The location of the rendering isn't the only difference here. The idea is to render multiple possible future outcomes, send them to the client, and only use the

The location of the rendering isn't the only difference here. The idea is to render multiple possible future outcomes, send them to the client, and only use the one that matches user input. The traditional way has been to predict player movement based on past movement. Also, I don't see how the improbable trademark issue is relevant to the topic.