The tables have turned! We made our own games with DIY game-tool Kodu, and Microsoft reviewed them. Will any of us make the grade?

Kodu is for kids. Straight up: Microsoft Research created the homebrew game-creation tool to let students make their own games and learn about programming in a fun and easy way. For adults who want to make independent games that run on Xbox 360, there’s the full-blown XNA Game Studio. For everyone else, there’s Kodu.

Well, we’re just like everyone else. We tried and failed to wrap our feeble minds around XNA, but we still had the lingering desire to make games. And if Kodu — a $5 Indie Game that makes, well, indie games — lets third-graders live out their game-creation fantasies, it should damn well work for us.

So after a few hours of messing about (really, only six or so per game), we asked Kodu’s creators at Microsoft Research to review our creations. That’s right — after years of passing judgment on games, we’re going to be on the receiving end. This could get ugly.

The Judges

Matthew MacLaurin
Lead “Kodu Game Lab” programmer, Microsoft

Matt is Kodu’s daddy, and the man knows his interactive stuff: He was a key member on the Windows Vista user-interface team. He’s also done a lot of work on data storage systems, worked at Apple back in the early ’90s, and was even a part of the team that worked on Myst Online at developer Cyan.

Scott Austin
Director of digitally distributed games, Microsoft

Scott joined Microsoft in 1999 and has been taking care of all that XBLA goodness for several years now. If it’s a downloadable game, it’s his bailiwick. He’s been intimately involved in the Community/Indie Games initiative, of which Kodu is a proud part.

Senior Editor Ryan presents...MehBike!

Bungie’s not going to be hiring me anytime soon. Beginning with a blank slate, I built my first gameworld far too small. After expanding it exponentially and running a few failed experiments with decorating my gamespace with water, I went wild with my world’s elevation. Using the hill-creation tool and then the smooth option, I crafted some bumps and uneven surfaces and then laid down some track. After spending a little too much time creating a village in the center of the course — complete with its own factory and backstory — I turned my attention to getting the actual racing component to work.

A lengthy but simple list of rules in Kodu’s programming script is all it took to get MehBike on the road. It was easy to make the left stick move the motorcycle and set LT and RT to brake and turbo, respectively. What I couldn’t figure out was how to fix the camera behind the bike at all times, so I had to leave the third-person camera controls in the hands of the players using the right stick. Oh well.

My real problem was the timer. My goal was to have players race around the track once, collecting as many coins as they could while finishing as fast as possible. But I couldn’t find a way to make a running clock. I opted to force a game-over at 20 seconds (a number arrived at after playtesting), with a point penalty incurred for each second after 15. That was my missing ingredient. After that, MehBike — named for its admitted inferiority to childhood NES favorite ExciteBike — was fully operational and ready to go!

The Verdict

Matt's Take
+ The scenery is fun — factories in the background are a nice touch. The thermometer is orange, which tells you that you’ve packed a bit too much in there; you’ll get a better framerate if you thin it out.
– You’ve sped up the bike a bit — 1.5X normal speed. I’d go even faster — after the Kodu update, you’ll be able to go up to 5X.
– You’ve gone with the narrowest road, which is tight for racing. Try other road types (select road, D-pad up/down).
– Way short on gameplay. It’s basically “move forward for 20 seconds, then die.” There’s not enough tension or reward to keep it fun.

Score: 4.5

Scott's Take
+ Clear instructions
+ Humor
– 20-second timer
– Track is hard to maneuver at maximum speed
– No end score summary, so no motivator for replay to improve score
– Easy to cheat by jumping into center world

Score: 2.5

Senior Editor Dan presents...Space Paranoids!

I began by thinking about Kodu’s strength: arcadey action. We don’t really need another twin-stick shooter, but I felt compelled to make one anyway, with the working title Paranoia. My hook: It’s a multiplayer deathmatch arena where four flying saucers avoid obstacles and snag power-ups…basically, violent hide-and-seek. I started with Empty Stadium as my basic arena and began populating it with…stuff.

