For me personally, I usually start at a minimum of 500, then add more as needed. I’m just a little curious as I’ve noticed a pretty large difference between authors when it comes to their variables.
But more than that, I’ve seen huge differences in how those variables are used. With some authors doing some black magic and using barely any, to do a crazy amount of things. Some authors have massive works, and not nearly as many variables as you would expect, while others have small works that have a variable for anything.
500 seems massive. I don’t plan a number of variables, just add them when I need them, so I’m not sure what my total will be, but it’s around 300 right now, I think. (Assuming temp variables count)
Really? That’s interesting to me. I’ll say temps count, and in that case, I have a little less than 1,000. 900 or so created ones, and 100 or so temps across two scenes.
This is so fundamentally at odds with how I develop that I have no idea how to even answer it. I make variables as I write, as things occur to me to track. The idea of sitting down at the start and going “I’m going to plot out five hundred variables” is very strange, that’s just not how I conceive of the process at all.
It’s more like… How do I word this, “A base level of reactivity”?
I know, for a fact, that I will end up with 500 or more variables no matter what I’m making, so I start with 500, and if that is not enough I make more. If that new one is going to be used more than once, I make it a *create (VarName) at the top with the rest. If it’s only going to be used once, it get’s a quick *temp (Gzt2b8).
If I don’t use at least 500 variables, I consider that what I’m doing wasn’t worth doing and scrap it. It meant whatever I was doing, didn’t even have 500 variables worth of interactivity, which means to me, I didn’t write something fun enough to play.
I think this is a bad way to think about reactivity, or at least one that creates perverse incentives. The idea that more variables = more reactive is how you end up with tons of choices that don’t do anything imo. One variable with three states that gets checked repeatedly throughout the game will get you a lot more mileage than making twenty variables preemptively and then writing to fill them like a quota.
This is a great way to burn yourself out in fractal complexity. Five hundred variables is a lot. To match the scope you’re talking about hundreds and hundreds of thousands of words, it’s alright to take smaller bites.
I do agree, I do, but this works for me extremely well. I’d rather have more options than less, it allows you to have say 7 different variables effect one scene (The course of actions that are happening, not the file, though that too.) so you can have as many outcomes as you can think of, covered by as many *ifs as you can think of, because you have the code to support it.
If var12, var431, temp6 are all true, you can have an outcome for if any one of them is false, and in any combination. Then, your stats, (vars1-10) can each have an acting affect upon the scene you got, based on what scene you did get because one of var12, var431, or temp6, was a false.
The player should always feel like what they did matters. And the most brute and absolute way to do that, is make what they did matter.
While I don’t set a minimum baseline of 500 variables, I did select 500 or so variables in the poll (including *temp) because I can get up to that point depending on how complex the game is. Many of my smaller games barely hit 80 variables.
Most of the 500 variables are set global variables as it relates to permanent MC customization. Afterwards, I make global variables to track unique character (RO) interactions and progress across multiple files. Next, many are just clues and investigation variables, as I like writing murder mysteries. Finally, I include temporary variables to see if a player is sitting in a scene, very angry, standing, et cetera so that I can adjust what ever is occurring in the scene for continuity. I don’t ever start with a pre-determined variable number, I make as much as I need and delete if I can’t handle anymore information.
I’m not following what you mean by this. In an average scene in a complex game, I might have:
*if (worried >= 4)
You shoot upright in your bed, your heart hammering.
*else
You spring up, palms facing her.
She hardly flinches at your sudden movement.
*if (worried >= 4) or ((friendliness >= 65) or (compliance > friendliness))
"I am so sorry! Am I late? I didn't mean to sleep in!"
*elseif (friendliness <= authority)
"Oh, shoot, I can't believe I overslept. I didn't mean to, really!"
*else
"What the — what? Did I just oversleep?! Oh no!"
Is this what you meant?
What do you mean by perverse incentives? I don’t understand if you mean there’s a lack of interactivity because there’s too much in a file, because it’s not fully remembered by the game, or because you worry an author might forget all their variables? Is this what you mean by meaningless?
It’s very kind of you to consider writer burnout with complexity. I also have to caution making 500 baseline variables and, instead, make as many as you need and delete if it’s too much.
I mean that, if you measure “reactivity” by “number of variables” and think that “more reactive = more fun = better product” it leads to making variables an variation for its own sake, without any real consideration of what might be best for the story you’re trying to tell. That’s not to say that highly variable games aren’t good, you should just be thinking about where you’re going before you start giving yourself a massive checklist of vars to pay off.
Are you actually calling your variables things like var12 and var431? Because that sounds like a recipe for disaster. It would be very easy to accidentally write var413 instead, either by forgetting which number you assigned, or just by pure typo, and that won’t necessarily cause a game-breaking error, meaning it will be very hard to catch.
It’s generally recommended to use descriptive variable names (e.g. stole_cake or steve_backstory) so it’s very easy to remember which variable is which, and if you do typo them, there probably won’t be another real variable with that name, so the error will be game-breaking and very easy to find and fix.
You don’t need to create 500 variables ahead of time; you can just create them one-by-one as you need them.
100% this. Instead of having to remind yourself of what a random generic variable refers to, e.g.
*set var314 6
*commen6 var314 is how many tarts I have
just have *set tarts 6.
If you write a long and detailed enough game, you’ll accumulate as many variables as the game needs, by writing the situations within which they have meaning. Better to hit the 500 or 1,000 mark that way.
