From Jason Green:

I've evaluated the proposals for both Red Light and Rum Runner and think that both games have a good core idea, so it'll be good to work with them and input whatever I can.

Here I'd like to simply give some advice that I think can be useful for both teams and they can interpret it how they like.

Physically prototyping a video game sounds tricky, personally I've never done that, but storyboarding is something we use in dev. Perhaps storyboard some key components from the game and put them together into a movie? You could draw this or maybe take photos of toys on a drawn background. Perhaps that breaks the physical form assignment? Perhaps you could just get the components together and 'act it out' for this assignment? It's commonplace these days to use actors to act out a cutscene once it has been storyboarded but before mocap, to see if it's flowing and making sense. RedLight could use action figures on a representation of their world, RumRunner the same but with toy cars.

[Note from Bob: The prototyping assignment is meant to test the game play -- so you should try for some playable game rather than a movie. You're welcome to "act it out" as Jason suggests, but should incorporate the computer player in the acting as well in order to capture the game mechanics.]

Keep your core idea distilled to the essential ingredients, chop out anything that isn't essential to the experience and work with the concept in its purest form.

Prove the concept works as quickly as possible, be ruthless and honest, you must know it's engaging at the earliest possible time.

Set your milestones ahead of the framework allowed for the project. Completing a simpler project and having the space to fine tune and polish it is a way stronger position for a dev team than rushing to complete a more complex project.

Work out an 'elevator pitch' for the project, a single, memorable sentence that describes the essence of the game.

Get the controls to feel smooth and intuitive, be cautious of including power-ups or mechanics that degrade the controls, this can be frustrating and push players away from the experience. In Split/Second, even if a car is pretty messed up, it still keeps 100% of its handling intact, it's frustrating to have 'numb' inputs and reversing controls momentarily is always an alienating experience.

If you have power ups, work with a memorable number and don't overload it. When you've trimmed the suite of power ups to the essential core, mention them to a friend or partner one day and see what they think. Then, the next day, ask them if they remember what they were, if they do then that's a very good sign, if not then maybe there are still too many or they're not distinguished enough. Too many power ups can also be tricky to balance.

Mock up the HUD and any front end quickly. Know exactly what you're making and agree on it in a pre-visualised state before you attempt to make it functional, otherwise you can explore some expensive blind alleys and possibly dress the screen with a cluttered HUD that doesn't complement gameplay or visuals.

Put mood boards together early, know the visual style and palette you're after before making assets.

Agree on the time of the setting, it'll help establish limits to work within, RumRunner already have this in the bag.

Keep the screen in proportion, don't overly detail characters or vehicles and drop them into a low-detail world, keep everything consistent and it'll all hang together.

Keep control inputs simple, only put in what's required to engage the player.

Keep this quote in mind, perhaps it's a little extreme, but there is value in the message: 'If you have five minutes to do a project, spend four minutes figuring out how to do it and one minute doing it' - Disney Imagineers.