#rgss3

I spent ages reorganising my events to fit accordingly (I acutally had a lot of fun doing that ^^;), but realised last night that the script somehow makes it impossible to save in-game. I’m proficient enough with RGSS(3) to sort most script issues, but this movement script is just way beyond my level.

This is me, crying a river.

Ever wanted to change the font of your game, but you don't have a clue about where it is or you just don' want to bother searching!?

Because you're a lazy bum like me, I wrote it up real quick for you. Am I nice or what. It's highly impractical, but makes for a nice little script for people that dont bother installing your fonts. Just write a message, switch the font to a default one and ask if they saw something or whatever, then switch it back! The possibilities are endl- highly limited! ^-^

Instructions in the script, feel free to ask any questions.

German English

To keep it simple, I was looking for a script that allowed me to store saves in a separate folder, like the appata folder, but the only scripts I found had restrictions on them, as in, non-commercial. So I decided to make one that has absolutely no restrictions.

You can find the script here.

So as some of you know I've been working on Free Movement for quite some time, although not in the last few months. This is mostly due to it being kind of shit. It's the same I had with Turbo Model Thingy for Minecraft, I thought it was shit, so I decided to just start over and make Nitro Model Thingy.

The difference is, Free Movement is a whole different kind of beast. If I am going to start over, I need to plan this. First, I'd need to re-evaluate how to do everything. Right now it's a mess that's built upon a mess that I've created myself. I'm now at a point where I'm like, fuck this shit, this piece of shit is too slow. I can't add new shit without it slowing down everything.

The main problem is that events and followers need to be accounted for as well. Now events can be rather easy to fix, but followers, they're pretty much bullshit to do right at the moment. This is why I also need to invest time in finally learning proper AI, especially the A* pathfinding algorithm.

But doing something from scratch should also be justified. Why should I do it from scratch? There are various things I want to do right. First of all, collision maps. Right now I'm using Tsukihime's Collision Map DLL to handle collision maps. This has the disadvantage that collision maps can only be of a certain size, plus they're only made for character movement. I'd need to take into account aircraft movement as well.

Another thing I want to do is split the script up in smaller subscripts that work modular. I'm not sure if it might be a viable thing to do, and it's almost certain it won't affect how things are done now, but still.

There's one thing I need to look into, and it's the main reason I'm thinking about making this change. Height maps. Height maps aren't really all that necessary, but they do add something that might result in more dynamic gameplay. For example, you could implement actual jumping mechanics, or have characters actually climbing stairs. With some work, you might even be able to define multiple layers of heights, allowing for bridges.

After over a year I decided that enough is enough. I can't seem to optimize this piece of code, and believe me, I tried. I even tried to write a DLL for it.

The main problem is that I've written it in such a way that the files it generates should be program independent. This means I can't use Marshal.dump to store the data. Well, I can, but that kind of defeats the purpose of this endeavour. Another problem is a bit more complicated to explain without explaining what CharasVXA is.

Short history lesson. Back in 2003 RPG Maker 2000 and RPG Maker 2003 were big, mostly due to piracy, but still. Now there was this community called the Charas Project. Its main draw was the online character generator. Then some guy came along and made Charas.EX, an offline version, which is where I'm heading to the point.

Charas is basically the character generator found in RPG Maker VX Ace, but with a bit more customization options, and an easy way to navigate through the several resources. Charas.EX built on it and added some new interesting stuff, like being able to recolor your stuff, kind of what you can find in Game Character Hub.

I myself have worked on Charas.EX a lot, added a lot of hacks to it, trying to improve it and stuff, but it never really worked out well in the end, and everybody lost interest. Come 2012 when I first got RPG Maker VX Ace. I loved it, and I still do, heck, every now and then I fiddle with it a bit. I never make games with it, but who cares, I'm still having fun with it.

One thing I found though was that the character generator was pretty limited, that, and you couldn't use the stuff in-game dynamically. Well, you can, but I wanted to go the hard way. Instead of just simply splitting the resources and put them in the game, coding a simple character assembly script, I wanted to replicate what Charas.EX did and improve on it.

CharasVXA is basically a character set assembly API. It doesn't actually generate the characters for you or provide with an easy way of generating the characters, but the core is there. Now, here's the problem.

Creating the assets is a pretty tedious job. You have to greyscale the resources, but not only that, you also need to color separate them. And yes, this process can be automated a little, but that's not what the main issue is, that's just something outside of the scope of my script. The main problem is assembling them into resources. You see, to assemble these resources, it needs to read the greyscales and write the data into a table, which not only stores the greyscale value but also the opacity. This process is really slow, especially due to RGSS not being all that fast anyway. If you download the demo I've posted below you can already see this.

The way I solved it is to just thread it all and put a loading screen on top of it, but, as you can see, it's really slow. And loading the data itself isn't fast either. The main reason it's this slow is because it also needs to write the data. Now, loading in the greyscale data is already extensive work, but saving it isn't that fast either. It's a big problem which I tried to solve with DLLs, but so far I decided to just throw my hands up in the air.

So yeah, there you go, here's the third WIP of a script I've been working on for a very long time. Feel free to study it, improve on it and use it. The script was intended to be released as CC0, and it still kind of is even though the script isn't complete. If you do decide to improve on it and release it on any of the sites, you're free to do this. You won't even need to give credits, provided you do actually touch upon the code in a significant way. Not because I need the credits, but it's kind of a dick move to just say, hey, I made this, without altering at least some of the code. Plus, when people find out, and they will find out, they'll ban you into the next generation.

As a bonus, I've also included the source of the DLL I made. Just in case you're interested.

Download

Demo