclined to remove all clamping — I'm not sure what purpose it serves. This feels to me similar to whether darkbulbs should be allowed. I'm on the pro-darkbulb team.

[trevordblack]

I was about to say more or less the exact same thing

[alkanen]

The code already basically does this. If fuzz is negative, it just flips the X and Y axes when creating a random distribution in a circle. Since there's no particular orientation to the random set of points, flipping in any direction doesn't really matter.

That said, I'm not sure why we bother trying to clamp values > 1. Given that we allow negative values, you could specify fuzz=-20 as a way to get the same effective result as fuzz=+20. +20 gets clamped, -20 does not.

I'm inclined to remove all clamping — I'm not sure what purpose it serves. This feels to me similar to whether darkbulbs should be allowed. I'm on the pro-darkbulb team.

Hm, bulbs emitting light-consuming darkness into a scene? I have to admit that I do like the concept.

[hollasch]

Ref #1285