Guardian Angel Devil is a solo jam game Play it here

The original game idea popped in my head when I saw the theme “Guardian” during voting, and when “You are the Villain” ended up the winner it came up again with a nice twist at the end. The idea is a take on the “chosen one” trope, as in the hero, Blake, is a young farm boy and he is the only one who can retrieve this evil artifact from the dungeon. Since you, a small demon,  are the villain, of course, you pretend to help him since he’s clueless, because you want to take the artifact for yourself after he retrieves it. You have cards you can play to provide various effects on the battlefield and Blake does all the dirty work of fighting.

The Good

JSIL:

I wanted to make a web game, and was about to use silverlight again when I heard about JSIL. It meant I could code in XNA but then export it to Javascript. I spent some time porting my silverlight tile engine to XNA that played nice with JSIL, which wasn’t too hard as long as I found workarounds for the methods that JSIL hadn’t implemented yet (Particularly having to read xml tile maps with an xmlReader because XDocument and XmlDocument weren’t implemented.)

Juicyness:

This came from a video that someone posted before the compo. Even though I didn’t manage to animate combat attacks, a simple particle effect and sound effect worked pretty well to make the game feel more finished, and took far less time than animating each frame.

Gameplay:

There are three different ways to finish, and all are possible. Plus it plays out a bit differently each time depending on which cards you get.

The Bad

Use of time:

By the end of the compo I only had a fairly generic RPG with the villain walking around and the hero equipping items and attacking, so I ended up entering the jam. I think I may have been able to squeak by on compo time  if I had planned a little more efficiently. I plotted my time out fairly in depth to see where I was using it and this is what happened:

Ideas: 2 hours  compo

Art: 4.5 hours  compo (about 2 hours thrown out)

UI Programming: 4 hours compo

Level Design and programming: 7 hours compo/1 hour jam (Some of this was unused)

Programming walking around the map: 2 hours compo/ 1 hour jam due to JSIL issue

Programming combat: 4.5 hours compo/ 1 hour jam

Programming items and equipment: 2 hours compo

Programming/Drawing cards: 4.5 hours jam

Sound: 1.5 hours compo (very good bang for the buck)

Conversations: .5 hour jam

If I could do it again I could have saved some time on art by deciding my style better in the planning stage. I also could have shaved some time off level design by only implementing things in the maps as I went. There was a bunch of tile flags in the final room that went unused because I didn’t have enough time to implement the cutscene I was planning. I also wouldn’t have bothered making JSIL builds until after the compo ended because it was really just a port. I lost a bit of time implementing something and then finding out I’d have to do it a different way to appease JSIL.

Bugs:

There are a pile of gameplay bugs (Not breaking and most of them work in the player’s favor) mostly due to the rush at the end to get 23 different card types working.  I think some of them might only be in the JSIL port but I haven’t had a chance to look into them.

GUI:

This game had quite a bit of GUI work and not very good base code for it (I intend to rectify that in my library before the next LD). And I didn’t have time to code proper window backgrounds so the text is a bit hard to read at times.

If you haven’t done so yet, Play it here

Per the rules, my personal library is located here.

I will be using C# and XNA, but cross-compiling with JSIL to make a javascript build.

I’ve already done islands (LD17), as well as exploration (LD16), so I hope I don’t end up with deja-vu.

My personal library is located here.

Tools I will use:

• C#/XNA
• JSIL for cross compiling C# to javascript
• Photoshop for graphics
• Possibly tiled if I want a prebaked map
• Possibly autotracker if I want to be lazy with music.
• Audacity for sound effect editing.

First off, I have a web build up so if you’re not on windows or just don’t feel like downloading anything, go here

Multiple Personality quest is a roguelike where you switch characters every 10 seconds. There are four characters, with different strengths and weaknesses and powers.

The Bad

Overambitious Idea

While the game was really fun to make it was a bit too much to squeeze into the compo and almost too much to squeeze into the jam. I didn’t get as many skills for the characters as I’d like, as well as pretty much no time to balance combat.

Game Balance

Game balance is way off. It’s ok at level 1, at level 2 you get a bit powerful, at level 3+ you get way too powerful, especially if you get over level 5 since I MEANT that to be the level cap but forgot to actually cap it (Which is why you see your character have an “Unknown Sword” or some such when you hit level 6+ Oops). It’s actually possible to finish the game at level 8 and the boss is fairly trivial by then.

This also takes away some of the differentiation of the characters. Who cares if Geoff is a kickass healer if you never need to heal? Who cares if Chloe can root if you one-shot everthing?, Who cares if Dirk has really good armor if nobody else ever gets hit anyway? Who cares if Gloria can teleport if she can one shot everything twice over. (Well I guess you care if it gets you around the dungeon faster :p)

The Good

Gameplay

While the balance is way out of whack past level 1, I think the gameplay idea is pretty solid and with some tweaks could become a pretty fun game.

JSIL

I’ve used JSIL for my javascript ports for the last 3 LDs and each time I use it it seems better and better. Porting this time around went really smoothly.

The Jam

I like how the Jam can be a good fallback if you have too much idea and not enough time for the compo. At the end of the compo period, what I had wasn’t really a playable representation of what I wanted, but I had the extra day to get it working and that made all the difference.

Saving Graphics for Last

Since I had to strip out so many features to fit my coding into the allotted time, this resulted on more code time and no wasted graphics work, since when I started on graphics I knew exactly what I needed and no more than that. If I had done graphics first I probably would have a bunch of dungeon, armor and potion graphics I didn’t use.

If this game sounds interesting to you. Go here and play it.

Tags: postmortem