Pages: 1
This is a tip for RPG development. Assuming that you have a certain array of charset resources, and you want each charset to represent a different NPC (i.e. you aren't generically reusing the same charset to depict 100 different characters)...

Create a map and fill it with each charset at your disposal. Then, whenever you use one of them in your game, remove them from this availability map. To navigate the map easier, you can organize the NPC's based on whether they appear to be good warriors, evil warriors, royalty, regular civilians, major NPC's, etc. Charsets that have extra frames for expressions like laughing and fainting should be prioritized for use as main characters or as major NPC's that have bigger roles.

Personally, I've found this practice to be helpful. But I know not everyone approaches game design the same way.
I'd really like to get rid of LockeZ. His play style is way too unpredictable. He's always like this too. If he ran a country, he'd just kill and imprison people at random until crime stopped.
The problems you have are problems I don't think I can even figure out how you have. Why would you even have a charset picked out for your game if you don't already know what character it's for?
Stand back. Artist at work. I paint with enthusiasm if not with talent.
He's talking about the specific kind of rpgmaker games where the devs plan to have unique npc's.

It's standard practice to reutilize NPC charsets, some palette swapped, in order to save disk space since they're supposed to be unimportant to plot details and only there to create an illusion of a world where people live, right? Then there's those games that put conscious effort in not having the same houselady sprite even in a second NPC, attempting to make each unique even if only graphically.

I personally think the practice danarkos talks about sounds pretty cumbersome. I dont think it's necessary to create a specific map where you fill it up with blank events wearing the skins you'll have available for your npc's and then erasing them as you go using them in the actual game, it sounds to me like it would be far easier to just use an excel document for this purpose if we're talking about an "did i already use this npc charset?" checklist thing.

Then again, in games I've played where each npc is graphically unique i've never seen the devs using that many character sprite options or that many npc instances as to create the "problem" that this "solution" brings to the table. Just as how in TV shows having way too many characters means you can't possibly focus on them enough to develop their personalities correctly, having too many unique npc's makes it very difficult to give each of them a worthy self identity. In cases like this it's generally better to keep the total NPC count on the lower side to help the player want to get to know them without overwhelming them.

If your game is so large that it requires 100 unique NPC's, I'd personally just put focus on NPC's that provide actual visual aid to the world around them, while keeping a "neutral NPC" sprite which i'd reuse over and over whenever an "utility NPC" was required (the type that only exist as vending machines, or as ticket booth officers, innkeepers and the like. if the plot doesn't require these characters to be important, I don't sweat giving them their own identity.).
Personally, I've found this practice to be helpful. But I know not everyone approaches game design the same way.

To be even further pendantic, this isn't game design. More like general gamedev organization.

But yeah copy pasting caps from the snipping tool into Google Shhets or something would be better. On the other hand you should just be tracking what's currently in the game rather than what isn't.

I don't think many people really value their NPCs being unique snowflakes, players just generally accept this unless there's some crazy reason for it, like you're making a Chrono Cross esque thing where every NPC in the game is a recruitable party member or something. At that point you'd be better off looking into procgen or something.
Usually if I have a large cast in a very story-heavy game, I'll do this because it makes it easier to group them up and know who is who and where they fit.

Or if they have special message boxes/conditions when they talk or act or something (like a clown that has a juggling animation or a guy that is smoking).
I sorta think people don't really care nowadays if you re-use assets for something like NPCs. But it also really depends on the scope of your game and the structure that you're going for. A world-spanning epic can more likely get away with re-using assets for generic NPCs, especially if you're planning on having hundreds of them. But if you're making something with a hub world, it's much more important that every character looks unique if the player is going to be seeing them over and over again.

The key here is the question: "Does this character NEED to be memorable?" And if it's just some generic townsperson telling me where the bar is, I don't really care if he looks like Farmer Brown two towns over. Otherwise what you're describing just seems like a lot more work with very little payoff.
Guardian of the Description Thread
For what it is worth, I had two separate note-files for character locations for this game. One was what location a character could be found before they were recruited by the player, and the other was their location in headquarters after recruitment.

Of course, having 108 characters, and having them distinct from each other, and from less important NPCs, was important to me. However, the game suffered by being short, and players probably weren't becoming attached to any one of them. Of course, Suikoden games tend to have a number of forgettable characters to begin with, but, I will forgo that discussion for now.
Pages: 1