Chibi Knight: Postmortem - By BoMToons

At the beginning of 2010 I released a Flash game called “Chibi Knight” which can be played here.

To-date it is, by far, my most popular game with millions of plays on many websites. I’m not quite sure what made Chibi Knight so appealing, but I’ll try not to bore you too much while I explain the process that brought the game to light.

I. My Favorite NES Game


First off, let me mention that I HATE RPGs with a melting fury. I find it absolutely boring and frustrating to have to sit there and wait for an enemy’s “turn” when I could clearly be running away, dodging, casting a spell, and attacking all at once. If you want turn-based game-play, play a board game where turns make sense. The whole point of vidya gamez is that they have real time action and adventure – pulse pounding predicaments precipitated by pixel proximity.


But, after having said that, there is one RPG I loved as a kid called Zelda II: The Adventure of Link. First off, what the flip!? The hero’s name hasn’t been “Zelda” this whole time?! Mind = Blown.

Second, hey, this isn’t an RPG because I have to button mash while jumping and dodging to kill enemies…this is actually fun! Oh, and my character getting stronger the longer I play, that’s pretty entertaining… so that’s why people play those turn-based RPGs!

I actually made a Zelda II minigame back in 2007 in my game “Boss Bash” which can be played here.


Making it reminded me just how cool Zelda II was and how cool it looked and felt with updated graphics. It also reminded me that it had an awkward spell system and some other little annoyances like a tall sprite that has to duck sometimes.

II. Castle Crashing the Beard

In 2008 I heard that Tom Fulp wasn’t shaving his face until he finished Castle Crashers, so while chatting with a Flash artist friend (Luis) I mentioned we should make a game about that. So two weeks later we released “Castle Crashing the Beard” which can be played here.

On some level, I was channeling the Zelda II Boss game I’d made in 2007, but Luis brought some really unique things to the gameplay just by the way he drew the game sprites. He had recently worked on “Newgrounds Rumble“, a brawler with varied “chaining” attacks for each character. Luis’ integration of “brawler” style attacks into a 2D platformer felt REALLY nice, and the quick “level up” dynamic with changing costumes was really addicting.


People loved it, and The Behemoth got so much traffic from the game they included Luis and my names in the “Castle Crashers” credits! For a while afterward, all people could say was “make this a bigger game!” So, almost immediately afterward, I started tooling around with a more full-scale treatment of the game-play in CCTB.

III. Expanding on the successful concept

At first I was just going to make a straight up “use your fists” 2D brawler with an overhead map view and some RPG elements mixed in and a main character whose physical appearance changed as you leveled up (shout out to “Altered Beast”).

I ran the idea past some artists with an animation list that looked like this:

Sprite Upgrades: Scrawny, Robust, Thick, Buff, Huge, Giant, Maxed out
States for each upgrade:
1. Rest
2. Weak Punch
3. Strong Punch
4. Duck
5. Weak Duck Punch
6. Strong Duck Punch
7. Special Big attack
8. Walk
9. Jump
10. Jump Attack
11. Knocked Down
12. Die
13. Hit

So that was like 13 animations * 7 states which is like 91 animations just for the main character. Yeah, no one was jumping at that opportunity. But I was gonna go easy on the artist and make the overhead “world map” tile-based so the map could be huge.

After the idea sat around for over a year, I decided that I just wanted to get this thing out there for people to test and see if it was really going to be worth expanding on.

IV. Simplifying the expansion

So, I didn’t want to be at the whim of an artist for my art assets and I have some art skills so why not do EVERYTHING myself?!

This decision’s main benefit was that my work-flow was fast, I’d quickly switch from drawing to coding in the same session and have something that looked good AND worked to get excited about. My vision was (almost) without exception fully realized in the end product.

The main drawback was that I didn’t have the time to really expand on game-play and features to the extent I could have if I was focusing solely on coding.

My day job as art director for a kids’ website had me doing cute little gumdrop-bodied characters and I thought “Hey, if I do similar simple small sprites, I just might be able to pull this whole thing off on my own!”

In the end, this simplified art style became the theme of the game: “Chibi.”

Not only was the hero simple, but now I could make the enemies simple, and the rest of the art simple too… JACKPOT!

I trashed the idea of different looking leveled up sprites as being “too labor intensive” and substituted that with color-coded armor and unique swords. This cut way down on the number of animations.

I decided not to do tiles on the main map because, while it would be faster and easier to build a map with a tile engine, actually making the tile engine was going to take longer than just drawing the small map I had planned. The hand-drawn look for the overhead map ended up adding a lot of charm that tiles wouldn’t have done quite as well, but it did have some drawbacks that you’ll see later.


I also decided to cut the number of chaining brawler attacks to a simple series of 3 sword swipes so it would still give the player a sense of brawling, but not be super complicated to “pick up and play.”


I knew that Zelda’s spell system was too cumbersome, so I simplified it with simple icons that appear around the hero instead of bringing up a giant menu. I also made all the spells take the same amount of magic (easier for players to wrap their minds around than “manna” points). So when you level up your magic you work toward being able to cast multiple spells on one magic charge.


I also cut out having mounts in the game, which I plan to put back in for a sequel 🙂

I started out with a very simple outline that involved 2 boss battles leading up to a final boss battle with one side-quest along the way.

