26 pages of drawings. I would never have imagined it would take this long, and I certainly could’ve done more. Hopefully my combinations will make a lot of fun varied items to find. It was pretty hard creatively to come up with the 8th belt design, etc… but now I just have to get them digitized and imported. That’ll be very rewarding to see!
I’ve been telling myself for a while now that my next game is definitely going to be smaller and simpler than Code of Arms. I’ve been interested lately in the capabilities of products like GameMaker, Construct 2 and GameSalad. Then I had the realization that I’ve just bought Flash and have been using only a small subset of it’s features. The criteria I had for these tools was:
No annual pricing, just let me buy it and freeze my tools in time
Preferably a native Mac client
Extensible with scripting/plugins/whatever-you-want-to-call-them
Tons of people make games in Flash and Actionscript, it runs great on my Mac, and duh, I’ve already paid for it. So with that realization I decided to do a quick single evening Air for iOS game to see what the process was like.
It took a total of three hours to slap together some graphics, set the environment to build for iOS with provisioning profiles/certs, and code up an AS3 tutorial of the classic game Snake.
TL;DR It was easy as dirt, lots of fun, and very flexible.
This post is long overdue, but artwork takes forever to create. This a small sampling of my game generating headwear at runtime. The art components and color palettes are all procedurally generated and assembled into a single item. I had a surprising amount of work to get from my demo code to it working with a wide variety of graphics, but here it is. Now on to drawing weapons!! Hopefully my process is better and it’ll progress faster.
RPG number crunching has been my primary progress lately. Lots of linking this stat to that stat with random rolls, etc… So many stats and calculations, I’m getting kind of numb to it. I would really like to do something graphical because seeing something visual change really makes me feel like I got stuff done. Unfortunately, that is not in line with my current goal of completing an end-to-end playable game. I did get damage over time, spell cooldown and mana regen working (see pic) and lots of cool stats for items.
The quest system has been my focus so far this Thanksgiving weekend. The idea is that for each region (Jungle, Arctic, Volcano, etc…) the player will have to undertake some number of quests to unlock the area boss. The quests will be randomly generated, though probably not to the depth of the item names. Since my world isn’t going to have actual places (at least for now) the quest will consist of objective and completed text, rewards, and the monster sequence which must be completed. I need to wrap this all up in an object which persists if a player loses or abandons the quest for attempting later. This article on “Self contained reusable plots” was very handy, as was this paper on quest analysis in MMOs. I’m really trying to get the code for the full end-to-end prototype done, art can wait until my ideas are more firm. Here’s a few of my rough UI screens thus far:
Much improved main character design… It’d better be because it took forever. I think I’ll dress him with a full gear set and work on some animations. I hope that looks ok when animating, there’s no good way to know of without trying. I have a feeling the artwork for this game might be very time consuming.
Fighting demo video! I’ve got the basics of rotating turns and calculating damage with a victory condition. The code is really rough, but I want to play a prototype soon without investing too much time in ideas which might change. Pretty exciting!
Lots of progress in a hurry this weekend. My code is getting really out of control as I rampage towards seeing two AI fighters duke it out. Haven’t been thinking enough about structure and I hope it isn’t too hard to refactor. I didn’t want to invest too much time in the UI, but needed to have some minimal indication of the item’s properties. The latest build has the ability to save your inventory and to set the active weapon. I’ve added some procedural stats and skill tokens to round out the item design.
Your item has a certain rarity (Common, Uncommon, Rare, Unique) which determines the min-max ranges for item stats and how many bonus attributes and skill tokens are attached. The attributes are pretty standard RPG types (+5 attack, +5% critical hit, etc…) but the skill tokens are different.
Here’s my idea for the “skill tokens” as I’m calling them now. Your character derives all of it’s passive/active skills from equipped gear. In the image above (assuming we’re not considering other slots like head, chest, etc…) your character would have Level 1 Heal and Level 1 Lightning bolt available for use in combat. I was thinking each skill would have 3? levels available and by equipping your character’s slots with items with similar tokens you could upgrade your skill to higher levels. An example would be if 1 heal token was present on your equipped helmet, boots, weapon and chest. You would have 4 heal tokens which I may determine is the right number to cast a level 2 heal. I might set the limits to something like 1 token=Level 1, 4 tokens=Level 2, 8 tokens=Level 3.
I’m not sure I described that well but I think it will allow people to customize their available skills and another reason to hunt for particular gear/attributes to maximize their character’s role (Wizard, Warrior, etc…)
I have been brainstorming how to implement my AI for a while now. Finally got everything working and I have a very simple countdown style fight working with a victory condition. It’s about the most boring screenshot ever, but illustrates what I mean. Now that the framework’s in place I’m pretty psyched to add the states which govern what the fighters actually do. I’ve wanted a reason to do an AI opponent for a while, and so far I’ve just made puzzle games with a right or wrong answer. Programming Game AI by Example was an invaluable book.