So! I discovered while unnecessarily mucking around with my menu system in an effort to further procrastinate on building my battle system that colour filters were broken in Verge3. I decided that as my update, I'd fix whatever the problem was and post it up for the whole world to see.

So I got to work. After two weeks here's what I've discovered:

• The internal colour filter functions LOOKED like they were still working all right.
• ColorFilter still worked when called from VergeC.
• I updated to the latest commit in the SVN repository and HookRetrace, etc. stuff was broken because Overkill was working on callback stuff.
• ACTUALLY updated to the ACTUAL latest commit and HookRetrace was just fine.
• If I rewrote code to directly call stuff inside of the v3 table instead of going through vx, then it seemed okay.
• I was using vx's ColorFilter wrappers entirely boneheadedly wrong.

So yeah it wasn't actually broken! Great, huh?

Anyway so I'm back to working on the battle system now. I'm having problems with blitting entity frames but I'm sure I'll just discover in two weeks that I'm doing it wrong.

It's been nearly a month; much too long. In that time I have been thinking about the battle system, rolling things around in my head trying to figure out how to lay things out. I've been trying very hard to figure out where the balance between interesting to play and easy to use falls. I'm somewhat stuck but am slowly working through it (and trying to apply my mantra of "do what's easy and make it more complicated later if you have more time"). I know that what I really need to do is start trying things out and figure out what works and what doesn't that way.

In terms of actual work I've begun to update the Character objects (basically bundles of stats and other information) to fit with the current state of the engine in preparation for using those Characters in battles. I've found a JSON decoder and slipped it into my general game-making library so that I can use it for data files (and, if I end up needing to, net communications). None of this stuff makes for terribly interesting new playing so I haven't bothered with a new zip for this week.

I've also been very busy with work.

In any case, I have also produced a little bit more artwork and I'm afraid that will have to stand for this update's pretty pictures.

What we have here are relatively "final" single frame sprites for each of the six playable characters in the game. In my original plans there were going to be twelve characters in total, and the story involved splitting them up at various times into groups of five or less. When I revamped everything to make it possible to finish, I cut out all but the first six characters that you meet, and you will never have more than five as an option for your party.

This simplified things a great deal but I admit that I was rather saddened to lose the rest of the characters and, sadly, some of the depth of the game along with them. That said, I am still convinced that it was for the best -- a game that I can actually finish is infinitely superior to one that will never make it off of paper.

Much of my free time this week has been taken up by getting together a new Verge engine release. So now:

This has actually been a few months in the coming, and Geas has had updated verge executables along the way; if you noticed the size of the .exe changing every few weeks it was because I was trying to keep up to date with the version in the SVN and doing testing with my code to see if things had broken (which was often!)

Anyway, hopefully now that this is out I can get back to worrying about Geas, and we'll see more game-stuff next week. Have a good one, everybody!

I've got the first section of the LuaVerge tutorial up. It includes one goody in the form of a nice clean workspace zip for building Verge games. The next section is in the works but not near finished yet since it requires me to have time to make some resources for fresh, bright-eyed newbies to use for MAEK GAEM.

By the way, I need to thank Overkill for listening to me ramble and whine. I was really unsure as to the direction that I should be taking with my tutorials, and he straightened me out like the stand-up gent that he is.

Unfortunately, I haven't had time to do any on-computer work for Geas this week. Things have been hectic.

Yet again, only small cosmetic changes.

Download this week's alpha of Geas!

New sprite added for Thea Wayman, Kiel's mother. Also, various under-the-hood changes.

I did some code reorganization this week. Nothing major, but some things flow a bit better.

Also on the coding front, I've added StringTable support to Geas. What this means is that (for now, only in the speechboxes, but I plan to expand this out to all gui display functions) I can define a language file that dictates a bunch of string replacements. Whenever I go to display a block of text, it first checks its language redefines and swaps in the other language's version instead. At some point I will hack Verge up so that it supports larger character sets, and then my dream of being able to translate Geas to other languages will be doable.

Using the StringTable there's two ways to going about producing a translatable game. First, you can just write the game in your language of choice, then pull all the strings (using some magic that I haven't yet created or decided on) and just directly translate from that. Secondly, instead you can write all your text to begin with as just IDs of some sort. "kiel_c0001_ln001" or something. Then, have a language file for every language, including your first one.

I haven't decided which of these approaches I will actually use when I get down to the nitty-gritty of having real conversations and not just placeholder text everywhere. I need to decide it pretty soon though, as I'm getting sick of having a completely pointless and storyless game. That said, I'm also thinking that I probably want to write a conversation system (like the one in Neverwinter Nights, for example) so that I can handle all of this a little more sanely rather than just writing all of the game's dialogue across hundreds of lua files.

Anyway, aside from Geas, I also started work on my LuaVerge tutorial. Sadly, in writing the first section I have discovered that I'm not sure exactly the direction that I want to take, and that may end up getting delayed. In essence, I am struggling with writing a tutorial for using LuaVerge "right now" -- in other words, in a state of mild disrepair and lack of good tools. I'm not sure I want to do that!

What I am considering instead is to write the tutorial in tandem with generalizing stuff I've built for Geas into a game-makery system, KISSKISS (mentioned in the LuaVerge doc index) that makes a lot of decisions for the user and does a lot of the hard work. Unfortunately this means that I have to do all that hard work and as anyone who has been reading this site knows I'm kind of bad at getting complex projects finished. Ultimately, though, I think KISSKISS + a tutorial on it use will be more helpful to the absolute beginner than a "here's how to use Verge ONLY LUA" that has been sort of done before anyway, and it will help differentiate Verge as a useful tool as opposed to another library that handles some of the graphics stuff maybe.

So yeah, I'll keep thinking on it.