Tag Archives: analysis

Analysis: Great Dungeon in the Sky

Great Dungeon in the Sky is available for play on Kongregate.

Brief Game Overview 

The main menu screen

Great Dungeon in the Sky is a platform game in which your goal is to go from level to level, looking for four dragons which you must kill to unlock entering the final boss’s level. Additionally, every new enemy you kill unlocks playing them as your character.

First Impressions

Great Dungeon in the Sky is easy to pick up quickly. Within a minute or two (and checking the short instructions on Kongregate), I was able to figure out how the game worked and what my goals were. The controls are simple: arrow keys to move, and Z/X/C to activate your abilities. Despite this simplicity, the controls are still a bit unintuitive; which abilities are assigned to which key feels rather arbitrary.


Great Dungeon in the Sky ran completely bug-free for me, and the gameplay was quite easy to get used to; however, there were a number of glaring flaws. The enemy AI for every enemy, including bosses, was effectively the following: “Walk back and forth along the platform. If the player moves nearby, or if the player attacks, perpetually run toward the player and attack repeatedly until either the player dies, or this unit dies.” Some enemies with ranged attacks would stand back a few tiles and shoot; any other enemies just keep charging toward the player, endlessly swinging. There was no real variety beyond that.

Because of the AI clinging to you so tightly, and because almost every unit moves at the same speed, there’s no real chance for developing a play style or improving your skills. In almost every situation, the best strategy is to stand still and use the same attack ad infinitum. I was even able to defeat every boss on my first attempt this way without getting hit by all but one. To be fair, there is one option I found that requires some skill: projectiles shoot at an angle depending on your character’s vertical velocity, so you can experiment with bank shots; however, this quickly gets repetitive and it is simply much faster to walk up to the enemy and bash it to death.

Although the game plays more or less exactly the same an hour in as it does in the first five minutes, there is some variety present.  For example, although most characters are unlocked by killing the equivalent enemy, killing a dragon only unlocks its egg. You hatch the egg, giving you a young drake which then evolves into stronger and stronger forms as you kill more enemies. As another example, in most levels a pit simply leads to another section; however, in one level, this brings you back to the beginning (though this could easily be considered as inconsistency rather than variety).

Finally, there is an “HD Graphics” option which implements some fun visual effects like motion blur. Unfortunately, this mode looks quite unappealing as everything is constantly blurred by your movement and enemy colors are washed out by the reddish tint. The artwork was clearly designed without HD Graphics in mind.


Jumping into the game, I was awestruck by the quantity of characters. To be more precise, there are probably around 300 of them! (Getting an exact count is tough, because some characters take up more than one spot on the character select grid.) You start with 32 characters unlocked: 16 unique characters with male and female equivalents.

Screenshot of Great Dungeon in the Sky's character selection screen

Lots of choices!

Unfortunately, most of these characters are very similar. All of them use different combinations of the same small handful of abilities, and to make matters worse, at least one (Living Statue) was completely unplayable as it could not move. Furthermore, there is no real sense of balance among all of the characters. The starting characters are somewhat in line with each other (with a few outliers), but everything quickly gets out of whack with all of the unlocked options.

Adding to the imbalance is the level design: every time you enter a level, it is picked randomly from a set of pre-made layouts. One level has an almost-unavoidable section with a large number of units in it that will brutally destroy you if actually try to fight rather than keep running, another is a straight shot forward from beginning to end. Thankfully, levels vary greatly in layout, but the lack of quantity causes one to see the same levels over and over. Flying characters can fly their way through every single level, ignoring everything; however, this is a bit of a boon once  you’ve seen the same levels five or more times each.


Conceptually, Great Dungeon in the Sky implements a number of interesting ideas! In terms of execution, however, the game falls short in a number of areas. There are some interesting mechanics present, but nothing new really presents itself past the first five to ten minutes of gameplay. Others have enjoyed the game, but I found it lacking in challenge, balance, and variety.


Current Goals

With fall quarter classes coming to an end, I’ve had time to unwind and start thinking about my current short-term goals:

  • Learn Lua – I have heard nothing but good things about Lua, and it seems to come up frequently with games. When one also considers that the first edition of Programming in Lua is legally available for free online, teaching myself the language is definitely something I plan to do in the near future!
  • Get a taste of DirectX – Having worked in XNA, and after getting a brief introduction to OpenGL, introducing myself to DirectX seems like the next logical step. Although I don’t plan on attempting to master it in the immediate future, I would like to learn the basics and reach the point where it does not intimidate me, at the least 🙂
  • Stay up to date on Games Industry news – I’ve been making a point to read Gamasutra and The Escapist daily to stay on top what’s going on in the Games Industry
  • Play (and analyze) a new game at least every other day – In an effort to both improve my knowledge of what games are out there and to improve my ability to analyze games, I’m going to push myself to play and analyze a new game at least every other day. If I can spare even 15-20 minutes every day or two to trying out a new game, I should definitely be able to pull this one off. I also plan on posting my analysis of the games on this blog!