A few hours later, I was tweaking everything: how much damage bullets cause, what happens to power-ups once they’re collected — the works. Once I tested some of the mechanics — the balloons weren’t scattering and they weren’t healing players when popped — adding little details made it feel more like a game. I kept tweaking the lighting, and I added a color-coded glow effect to the saucers. When players get below 20% health, their glow starts cycling through different colors. I thought it looked and played pretty nicely for something that I spent about six hours on. And in honor of my retro roots, I changed the title to Space Paranoids. Greetings, programs!

I learned a few tricks that made things easier as I went. One, create and alter one object until it behaves exactly the way you want it to, then clone it (not the other way around). Two, don’t throw everything in at once. You can always go back and add or change things once you get it running, and you can save multiple versions of your game — assign version numbers and the Y button to save an iteration.

The Verdict

Matt's Take
+ A variety of actors, well-suited to their roles with respect to speed and movement style
+ Nice balance of threats and goodies. Simple party fun.
+ Cool look (although some dynamic lights would have made it even better…)
+ Good point of departure for a multi-user code-and-play; would be even better if each user could improve their character or add a new type of A.I. enemy
– Balloons go wandering off into the weeds. You should assign them different paths within the playfield.
– Would be better if things got more intense over time — perhaps each minute or two a new type of enemy could enter the playfield.

Score: 7.0

Scott's Take
+ Multiplayer dual-stick shooter
+ Clear instructions
– No single-player mode
– Environment isn’t destructible

Score: 5.0

Art Director Juliann presents...m33p!

At first glance I was thinking: “This is for kids? Good luck, kids.”

I opened two different tutorials and the lack of guidance made me lose interest immediately, so I decided to just start fresh and figure it out. This worked well for me, and after I got past the 20 minutes it took to realize you have to lay water OVER land, the rest seemed way easier to do than I had initially thought it would be. Also, the cute-as-hell interface helped keep me going just to see what the different tools did.

I actually started four different times, but I’d get to a certain point and think: ”No, this isn’t it. Start over.” So with the small amount of time I had left (our internal challenge deadline was pretty intense; I would’ve liked to have had more time and add more stuff), I ended up making a game as it came out…no planning, which probably shows. But it is kinda cute, and it totally exposes that I’ve spent countless hours playing platform games.

In the world of m33p, your little m33per — who’s very talkative — cruises around eating apples for 5 points and hearts for 10 while being shot at by spaceships and steamboats. When their bullets hit you, you lose points, but shooting them will give you more points back. The game ends when you find a star (I couldn’t help it), which isn’t so easy to find.

It was a very low-stress experience overall — in the real world as well as in-game. I look forward to making many more!

The Verdict

Matt's Take
+ A variety of characters — flying, sailing, and, ah, hovering
+ Extra points for having an ocean
+ Points for control use — shoot, secondary shoot, and jump
+ A variety of interactions with different actors. (But I didn’t realize the potential until I looked at the script.)
– Terrain is nearly impassable. Use the flatten tool or the road tool to make it more obvious where you’re supposed to go. Rethink the terrain to be more of a progression rather than an open wander, and a lot more of your programming efforts will pay off.
– I can shoot, but nothing is at my height, so I can’t seem to actually hit anything. Adjust Kodu’s height by selecting and using D-pad up/down.
– Is there a goal? I see the hearts, but I can’t seem to win or make progress. A little RPG-style mission would fit nicely with the style. At a minimum, use the “say” verb to explain the basics to the user.
– Most of the cool triggers didn’t actually occur. Try constraining the character path so you can make sure that each trigger fires, and that the user notices.
– Gentle is fine, but this could get a lot faster and still be mellow. Use the object settings to multiply the speed (up to 500% with the Kodu update).

Score: 3.5

Scott's Take
+ Destructible environment
+ Finding the star is rewarding
– Game sometimes ends suddenly with no clear reason why
– Wish I had the ability to control speed
– Wish I was able to eliminate enemies
– The game needs instructions

Score: 3.5