2 ms·
It's not necessary, but it's convenient. What's actually happening here is Monte Carlo integration. Randomness guarantees you get the right answer. It prevents
by dahart 7y ago
It's not necessary, but it's convenient. What's actually happening here is Monte Carlo integration. Randomness guarantees you get the right answer. It prevents correlation artifacts. It also gives you a nice film grain look when you don't have the time to take samples until complete convergence.