Hi all!
I was thinking about the more classic style of text adventure scoring lately, and am really just curious to hear how folks address scoring in their own games. How do you weight specific actions/tasks, and how do you decide what’s worth scoring?
I’ve used scoring in three games, I think. Here are three approaches:
So it depends what you want to communicate to the player with the score and what the outcome is from having different scores might be.
Although I like score a lot, I haven’t used it as much as intended.
I think Joey’s points are very good. Although it doesn’t have to add up to 100 unless you’re purely measuring how far you are through the game! Infocom used scoring of 400, usually (going up in 5, 10, etc.) or 80 for Planetfall (going up in 2, 3, 4). My favourite scoring management is from Curses. It goes up in scores of 6, 3, 2, etc. I don’t know why. Also Trinity, which uses a similar method, but with a nicer rounder number.
I talked about this (although slightly differently) in this thread: Score, WHY?
I prefer the classic adventure game score, to award puzzle solutions and story progress. Also it’s a nice progress indicator. Also some people find it very motivating to receive a ranking in relation to your score.
I personally don’t find scores particularly important, but nice and informative. More comparable to the number of pages in a book than a real reward.
in Creative Cooking the 101/100 score isn’t a bug, but that the classical “last lousy point” of yore, albeit easy to gain, is an hidden bonus point, averting the classical frustration in hunting for this last point 
Best regards from Italy,
dott. Piergiorgio.
In my WIP, I mostly use it because scoring is a convention of the game’s genre. The score indicates progress or punctuates story beats.
In a couple of cases, though, it is a way for the player to know they have done something productive when it wouldn’t otherwise be obvious.
E: in terms of the numbers, I have only used ten-point increments so far. I also assign a rank based on score, because I always thought that was fun in the old days.
My philosophy on scoring systems is that they serve three purposes, in a way that’s pretty easy on the developer’s side:
If you’re having trouble with any of these issues, adding a scoring system is an easy fix. It’s an unsubtle but useful way to tell the player “here’s what I want you to be doing”.
If I use score as a progress meter, I kind of prefer one point for any successful action milestone like solving a puzzle or discovering a secret. That way the player knows if their score is 47 of 50, there are three more things they didn’t find. I’m less enthused when successful actions net an arbitrary number of points. I opened the magic umbrella which is worth ten points, but I defeated the wizard which is worth only five? In that scenario total score means nothing. If you’ve got 35 of 40 points, is there one action that will net you five points left to find, or five things each worth one point, or two, worth two and three points?
Alternately, if the score represents something like money earned, then different amounts make sense score-wise. I was awarded ten gold for a side quest so my score goes up by 10. Makes sense. Or if it’s clear that points represent something like XP, killing a single imp might be worth 5 where a marauding ogre is worth 15.
It kind of depends on what “points” represents - is it story completion, or XP, or money? In an RPG-like you probably don’t just have a point score, but you have multiple variables representing XP and coins and such. “Points” is an abstract concept that carries over from arcade games where higher point total is literally just a score that denotes success and completion or “how well you played” in comparison to other players separate from game tasks or achievements.
The other variable is if there’s a “maximum score” the player can work towards. There might be some games where maximum score is hidden and thus not a thing a player can use as a progress meter to the end except in comparison with others “I only got 83 points, but you got 120 - what did we do differently?” or in a game that simulates arcade mechanics. if you can kill multiple regenerating beasties for points as much as you want, you can push the score as high as you want and there no limited maximum.
As Hanon mentioned, I’m also a fan of the score system just telling the player how many secret things there are to do. Otherwise, a percentage complete would be better than 147/208 points.
My favourite progression indicator was the inventory screen in Ocarina of Time. (I understand that we’re talking about text adventures with a standard header bar though.)
I wonder what progress bars would feel like when playing. Maybe it’s already been done before? There are lots of unicode characters to play with.
■□□□|□□□□|□□□□|□□□□
■■■■|□□□□|□□□□|□□□□
■■■■|■■□□|□□□□|□□□□
Anyway, food for thought.
@kathrynli’s Goat Game used a sort of dot bar to display which endings you’ve achieved, though it’s not a linear progress bar.
It also has some player stats written separately.
I haven’t actually made any IF games, and how, and even if, I implement score would depend on the game.
That said, I’ve long been a fan of games that tracked multiple overarching objectives and aggregated them into a completion percentage, so I kind of want to do something in that vein.
Something like:
THe map is devided up into a number of major regions.
Each region has a boss monster guarding a major Macguffin.
Each Region has numerous minor objectives that each earn you one of numerous minor mcguffins.
The game tracks how much of each region you’ve visited.
And beating the game requires collecting all of the major macguffins, a significant fraction of the minor macguffins, but not all of them, and you earn say, 5% completion for every major macguffin, 1% for each minor macguffin collected, and how much of the map you’ve visited impacts completion as well, perhaps with a few secrets that net you an extra percentage point past the 100% mark, or a second quest style post game that lets you net 200%.
