🎉 Celebrating 25 Years of GameDev.net! 🎉

Not many can claim 25 years on the Internet! Join us in celebrating this milestone. Learn more about our history, and thank you for being a part of our community!

Game design problems: from story to algorithms..

Started by
29 comments, last by Niphty 24 years, 1 month ago
Hey everyone! I signed up recently, and have been perusing the discussions I''ve got a very good question to ask regarding game design, and I think it''s an interesting aspect I have yet to see covered. That is, when designing a game, you write out a nice story or something to companion the game (mostly in RPG.. although even tetris has some kinda lil ''this is what you do'' story). Then you make the game off this story. Well.. there''s the translation step that I''ve yet to see drawn out very well. Now, what I mean is this: When you say "XXX player hits XXX monster of X damage" in a story, and then try to translate it.. you hit a problem. Math. Designing how to determine many factors. Does the player hit? if so, how much damage? It''s all about the balance aspect. This is, of course, from an AD&D setting, but all games have some math involved. Even in pong.. you have to determine what parts of the paddle bounce the ball back at what rate and at what angle. It''s a lot harder, however, to try to balance an entire game system when you''ve got nowhere to start from. So my problem is this: how exactly do you translate the story into a working, balanced game system without losing out aspects of the story (eg. reality, fairness)? My girlfriend and I both have a nice story designed out in our heads, but we''re forced to wait till we can code it to try to balance the system (it''s a lot of numbers.. RPG style game with stats and skills to factor in!). Any help would be appreciated, thanks! J
Advertisement
What I can say is: Balance is HUGELY important. A little anecdote for you...

My favorite game EVER is Panzer Dragoon Saga on the Sega Saturn. Play it if you can. But even this perfect game has a huge lull in the middle of it''s four cds, brought on by a lack of balace in the difficulty of the battles. For a stretch, nothing can hurt you, at all. It bored me to tears, there was no threat.

So anyway, you''re right to worry about balancing the game well. Final Fantasy Tactics was great at this, because it calculated the enemy difficulty based on the player''s current EXP levels. So as the designer, you can describe a battle with levels such as: Very Difficult, Easy, Hard, etc. The programmers (you, I presume) need to make it so the program looks at the players EXP, then uses that to calculate the opponant''s attack rate, speed, damage, powers, Ai difficulty, etc.

This avoids the syndrome in the above-mentioned anecdote. Also undermines player''s efforts to horde EPs, and in doing so gives the hack/slash element a back seat in your game.

On a side note, I hope you haven''t grown attached to your story down to the letter. I find that involving your whole team in the creation of the story''s elements ups the enthusiasm of the whole team. Just a tip.

Your girlfriend does CRPGs? Hang on to that one...
-Landfish

Where does the Landfish live? Everywhere. Is not the Landfish the Buddha?
======"The unexamined life is not worth living."-Socrates"Question everything. Especially Landfish."-Matt
AI that weekens itself to suit the player? Hmmm isn''t that cheating?

The_Minister
1C3-D3M0N Interactive
[email=mwronen@mweb.co.za" onmouseOver="window.status='Mail The_Minister'; return true" onmouseOut="window.status=' '; return true]The_Minister[/email]1C3-D3M0N Interactive
No, an AI that is always as challenging to the player as the designer wishes it to be. No matter how much exp gaining you do, it always comes down to how good you are at the game. With this system in place, the player is never mindlessly pushing the same button. That''ll get him killed.

It''s subversive, but believe me, it''s effective.
======"The unexamined life is not worth living."-Socrates"Question everything. Especially Landfish."-Matt
If the computer always adapts to your current condition, what''s the motivation to become stronger?

When figuring out points, you should start at at strength of 100, then figure out what the player''s strength should be at the start of every level if they played well. Use this number to figure out the strengths for the baddies for the next level. Try to make the end of one level lead up to the difficulty of the next. To make difficulty settings (easy/normal/hard), just make it harder or easier for the player to get to the predicted strengths.

This is a simplified version but you can extend the idea across the different variables of the game.

E:cb woof!
E:cb woof!
Landfish, your ideas are much appreciated, however.. you run into a barrier in an MMORPG. You can''t have goblins above a certain level, at least not many of them. Assuming goblins the lowest, then you''d have people killing these things off day in and day out. How would a goblin manage to become a leader? LOL!

You also note another good point.. slaughtering hordes of low-level monsters doesn''t offset the lack of higher-leveled ones. However, in a realistic world.. how do you settle this problem? Certainly there''s no persisting menace that will last ages? I mean, goblins in most MMORPGs must have a life expectancy rate of 2 hours, and a gestation period of 45 seconds! That to me seems a bit off. I disagree with the whole "monster generators" to a degree. Having constant baddies pop up from nowhere doesn''t seem like a very big roleplaying aspect. If a monster crawls out of the bushes, and the game disallows the player to go into the bushes, seeking the goblin horde.. well.. there''s no fun there! To me, it should be that goblin troops rise up every so often and do battle, perhaps even try to takeover a city! A spying ranger or passerby could warn the town of a giant camp full of goblins, and then a war party would form up to battle the threat!
Even if you have some form of portal allowing baddies entrance.. they would eventually run dry, unless they''re undead and killing them merely sends them to the other side of the portal. This is an idea i''ve toyed with in a D&D campain.. it''d take time afore the adventurers found out that they need to focus on the portal while a few hold off the undead by then, they might not be alive enough to fix the problem! It adds a surprise! element to the game. Forces people to really think, instead of just hit a button over and over, which we all agree is useless on both ends.

