March 30, 2010

The Experimental Games

It's been nine months since I started doing Experimental Games and I think it's time to evaluate and maybe go in a new direction.

At first I had been stuck for ages with my never-ending projects and I had literally nothing to show for all my coding. Well now things have changed a bit: I've successfully released seven experimental games, games that more or less are playable, working and sometimes even fun! These months have taught me that games aren't impossible year long projects, they don't have to be perfect to be enjoyable and above all they taught me to get things done!

Here are a few lessons I learned the hard way.

Iterate
Don't try to make the game perfect from the start cause it won't be perfect and it won't be pretty. Instead create the ugliest, crappiest playable version of your game you can and then work from there. The best games where those who got playable the earliest: MenuCity, Bugger and Jonas IceCream Stand. The other crappier games such as A Geek Valentine and Black and White got playable just in a very late stage and indeed they also became the least fun.

Fun isn't created on the spot - it's created through iterations. It's like carving a wood figure. You don't create it with a big slash, you carefully chop the wood off little by little until the figure is complete. This is how I think good, fun games are made.

This quantum leap wasn't achieved with a big chop, but with many smaller ones

Get things done
This is the thing that changed with me the most. At first I thought games where all about the planning stage - the idea stage. It's important to have a great idea and it's good to write down things sure but you can't just sit on your ass doing the big talk and not actually do anything. As said about iteration you can't make the game perfect from the start and you can't the whole game from the start either. The game will change and your idea about what's good for the game will change too.

Take for example my latest game A Geek Valentine. I had envisioned the game as a game where you built trap combos from a top down point of view. But the game changed into a pac-man type (I'm not all too happy about it though). For better or worse games and ideas change, stop planning and start getting shit done.

Set Goals
In December I had the truly great theme New World Order and it spawned the best game idea I've had, a sort of SimCity game but instead of building stuff you plot to take over the city and taint it with your evil propaganda. Like in North Korea or in the old Soviet. But the game was not to be.

Why you ask? I think it's because I never had a goal - I broke the one week rule and I could just push the deadline further and further and I never got anything done. With other games when I was at the 2-3 last days I always got a big energy boost and I always managed to get the game finished. But here I could just say "meh, I'll do it later" and so it never got done. It's Duke Nukem Forever all over again (maybe a bit smaller okay). Goals will help immensely to get things done.

Have fun
In retrospect I should have had an overall goal for my experimental games too. I never put an end date - it was just new month = new game and in the end it wasn't very inspiring to just churn out a game. It was as if I just had to make a game, it wasn't something I wanted to do and it was almost as if it had become a chore.

Mind you I've had lots and lots of fun making these games and the games who were the most fun to play were also the ones I had the most fun making. You could divide my games into three categories I think.

The beginning: Fun
Balls
Black and White

The middle: Really Fun
Bugger
MenuCity
Jonas IceCream Stand
Failed game: The Zed Worker Party

The end: Not so Fun
The Chronicles of Bim: The 100 Fake Afros
A Geek Valentine

The funny thing about this is that it works really really well both ways. Both how fun the games are and how fun I thought it was to make them. I think this is an indication for all you game developers out there: Have Fun, not only will you live a happier and longer life but the game will also be a lot more fun!

Conclusion
I'm not sure what the meaning of this post was. At first I wanted to state a new beginning for myself and then it turned out to a lessons-learned post... but whatever.

I won't be doing monthly experimental games for a while now, I'm happy with the results but I'm not at all motivated right now and I need a little break. I will continue my game making but it just won't be exactly the same - maybe I will try to release a game, any game, each month or I'll focus on making "real" games - games which are everything I want them to be, not just a proof of a concept. I guess that's something that's been bothering me about my experimental games - it feels like something is missing.

March 25, 2010

No game this month

Well here I am in the end of the month - without a game. It's a failure I know but once again I lack inspiration and then it's really hard to do something. Instead I've been thinking of something I could do that would make me wanna code like really bad again. I'm thinking of patching up this website, the forums are bugging me like hell as I didn't write it, switching rendering framework for my games and just now today I've done a few Project Euler challenges in Haskell which was pretty fun (but hard as I'm really bad with it).

Next time I've got my sights on the Ludum Dare which instead of a seven day game is a two day game! We'll see how that'll go.. If I have the power and the time to.

I've also written a small piece about the Evolution of RTS games, heavily shortened to fit a perfect 5 pages a feat I'm quite proud of. Of course I wanted to write a hell of a lot more.

In short: Not much has happened and not a lot is happening but we'll see what the future has in store.

March 10, 2010

Postmortem: A Geek Valentine

Ah man the mush in my brain is finally letting go and I'm starting to feel this tiny little programming urge again... This time it's not Haskell or a new experimental game that's luring me on, no this time it's me longing to create this fantastic awesome epic RTS game. Sadly it's a long way to go there...

Anyway let's get this going!

A Geek Valentine isn't a good game by any means, it's really nothing special. The gameplay sucked really bad. It's kinda funny as I told Sundb00m this would be my greatest game gameplay wise. Yeah right..

Hard Work Work!

This is a small graph which shows how I've worked for a straight week from the 22th to the 28th. The Y-axis is hours and the X-axis is days. The labels are for either work or break and the height of the green peaks if you check it on the Y-axis is the sum of both green and red. Green is for hours where I actually worked and the red is for when I had a break. You know food, clipping toenails, reading manga and playing counter-strike.

Total work time: ~25 hours
Total break: ~15 hours

Weekend work: 13.8 hours
Weekend break: 8.3 hours

The noteworthy thing about this is that I spend more than half of the whole week's worth of time in the weekend and I probably got even more done during that time. I think it's the whole deal with being chased with a the brutal thing of failure that motivates me.

Graphical Adoration

About the game let's start with the positive stuff. I really like how the game looks, it actually looks pretty darn good (even though I forgot about the girls' turning, now they're just looking forward). I'm really improving in a graphical sense, something I thought impossible when I began this journey.

Do something you Like

Like with a Big L. I felt I had a really nice idea going for this one. In the end it looked and felt like a bad pacman clone but this wasn't what I had in mind at all. The idea was to build cool trap-combos sort of like in the Epic Game Evil Genius.

Here's a plan for an über-trap in Evil Genius

I really love Evil Genius and especially the base building but let's face it - I failed. But the idea kept motivating me and it was really fun to try to make it happen.

A first small taste of an "AI"

How do the girls move? They have a 5x5 vision with a 2 radius (they're in the middle) and they choose what to do. If they see the dude close to them either in front, left or right they'll go there. Otherwise they'll try to follow a path in front of them or to the left or right. Otherwise they'll just go in a random direction.

They are extremely stupid and you can trap them in a never-ending loop fairly easy and it's a stretch to say they have intelligence but it was pretty cool to dip my hand into AI programming. I know it's extremely shallow and bad and stuff but still it was pretty rewarding to see them actually move around on their own...

I might actually develop this game further with special regard to AI. Change the core of the game so you can script both the player and the girls, just for laugh and giggles?

The game sucks but hey...

Yupp it's not my best game, in fact I think it's one of the worst. But I still like it a lot and my trap-building idea is still alive and I might develop the game more, improve the AI and focus on building traps (need a lot more cool traps damnet). Yes I think I might do that...

March 2, 2010

March Theme: 10 seconds

Well well here's something interesting. The Experimental Gameplay Project has given us a really interesting thing to focus on: Time. Like Braid, it's time for us to make something interesting with time itself. 10 seconds is the constraint and that's not a lot, but perhaps enough?