Book 1: Image 8 contains incorrect rendering after Section 8.3

[ronnieholm]

The image rendered at the end of Section 8.3 appears to be the result of fixing shadow acne in Section 8.4, and should probably be placed at the end of Section 8.4.

Placing the image after Section 8.3 causes one to go hunt through the code for bugs because the image is too dark.

[hollasch]

We'll double check this during the v4.0.0 sanity pass.

[johannmeyer]

I am unable to reproduce image 7 in sec 8.21(The first render of a diffuse sphere) and image 8 in sec 8.22 (gamma correction). My images are much darker than the images shown but removing the shadow acne by setting t_min = 1e-3 corrects it mostly but the produced images are then lighter than image 7 or 8. Image 9 is reproducible though.

[hollasch]

Thank you for the additional information!

[webanck]

I concur, image 8 (section 8.3) is much closer to the result I get after taking care of shadow acne (section 8.4), even with the gamma correction.

[ISchenhaiqing]

I fix this bug.

if (root < t_min || t_max < root) {
        root = (-half_b + sqrtd) / a;
        if (root < t_min || t_max < root)
            return false;
    }

[sphere.h] Ray-sphere intersection code (now)

if (root <= t_min || t_max <= root) {
        root = (-half_b + sqrtd) / a;
        if (root <= t_min || t_max <= root)
            return false;
    }

[sphere.h] Ray-sphere intersection code (after fixing)
The inconsistency between the 2020 version and the 2016 version results in a different energy transfer that makes the current result darker than the previous one.

[tysonliddell]

Confirming @ISchenhaiqing's fix yields an image that seems to match Image 7 in the book much more closely.

Before the fix

After the fix