VagrantEnchanter

I just realized, I feel like a jerk for never getting back to you. I got it to work after a couple days of fixing some silly eventing. I just wanted to pop in and say thank you. So.... Thank you.

Seems like things got interesting while I dropped off of the face of the earth! Darth, I wish you the best. I think everyone here would be more than happy to see it done, regardless of how long it takes. Besides that, maybe a collaboration wouldn't be such a bad idea? I mean, a file open to three coders would be the fastest way, wouldn't it? Giving each other notes, correcting redundant or buggy coding and even coding with whatever time someone can spare. Even a moment can make a difference. Anyway guys, just popping in. Keep being awesome.

Alright. Runic Enchantment was close, but not quite there. It took me an hour to get the notetags to work correctly with the limiters and even when I did get them to work, it didn't do quite what I had in mind. I have to say though. Ridiculously close. Just that tiny bit, you know? I'm sorry if I'm seeming picky here.

Mm, not the exact method but something close to it. I like the concept of the drops becoming PART of the weapon. I just don't like that you can't pay someone to REMOVE them. That's the part that irritates me. I'll try Runic Enchantment and get back to you. (By the way, this request is getting cut into pieces...Should I ammend the original post?)

Just out of curiosity, are you still working on this?

That kinda looks like a script call....In which case, the enemy note is the wrong place. You COULD have it set up so that the enemy drops an item instead. The item could call a common event which simply calls the script. It's a bit tedious if you want enemies to drop singular values, though. Anyways, there's my two cents. Edit: Actually, I take that back. You could make an item for each denomination of enemy. Small Gold Piece: 10 (Whatevers): Small enemies Gold Piece: 20/50 (Whatevers): Medium enemies Large Gold Piece: 100 (Whatevers): Large enemies Obviously you aren't calling your money "Whatevers", those were just examples. In any case, that's probably the easiest way to do it. Just set the drop rate to 1/1. Come to think of it, it kind of makes sense that way. After all, when you find a dollar, it doesn't just poof into your wallet, right? You have to put it in there. Anyways, my two cents. ....again.

Yeah, I honestly like Envelon's the best as well. I tried eventing the upgrade system and I realized.....I have no way of doing it without making an item for each variation of the item. That's hundreds of thousands of items I would have to create. XD

Edit: No, I was wrong. That was an entirely different shop. I made a fresh demo with the error and my edits to the script if you'd like to see it for reference.

Ah! I hadn't even considered changing the items! Well, that would be necessary, wouldn't it? I'll check those out after class tonight!

You can play cards, capture cards, and win/lose. If you can't even play a match, then it doesn't work. That depends on how you define the utility function. If your utility is based on maximizing your score, then you don't need to consider self-defense anymore: the AI would choose the move that would maximize its score, which means it will try to avoid moves that would leave itself open for capture. If you consider an arrangement of cards to be a particular game state, then you would try out different combinations of card arrangements and calculate the "total point value" of that state. Then, yes, you would then pick the state that has the highest value, because presumably that is the move that will give you the highest chance of winning. Of course, how you determine this point value is up to the developer. For example, if you have a card that looks like 1 1 A A Then any squares that leaves the card open to attack will be given low values, while any squares that protects it will be given higher values. Place the card in top-left corner at the beginning of the game, for example, would be a high value state. That's pretty much what I had in mind. If I could script, I would be all over that.

An AI basically needs to follow the rules of the game. If you have a working game it is not difficult to do the AI. Whether the AI is "good" or not is a different issue, but you can practically put together an AI in an hour. Consider how RM's battle AI works: just randomly choose an action and randomly choose a target. Kind of silly but people don't seem to have an issue with it. If someone gives me a working triple triad Ace implementation I can look at the AI. Hmm....How much would you need to consider it "working"? Edit: Thinking about what you said, I had a thought....What about scripted battles? The user chooses what moves are used on certain turns.....Would it be possible to change the conditions? Like you said earlier with the "Able to take card?" if/else statement. I mean, the only real issue would be "self-defense". Making sure the opponent doesnt place stronger sides against borders or leave weaker sides exposed. Edit2: Actually, that might not be too hard.... Maybe. If you could give the card a total point value, then recalculate that value after each move and see which leaves the highest value, would that be reliable? Or maybe a literal vulnerability factor? As in, if <5, hide, else show. You could just apply that to each side of the card to find the best move. I don't know, something simple like that.

Just curious.....How hard would this be to throw together? I mean, we already have quite a few character portraits in-game. Most users wouldn't mind setting it up or even making their own if it meant using a script like this.... Any ideas how long the A.I. would take?

If you don't like the "hanging around" portion of it, you could do the same thing with choices and conditional branches. For example, say you have around seven members you can hire. Set up your vender npc to show the buy or leave options. Select buy? Show the first three, have the fourth option be a "more" option leading to the next three, followed by anther "more" option ended with the last character, a help option (If you want to add any notes on characters), and the "back" option. When someone selects a character, this is where the conditional branches come into play. Have it set so that if "Henry" is in the party, it says "Hey! I'm already here!" or something to that affect. If he isn't, subtract the gold and add him to the party. It's not too hard, I played around with it myself not too long ago. It just takes some testing to work out all the kinks if you haven't done it before.

Haha yeah, sorry. I was tired when I wrote that so I was a little lazy with details. I started a new project and there was no crash at all. It might have something to do with the 20+ scripts running at one time. Hmm. I'll look into it a bit more and tell you where the crash happens. In regards to the item itself, it was just a simple hp/mp booster. Edit: That's the strangest thing. I moved the script up a few spaces in my script manager until it was RIGHT below the class script and the error vanished. Hmm. Weird. Oh well, I suppose that one solved itself. Sorry for the hassle.

I ran into a tiny bug while testing the other night. I'm not sure if it's just me or others as well but whenever I use an item to boost a stat, it crashes. line 213: NoMethodError occured. undefined method `+' for nil:NilClass Is that something I messed up on my end? I don't remember touching line 213...