D-mo

@hollasch your #1390 commit is going backwards in time. Brace initialization has been valid since C++11 (for over 10 years). Judging by the warning _"'auto' type specifier is a C++11...

And, since we are on this subject, there is actually no need to have constructors for simple classes such as `vec3`, `ray` and `interval`. You can just use brace initialization.

> When I change samples per pixel from 64 to 100 (as it is in listing 17) my image is still much noisier than Image 3. In my reimplementation `1000`...

The brighter image part is a dup of #1321. @hollasch this issue can technically be closed.

This is definitely a typo. From the wikipedia artice on [Monte Carlo integration](https://en.wikipedia.org/wiki/Monte_Carlo_integration#Importance_sampling) the correct formula should be: $$ \int f(x) dx \approx \frac 1 N \sum_{i=0}^{N-1} \frac {f(x_i)} {p(x_i)}...

Related to #1302

FWIW, I got different results between 6.1 and 6.2. 6.1  6.2  This seems to make sense, actually. In 6.1 we multiply and divide by...

To make sure the problem was not with my implementation, I took the code from `TheNextWeekend` folder and applied the changes introduced in 6.1 and 6.2: ```diff diff --git a/src/TheNextWeek/camera.h...

Oh hey somebody is alive in here. 😃 I've made quite a few changes. Will send PR shortly

As a starting point I was thinking if I pass 0 length and tweak this line: https://github.com/CasparCG/server/blob/6fec6f8f6b5276db86536dd551e047f2a8f3dcdb/modules/ffmpeg/producer/input/input.cpp#L366 to read : `... || (file_frame_number_ >= out_ && out_ > 0) ...`