The Impossible Bottle uses one of these, and since it’s a standard feature built into Dialog I’m going to be using it too in IFComp.
I jumped onto CoG since ChoiceScript has very nice built-in progress and relationship bars. I forgot they also have a built-in “Achievements” tab. This game spells everything out including point totals from screen one:
But if you use ChoiceScript it would be exceedingly easy to make a progress/game completion bar. In this example “Doomsday Weapon Status” is definitely a completion percent.
My first two games had scoring, and I preferred using 1 point for each achievement. Crash had an in-game list of things to do, plus a few extra points depending on what ending you achieved. Galaxy Jones scoring was simpler, but leveraged it by giving the user a visual reward for getting a point.
Other than the ending-based points in Crash, I’ve always used score to indicate progress. “Ah, you got the fromitz board out of the Expando-Brain Computer, you get a point because the fromitz board will be important later.”
(No that isn’t a spoiler for my IFComp entry. It contains no fromitz boards and doesn’t have a score for the game as a whole, even though it has a score in some places.)
Now that I think of it, I have scoring stuff in everything I’ve messed with. Repeat the Ending has a point system. It awards one point for each of 33 in-game collectibles. The number affects the game’s final ending. RTE offers both detailed and brief reporting on the player’s score.
The status bar prints an “assessment” at the right side based on a calculation of collectibles found vs collectibles available at that specific point in the game.
the assesment is “unmotivated”
Brief:
.> score
We are no stranger to failure.Total Fail States Discovered: 5 out of 33.
Nineteen Eighty: 2 out of 2.
Trailer Interior: 3 out of 9.
Trailer Exterior: 0 out of 8.
detailed:
.> cs
You have discovered two of two fail states in the 1980s:
- I amused myself with a rather unfair fight against a magic-crazed primeoid male.
- I grew bored with my search for powerful magic, instead casting the WIN spell.
You have discovered three of nine fail states in the trailer’s interior:
- You wasted a powerful source of entropic magic to tidy up our kitchen, leaving us utterly pantsless when Fast Eddie’s trailer exploded.
- You used the power of the SEETHING ORDER to summon Battle Princess Chiyo to your kitchen. She rewarded our impertinence by cutting us—cutting me—in half.
- You used our only means of getting dressed to destroy a television.
You have yet to test the narrative boundaries of the trailer’s exterior.
In another work in progress, I track “achievement” level which awards one point for every possible ending discovered in the game. Achievement level dictates output in a lot of cases. Since there’s a small number of points, I can display that in a more visual way (the ^ _ ^ indicates completion):
You have discovered one of eight endings:
Is That Winning ^ _ ^
Cats Are Our Friends –
Be Sure to Eat Your Vegetables –
Can You Run in Those Boots –
Space Is Forever –
[Hidden] –
[Hidden] –
[An Impossibility] –
Then we have the games that abuse the scoring system or use it in unexpected ways. This was actually a gimmick almost from the start, and there are plenty of old-school examples (creative uses of scoring tend to be more common as a modern feature, to be clear).
Some examples are outrageous to the point of self-parody: In GIBI0375, five points are deducted if the player answers “yes” to the question “Do you want instructions?”. And in some unofficial ports of Advent, one point is added if and only if the player never saves the game.
In the 90s I used to laboriously put in code to increase the score flag every time a key action was completed. It was always a pain to make everything add up to a nice round number… to choose how to weight each action… or to avoid instances where bugs meant the score could be repeatedly increased by doing the same action again and again.
These days if I do include scoring it is much more based around a list of tasks; especially as I have flags that I am naturally using to track that progress in the game. So I may tell the player they’ve completed X of Y type of tasks or found X of Y items. Or even give them an actual list of tasks and tick them when they’re completed them… adding new tasks when they discover new puzzles or locations.
In the old days, lots of players were completionists… they were willing to replay games again and again to optimise their strategy, to reduce their number of moves, to find all the missing points to achieve a 100% score… I don’t think the majority of modern players are quite as interested in all that.
As a slight aside… although related to an earlier post about how you can use scores, particularly deducting from scores in a fun way… this is my favourite story ever about scoring in text adventures, which comes courtesy of Larry Horsfield writing on the old Adventure Convention website… I hope he won’t mind me reproducing it here… He was talking about the Megapoints Competition where scoring was key…
This game, a feature of every single convention since the first, was the brainchild of Jim O`Keeffe. The idea is that contestants are given 20 minutes playing time at a specially-written text adventure, with the object of scoring as many points as possible. The first game used for Megapoints was a specially adapted version of part 1 of my own text adventure, “Run, Bronwynn, Run!”, a game which was eventually finished and released a few years later! This game was written using “P.A.W.”, the Professional Adventure Writer utility. Winner of the first-ever Megapoints was a chap called Jack Higham, who won mainly because of his incredibly fast typing - his fingers flew over the keyboard so fast we swear we saw smoke.
