Butterfield's presentation focus on the challenges of making a game that seemingly transitions from 2D to 3D without effort. Crush is designed so that your character, who is under hypnosis to cure insomnia and forced to find a way through 40 dream sequence levels, can only "crush" (transform 3-dimensional space into 2-dimensional platformer) horizontally and top-down.
Of the challenges discussed, Butterfield talked about the shortcut challenge, whereas the clever player would be able to use the crushing ability to move from the beginning of the level to the end without effort. Thus, the team devised three different blocks whose behavior differed when crushed: ghost block (insubstantial), solid blocks (impassible) and hollow blocks hollow block (somewhere in between).
Other issues include disorientation, which Butterfield tried to fix by way of camera work and the placement of prominent landmarks to help the player. There were two design flaws that the dev team worked to solve. The creation of cut-zones let the developers section off part of the level so that one does not accidentally crush themselves next to a horrid enemy without prior knowledge. A safety feature was also implemented so that a crush process does not cause the character to fall helplessly to their doom; "an explanation as to why you failed the puzzle" will be shown instead. Many of these problems were also solved, of course, through countless QA and debugging.
Following the talk of problems and problem solving, Butterfield moved into the realm of conjecture, philosophy and other higher-order thought. What if you were allowed to crush at any angle? What if you could crush outward into four dimensions (with time being the 4th candidate); i.e. a block could become a bridge, a cockroach would crush into a centipede, etc. How about multiplayer? Butterfield suggested separate realities for each character, whereby only you control the crushing in your reality and only your movement would be reported to the other player's screens.
There's no clean-cut solution, but some of these problems give us an idea of the far future of platform puzzlers. Crush is looking good and the game's twist well-executed. The game is slated for release later this year on the PSP.