If there’s one thing I had learned from previous projects, it’s that keeping your scope small means your project gets DONE. It also helps ensure that it’s not full of ridiculous bugs from your experimental creativity.

The urge to make everything BIGGER AND BADDER AND MORE IMMERSIVE, is a temptation I have to constantly fight when making games.

V. Why a Knight?

To be honest, I saw a sketch of a knight I liked and thought it would fit the game perfectly. I had also been talking with Armor Games about sponsorship and they are suckers for medieval-themed games.

VI. Big Bosses

I just don’t think any game is complete unless there are giant intimidating bosses. Bosses that outweigh the hero by at least 20 times and seem impossible to beat when you first see them. Abandon all hope little hero! (think Little Mac in “Mike Tyson’s Punchout”)

That initial rush of “Big Boss” intimidation pays off in spades when you finally come out on top and, not to be too flowery, it reinforces your faith in your own ability to overcome “no-win” odds in a world that assaults you with them repeatedly.


Bosses should always be extremely over-powered, but so stupid that they follow a recognizable pattern. Humans have some innate ability to learn patterns, so this method naturally leads players to continuously improve and keep coming back at a boss for more even though they keep losing. In the end, you’re able to really impress your un-initiated friends with your prowess.

Chibi follows this pretty well, though I’d say the Canyon Boss’ pattern is a little too complex for a “first boss” and the Island Dragon is probably too simple for a “second boss.”

VII. Expanding Again

So I had a very-near complete build of the game done and sent it over to Armor games. They loved it but wanted MOAR, so they offered me a significant sponsorship boost if I would double the game-play time… Oh NOES!

Greed wins!

This challenge led to the creation of the dungeon Level and the 3 Knights. Those 3 knights turned out to be my favorite part of the game and also the part I’m most proud of. I think they not only extended the game-play, but really helped deepen the “story” and “lore” of the game and are just fun to play.


This challenge also led to the addition of the blacksmith, the tree chopping side-quest, and the addition of the “fire” spell and the “life” spell. These also make the game feel much more adventurous, complete, and fun so I’m glad they made it in. I can’t imagine the game now without those elements.

VIII. Polish

My game was done but without music. All I could think about was that I had made every single aspect of this thing, but have no talent for music. Luckily, I had worked with Brian Holmes on a couple other projects and he was instantly excited to do a fully custom score for this little Flash game. The music took the game to a new level and made it seem way more epic than it probably deserved.

Since I was doing art and code all along, most of my vision for effects and transitions had been built as I went along, so this phase was fairly straight-forward, except for one thing…

IX. The Voice

If I had to attribute the game’s success to one element, I’d have to hand it to my 5 year old daughter who did the voice of Chibi knight.

Only a few days before releasing the game, I was remembering that one of the big contributions to the success of “Castle Crashing the Beard” was the voice-over work done by Tom Fulp himself. A boss that talks to you and taunts you while you beat on him sent the “fun” level of the game through the roof.

I also remembered games like “Smash TV” where the announcer kept you entertained while you were playing and the “X-men arcade” where the voice-work was one of the most memorable parts for me (”Welcome to die!”).

I had just gotten a new microphone and tried recording some “Hi-ya!” and “Oof!” sounds for the knight, and it was sounding extremely lame. My daughter started imitating me in the other room, so I thought I’d give her a tryout on the mic to shut her up.


Once I cut the sounds down and put them in, I knew I had struck on something special. The cute little voice expanded on the “chibi” theme and made the game not only fun to play, but entertaining on new levels. It now had elements of “cute” and “humor” that I knew would keep people playing for long enough to get them over the hump of “casual gaming” and bring them to the addicting aspects of leveling up, finding spells, side quests, and the story of the cute little kingdom in peril.

X. Bugs & Release

Because I had been chinking away at the game for over 2 years, I had squashed most of the show-stopping bugs but, as always, having millions of people play your game inevitably raises other problems.

Some people had problems with how hard it was to grind and “max out” your levels because I removed enemies after you’d defeated them. Not being a big RPG guy, I never realized that this is a common “replay” thing players do when they beat the story arc of an RPG… who knew?!

There was one bug with not being able to walk up to the final boss after defeating the 3 knights, and that took me a long time to squash.

There were also some problems with performance on older versions of Flash player and low-end computers when entering the dungeon because I was doing so much bitmap caching on such a huge map…that one I never fixed because it would have required a complete redesign of my map engine which, if you’ll remember, was designed for a much smaller hand-drawn map initially.

There were a few exploit bugs like purposely jumping off the bridge after beating the Canyon Beast to get back to the main map faster. Holding up down and right on certain keyboards also made you move super fast for some reason.

Also, apparently if you leave the game paused for 24 hours, when you un-pause you’re outside the map… never fixed that one because it makes absolutely no sense to me.

Oh, which reminds me, I never implemented a “lives” system just because it seemed like it encouraged more experimentation to not have to worry about dying. It also took some pressure and frustration out of the game and kept people from “rage quitting” which seems to be the standard for “casual” games nowadays (whatever those are).

XI. Advice

I’d say try to build from concepts you love. Cut out what annoys you about your favorite games, mix and match elements from other stuff you like, experiment a little, keep your scope in check, send your game around to lots of people to play before you release it… and take their feedback to heart, add in the stuff you always wished was in your favorites originally, update and expand on the classics, and shamelessly use your family members’ talents to make your games better…after all they’ll only be 5 for one year!