One fascinating aspect of the Megapoints Game for me, as the author of the text adventure being used, was watching all those different people playing my game in front of me. I couldn’t believe the things they were inputting and was constantly taking notes - it was like a great big play testing session. We thought we’d eradicated all the bugs in the game, but a few more were found (fortunately they didn’t prevent anyone from continuing), as well as a few spelling and grammatical errors.
Various text adventure authors have contributed games for the Megapoints Competition over the years. At the third convention, young Scott Denyer, proprietor of “Delbert the Hamster Software”, wrote a game called “Exploits in a Wheelie-Bin” for the Megapoints. As I had provided the games for the first two years, it was a pleasant change to actually be competing in the competition, instead of running it. When it came to my turn we had, as usual, three Spectrums in use, but I didn’t notice that I was asked to sit at one particular machine, rather than choosing one of the three at random, as everyone else had. Once I started, I found it incredibly hard to score points, even though I seemed to be solving the puzzles in the adventure. Whenever I did manage to score any points, they were swiftly taken away a few moves later! Total score after 20 minutes: 0 points! Everyone else managed to score at least a few points, and I put my failure to score any points down to the fact that I was much better at writing games than I was actually playing them.
Afterwards, when I was discussing the game with some other competitors who had already had their goes (we were under strict instructions not to talk to anyone who hadn’t had their turn, of course), they were mentioning things in the game that I had not seen… and I began to smell a rat! Sure enough, when the winning trio was announced, Scott admitted that I had been the victim of a wind-up and the particular version of the game I’d been playing had been doctored so that any points I scored were deducted a few moves later! It turned out that this plot was hatched at a meeting of The Hamilton Halls Gang in London which I had been unable to attend.
This caused much merriment amongst the people at the convention, and I was rather embarrassed, to say the least, although I did take it all in good heart. I was, however, determined to get my own back, but the next year all but one of the co-conspirators were, for various reasons, unable to attend the convention. The only one who was there was Martin Freemantle, but as luck would have it, he was doing the Megapoints game that year. I suddenly had a brainwave - if they made it so I couldn’t score any points in the last Megapoints competition, then I would score MAXIMUM points at the next one! I knew that Sharon Harwood (another of the conspirators not at the convention) was play testing the adventure, so I phoned and told her what I was plotting and she readily agreed to send me a copy of the final version which would be used in the actual Megapoints Competition.
I played that game to death during the weeks before the convention, learning all the point-scoring inputs so that eventually I was able play through the whole game, scoring the maximum number of points (250) without referring to a solution. On the day of the convention, all went according to plan, and I played the game, “Curse of the Serpent’s Eye”, all the way through, although I did make one mistake near the end and missed a few points. Jenny Perry was supervising the running of the game when I took my turn, and when she saw how many points I’d scored after 15 minutes play (it was over 200), she rushed over to Martin and told him. He was busy attending to customers who were buying games from his “DreamWorld Adventures” table and I remember that it didn’t seem to register with him that I’d scored so many points.
Afterwards, when word went round of the incredible score I’d achieved, people asked me how I’d managed it. I told them - quite truthfully in some respects - that solving a certain puzzle, involving a letterbox I seem to remember, opened up the whole adventure and I was just plain lucky! When it came to awarding the certificates (I had to let a couple of people in on the secret so that a certificate wasn’t completed with my name on it) I stepped up and, instead of taking the winners certificate and retiring gracefully, I asked to say a few words. I told the conventioneers about the previous year’s wind-up and revealed, to howls of laughter, how I’d been able to get my own back on the only conspirator present. As I said all this, Martin just sat with his head in his hands, absolutely squirming with embarrassment. Oh sweet revenge! Afterwards, he phoned Sharon and said a few choice words to her, most of which are unrepeatable here. The 1st Prize Certificate did, of course, go to the person who came second, but who really came first… if you see what I mean?
I don’t worry about a final score–I don’t want to fit things into, say, 100 points, or give the player 2 points for one puzzle and 1 for another.
(I did so once, in A Roiling Original. You got 1 points for the cheap solution, then 3 for the deluxe, then 2 for the deluxe with hints.)
Also in my anagram games I had per-section points, which I think worked well. Otherwise you’d start with 120 (or more) points to get. You could see how big it was immediately.
I don’t mind having 87 points or whatever as that just says, okay, you have 87 things to do.
I also like keeping separate track of necessary and bonus points, so the score in the upper right status line will read 17/75-87, for instance. Then as bonus points become unavailable, or bonus points get scored, the scope narrows e.g. 22/76-87 and if you see for instance 22/75-86 you know you’ve missed something and should undo for a perfect score.
I am putting the final touches on my first inform7 game… so my method might seem simple compared to some… I have:
Total is 107. When you complete your last mission puzzle, there is a “finale” and you get awarded another 107 points, making the total 214 points. Upon completing the mission, a “treasure scroll” appears… that if you examine it 2 times, tells you where all of the treasures are… (first examination gives you a spoiler warning.)
There are ranks along the way.
