This is version 0.01 of Nicewrench, a game prototype. Nicewrench was created as a riff on the design of FLYWRENCH, a game created by the independent developer messhof. I recommend playing FLYWRENCH before playing this. As of this writing, you can download FLYWRENCH at: http://www.messhof.com/games/flywrench.php I found FLYWRENCH to be a very original and compelling game. However, I also found it very frustrating to play because it is so brutal. I began to wonder what the game would feel like if it were much more forgiving -- if there were no way to die at all. In a way this was a tiny follow-up to Braid: one of the many lines of thought being pursued in Braid is, what happens when you take a game that is traditionally very player-death-oriented and make it so that the player cannot die? How does the design change, and how do you keep the game interesting? The goal of FLYWRENCH was just to reach the end of each level, but when you can't die, this is not nearly so interesting. Perhaps because I was still in the Braid "wander a world and find collectables" mood, that's what I did here: you fly around and collect stars. Some of the stars you just have to fly through, and you can get them at any time. Some of them, though, are precariously tied to the colored beams; if you fail to pass through the beam on the first try, these stars are lost. Whereas I think this was a good experiment that I learned a lot from, I do not feel Nicewrench is a very successful game; FLYWRENCH is much more compelling. Nicewrench feels clumsy; the player just ends up bouncing off things a lot, and flailing through the level, which is not very fun. At the same time, though, this prototype highlights some interesting gameplay situations that were not possible in FLYWRENCH (because they involve the player bouncing off walls and beams). That's part of the reason I feel it was successful. Another reason is that Nicewrench lent itself to a bigger, more-sprawling level layout, and it was interesting to explore that (even though I did not spend much time on level design, given how quickly this prototype was made). If I were to make Nicewrench 2, which may happen sometime, I think I would try making it a racing game where the player is trying to reach some end-goal and every second counts. The player would not fail to pass through beams if he is the wrong color, but they might disrupt his course and speed. This might permit an even-more-open world in which there are many viable courses, the player choosing the one that best fits his flying style. Hopefully this would bring back some of the intensity of FLYWRENCH, and remove much of the clumsiness of Nicewrench. There are also some basic tweaks I would make to the movement model. When the player is green in Nicewrench, he is mostly inertial but has a little bit of air friction. This causes a lot of bad-feeling situations where the player is trying to drift across a green beam but can't quite make it (though it does feel good once he crosses the beam and is accelerated by it). I think I would make the green mode accelerate the player up to some equilibrium speed if he is moving too slowly, which again may remove some of the clumsiness and stickiness from player motion. This game was written in 2 days or so, during February 2008, while jetlagged on a trip to Copenhagen -- so don't expect much in the way of audiovisual niceness. The latest version of Nicewrench, as well as other prototypes, can be found at: http://number-none.com/blow/prototypes/ -Jonathan Blow 5 July 2008 San Francisco, CA