Being responsible for much of the level design on Egg vs. Chicken, I think I must have logged somewhere around 300 hours playing the game. No seriously, for five months almost all I did was fire eggs into waves of oncoming chickens.

Let me describe my day for you: Wake up. Ride the train to work. Launch eggs at oncoming chickens. Read the Times opinion page. Slaughter more fowl. Eat lunch. Gab with co-workers. Slaughter still more fowl. Go get a cookie. Continue killing birds with their own eggs. Take the train home. Throw eggs at cars on the Brooklyn Queens Expressway. Go to sleep.

Level design requires that you play a level over and over to see if you have properly orchestrated the experience. Is it too hard? Too fast? Too slow? Too boring? You think you have a fool-proof concept for a level. You carefully calculate and set dozens of variables so that just the right number of chickens will pop on the screen and mix with a player's skill and anxiety to produce that elusive grail: fun. After half an hour of setting up the numbers you finally run the level through the game and find that completely unexpected behavior has been produced. All of the weaving chickens are bumping into each other forcing them all to the left-side. So you go back and tweak the variables, adjusting the number of chickens down and spreading them out. Rinse, lather, repeat.

Soon you have played the level a dozen times. The pattern of the level has so ingrained itself in your poor brain that you know where all of the chickens will emerge, where they will walk, where you will have to move eggs. Soon you can play even the hardest level with one hand while facing away from the computer as you kvetch about the stomachache your lunch of Indian food has produced. You forget that the challenge and excitement of the game derive from the surprise. The player doesn't know what will happen next; it's all new to them. In a feeble attempt to restore a bit of this surprise to my play I was forced to drink myself into a stupor each night so that when I came in the next day I could play the levels I had just created without foreknowledge. Okay, so I didn't do that. But now that I think of it, perhaps I should have.

Egg vs. Chicken presented some advantages and challenges as I designed over fifty levels for the game. With so many different chickens, power-ups and special items, I had a wide palette to deploy. I would think of one interesting idea and try and focus the game around that concept. For example in this level waves of little chickens will rush from one side, while big slow chickens creep up on the other wall. I could load one level down with bombs and let players blow chickens to kingdom-come or increase the number of ice patch drops to let players forestall the impending doom. But after a while it all felt the same. The constraints of the board and core-dynamic limited the geography. You were always going to have only four sides to defend. The chickens were always going to march slowly towards the fort. I found that somewhere around 35 levels I began to stall out on ideas. Fortunately, my fellow designer Naomi Clark was making some levels. Just playing her levels revealed interesting new ideas I hadn't thought of. You tend to design yourself into a rut after awhile. You can see why Rothko painted so many damn blurry squares. A fresh perspective is always appreciated; it allowed me to play a few levels like a real player, with the surprise and delight of the unexpected.

During development we also did a lot of playtesting. Sitting behind fresh meat - oops, I mean fresh players - careens back and forth between delight and torture. To watch a player struggle with what to you seemed so clear and easy can really shake your confidence. You begin to think that the system you built is absolutely inscrutable. But watching a player get excited, their pulse quickening as they polish off the last of the feathery foes and cheer with joy and relief at winning is wonderfully gratifying. But even from the failures you learn something. You see what doesn't work and you can usually find a way to adjust it. In response to player feedback we adjusted level difficulty and tweaked core dynamics, like how players move eggs and fire them. We added features like picking up eggs from the field to pull players into a different activity, dividing their attention.

I'm not sure if we ever got the best egg movement interaction down. I think the egg flicking works well given the constraints of moving 2D objects around on the screen with a mouse. In terms of levels, I wish there was more variation. Towards the end of development I discovered a few new board shapes that produced new experiences. If we had more time, I would have liked to have gone through and varied the forts a little more. In the end, I found shape to be one of the key factors in defining the experience of the level. But overall I'm quite pleased with the progression of the levels. It was the first game I worked on at Gamelab. And even I still find the levels in the fifth era quite challenging and exciting. I even lose occasionally. Very rarely. After all, I have the build with the cheat codes.

Essay by Greg Trefry