Gamedev Lounge 0010

[puggsoy]

Hmm, I see. I feel that you should call this Stat, and rename your enums to something else. Personally I tend to do something like "StatEnum" or "StatType", since that essentially is what the enum is; an identifier for the type of stat, not the stat itself.

In the case that this isn't an option, maybe something like "StatHolder" or "StatVal" would work.

[Phaze]

Seems good, I'll probably use one of your suggestions (StatValue sounds nice and so simple in retrospect). Thanks

[Phaze]

As I start to flesh out this rewrite and populate it with the Classes and setup I had in the Java version, I'm finding myself constantly questioning previous (seemingly sound) assertions and now I'm completely indecisive about what to do (one example is if the lifespan of objects and event management). Makes me feel like crap

I hope I can deal with these issues by looking around to see how other people managed this kind of thing.

[TheShyGuy]

A game I've been working on with mit. Very early WIP:

gifs

Spoiler

[Phaze]

Here's hoping I could do something half that impressive in the future.

[TheShyGuy]

Working on pathfinding for partner following and enemy chasing:

[MrYoshbert]

Really impressive stuff, ShyGuy.

[TheShyGuy]

Thanks guys,

Heres an update with walljumping. I gotta polish the node traversal after this.



if anyone is interested in pathfinding: http://www.policyalmanac.org/games/aStarTutorial.htm

[MrYoshbert]

Which program are you using to make that?

[TheShyGuy]

The game is being programmed in C# using the Xna Framework from microsoft.

[TheShyGuy]

I'd just like to post this link which links to two AI videos. They're about utility theory as a way of improving decision making and can also give the AI a bit more personality.

http://intrinsicalgorithm.com/IAonAI/201...gdc-vault/

I'll probably try it out soon to make the AI for the game I'm working on a bit more interesting.

For those who checked them out, what do you think?

----

(I know that I could've edited my previous post, but I figured that it's been a few days and this would be an OK way for others to know that the thread had been updated)

[IndyGo]

If anyone remembers my horizontal Parodius-clone project "Comedius", currently I've had difficult time catching up and using E02 efficiently to develop it further on. Relying on Stealth to teach about E02 isn't helping too much either as it's not entirely instant for me to get new info.

So I thought lately about moving back to Game Maker (Studio, perhaps) and start all over again. Not that it's too entirely difficult to catch up with what I had in E02.

[Phaze]

If anyone remembers my horizontal Parodius-clone project "Comedius", currently I've had difficult time catching up and using E02 efficiently to develop it further on. Relying on Stealth to teach about E02 isn't helping too much either as it's not entirely instant for me to get new info.

So I thought lately about moving back to Game Maker (Studio, perhaps) and start all over again. Not that it's too entirely difficult to catch up with what I had in E02.

As long as you didn't have everything but polish added, you should be able to reproduce/port what you've already done without losing too much time. Heck, it might even make you think "gee, this code could be written better...", leading to improvement of the game in general; I know I've sure thought that when I decided to rewrite/port things in the past.

Edit: As for E02 vs GM, I really can't tell you which would be better to use. I have p. much 0 experience with tools like that.

[Hoeloe]

I'd just like to post this link which links to two AI videos. They're about utility theory as a way of improving decision making and can also give the AI a bit more personality.

I just have one question about your AI - did you place those waypoints manually, or are they automatically constructed by the algorithm?

Also, you may want to watch out for XNA - Microsoft have admitted it was a failure, and are dropping all support for it. There are some open source continuations of it, though, if you want to release for PC.

[TheShyGuy]

I manually placed the waypoints by hand with a small tool I made in a few hours. Also, utility AI has nothing to do with generating a graph. I haven't 100% tried out the theory though but it sounds interesting heh =/.

[feel free to skip this paragraph]
I did a simple test with it a few months ago. The test game was a text-console-based battle with the AI. The only stat was Health and both me and the AI can either heal or attack. The AI considered it's own health and my health if I remember correctly. If I had high health and he was low, he'd heal. If I was low on health and he was high, I'd end up healing (or occasionally attacking) alot. He'd be pretty aggressive and continue to attack. If we were both very low on health, despite the AI's health being low (or lower), it would continue to aggressively attack. If I panicked and tried to heal up towards 100%, the AI would continue to attack until I gained "too much" health that attacking was no longer worth it - and it would start to heal also. That's what I was going for and I was pretty happy that it ended up like that. I'd say the hardest part was coding up a way to make the equations data. Today, I just use lua to create the equations instead of trying to parse them in c# or through a string. *I didn't implement utility AI in the game yet*

Thanks for the heads up. I know about Monogame, and have tried it multiple times, but there's always some issue that gets in the way =/. I don't need to move away from XNA, so I decided not to yet. The only con I see from no more future XNA support is that there won't be any cool updates, but I don't really need any update as far as I'm aware. If there is a high enough demand for support for another system, then I'll try to move to Monogame again or something else.