My work in progress has multiple endings - two ‘wins’ and a third, anticlimactic sort-of-win - and I wonder if anyone has any thoughts on how to implement score (or even whether to bother at all) with this in mind.
I’ve only written single-ending games before, and I normally just give the player one point for every required action (whether or not it’s a ‘hard’ puzzle, since players notoriously disagree on that anyway) and express it as a percentage. This won’t be great for multiple paths if they don’t all have the same points total. And “percent complete” becomes a bit of misnomer, since you won’t see 100% of the game in any single play session. Similarly giving the score as something like “31 of a possible 45 points” seems misleading, since you don’t need (and can’t get) all the points to win.
Some options that have crossed my mind:
give a score as e.g. “31 out of a possible 45 points”, where 45 is the total number of points available from all the puzzles, even the mutually exclusive ones. This seems misleading as you don’t need all the points to win, and can’t get them all.
make sure every path has the same number of points, perhaps by rewarding some puzzles with more than one point, so it can be “31 out of a possible score of 40” or whatever. This seems kind of artificial. I do want the two wins to be roughly equally hard to achieve, though.
abandon score altogether. I like it, though, as it’s an easy way to let the player know how close they are to finishing. I suppose that can be done through the prose. Maybe keep a running percentage of rooms visited or something.
The trouble with that is that the game doesn’t know which storyline you’re on, and probably neither does the player, at least on their first attempt. There are different puzzles leading to each ‘win’, some shared and some not, so if you solve puzzle X it might take you closer to ending A but be irrelevant to ending B. I suppose I could work out a percentage for each and display the higher one.
(Sorry for being so abstract; I don’t want to give away my plot!)
Or I could keep score towards only the more obvious, conventional ending, and have the other one be more easter-eggy. This means the player will stumble across puzzles they can solve that apparently don’t increase their score, and maybe that will make them wonder if they’re in there for a reason? And I’ll explicitly say “This is ending 1 of 2” at the end, so they can go back and investigate. Having just thought of that, I’m leaning towards it, but I’m still uneasy having some puzzles scored and some not, especially since the alternative ending probably has harder puzzles.
That’s an interesting suggestion. I’ll be replaying Pandora Directive soon-ish, and I plan to use the cheat that allows you to see which path you’re currently on (because it’s Really No Fun for the game to track these things invisibly, as it usually relies on variables that are either blatantly obvious or too obscure). As a player, I really don’t mind it when the game’s that transparent.
In Andromeda Apocalypse I used an Achievement-method, in which you can achieve a total of X out of Y points. To get them all you need to replay the game several times, as many points are given for peculiar deaths or for different ways to solve a puzzle. I used a file to store the Achievements over multiple plays.
Yeah, that’s another good example - it goes back as far as Indiana Jones and the Last Crusade. I never mentioned, Jamespking, how I enjoyed that little touch!
Thanks everyone. I’m trying to go down the route of making all the paths take the same number of points (both for simplicity of scoring and because it seems like a good idea design-wise.) This is the first time I’ve done multiple endings, or even multiple solutions to the same puzzles, and it’s great fun if a bit mind-bending to have to think of all the possibilities.
This does mean that there are some ‘clever’ actions that effectively don’t get scored. A made-up and somewhat simplistic example:
You have to get rid of a dog that’s guarding the MacGuffin
You could either shoot it with a laser gun (for 1 point), or feed it a sausage (for 1 point)
To shoot it, you’d have to find a battery to put in a gun (1 point), and to get the sausage, you’d have to put some meat in a sausage machine (1 point).
So solving the problem gives you 2 points however you do it… but you might shoot the dog, and also put the meat in the sausage machine, giving you a pointless sausage and a stray point. My present approach is: only give you the point for the sausage-making if you haven’t powered the gun yet, and vice versa. But of course this is going to get twisty as soon as there’s a puzzle down the line that might also be solved with the gun (which there probably would be… which is why there isn’t actually a gun in my game.)
Even typing that out has made me lean back towards the “Sod it, let different paths have different scores” route, but who knows, I might change my mind again.
Another question: I’m also thinking of having some of the endings mutually exclusive, in that they couldn’t both be the ending of the same story without the background being different to begin with - a wavefunction of background information that doesn’t collapse till you find it out at the end of the game, if you like. It does relate in a sensible way to your decisions as a player, I think. (Sorry - this is another thing I don’t want to elaborate on because I don’t want to give my plot away; hope I’m making sense.) Is this “allowed”? Has it been done much before?
You can do that, but it will wrongfoot a lot of players unless you cue it clearly from the beginning. Try making that element of the background a blatant mystery, rather than just leaving it unremarked.