However, high level characters giving low level characters a tour, or some practice at combat would be going beyond mere hack''n slash. This way the fighter might be doing nothing more than whack-whack-whack, but it''s for a better cause! And believe me.. this is something we heavily support.. as we''ve even put in plans for a system to gain minor experience from watching someone fight. You must actively watch though, not just be partied with the person!

But once again.. it''s all down to how to translate that into the actual game system without killing all other aspects. I feel like i''m building a card castle without a steady hand! One card doesn''t fit and boom, the whole thing tumbles. Leave out a key card.. and boom, it''s gone. And you put a lot of time into something like that.. so you wouldn''t want it all to colapse! It''s hard to pick up and get going again after such an incident. So.. i guess i''m looking for some support in this aspect. I know balance is key, but.. how do you achieve that balance? where to begin? do you make the player and then set the world around them.. or the world and then put the player in the world? LOL how would you work it if you had to fill god''s shoes?! LOL! well.. just some food for thought..

J
The best way to balance an RPG without actually coding is going to RPG''s roots - paper. Stage mock battles/encounters and balance them on paper. Set up various encounter scenarios and battle them out. Use dice for your random number generator. You don''t have to limit it to encounters, you can do the whole shibang on paper - from world/map layout to encounter probabilities to skill sets to obstacle situations (doors, lock picking, etc). Not only will you get a more balanced game, you''ll have a good start on a design document for the game.

Where did you think D&D got it''s start?
The only problem is the complexity of the game. We can''t exactly just use pen and paper because the formulas for things are going to be hella big I want it as life-like as possible, no "roll 1d20 to see if you hit" BS. It''s all about the skills you possess vs the skills they possess. This way it''s more like reality instead of some huge random element generated by a pseudo-random generator. It''s complicated to say the least And it''s why we need to work on testing every aspect of it, but pen and paper simply won''t do.. it''s far too complex. we''d have to use calculators, and even then it''d be slow going.

J
Were we talking about MMORPGS? I thought that was the other forum. Oh well, in that case...

It should be noted that I am a pretentious weirdo. I don't believe that expendable creatures should have any role in RPGs. I don't agree that any creature is expendable, I feel that that is a relic from the kinds of D&D games that I hated playing.
Ever considered finding a player for each and every goblin? Set them up as a misunderstood tribal society; that way, they still raid the occasional town, setting up for some adventuring party to go massacre them, if they can...

This comes from a D&D game I played when I was twelve. My brother had me encountering suff like the mother bugbear in her hut trying to save her kids from (us) the marauding adventurers. You can see how that would shift my perceptions a bit. Has anyone ever considered making a game about the reprocussions of goblin genocide?

Frankly, Leveling sucks (IMO). All it does is reward those who spend more time hacking buttons. We need a system that rewards good roleplaying, or good twitch skill, or anything BUT pressing a button mindlessly. (I'm sorry, that was a rant i've needed for a while.)
I suggested something in the other forum of Niphty's, I'm going to respond to it there.

(PS- Ha! This is great! Finally a good topic in the design section... no offense to previous posters)


Edited by - Landfish on 5/4/00 1:01:31 PM
======"The unexamined life is not worth living."-Socrates"Question everything. Especially Landfish."-Matt
Niphty - I still can''t really tell how much of a programmer you are from your posts, but the solution to the complex battle stuff is really just good programming. You said something in the first post about waiting ''til you can code the game to try and balance things, but I don''t see what the overall design and story have to do with the details of balancing the combat system. You should be able to finish the design and the story without writing any code, and if you make the combat system code modular, with each part well encapsulated, you should be able to tweak the numbers all you want after it''s done to get that balance you''re looking for. I mean, it''s all just formulas, right. Get a damage number from the weapon, multiply by attacker''s skill, reduce by monster''s armor rating, etc etc. You should be able to code in the formulas and modify those other variables (armor class, hit points, whatever) to balance it later. Of course, if there''s a fundamental flaw in the formulas you''ll have to start over, but if you''ve kept the battle code isolated you should be able to rip it out and put in something new without affecting the rest of the game. Does this make any sense, or am I misinterpreting your problem?

Landfish - I have a question about how you could use the modified enemy skill thing in an RPG setting. If you want to have a large, wide-open world for the player to explore, where he may encounter any kind of monster at almost any time, how do you avoid a situation like a level 20 character actually being challenged by a formerly low-level monster? I mean, say early in the game he runs across something like an Orc. The game looks at the player''s experience level and adjust the Orc''s stats to be a fair challenge. Now when the player comes back much later in the game he should be able to totally slaughter that same creature. It seems like you''d have to come up with a mildly complex system of constraints for the monster stats, because the player will go through such a huge range of skill and experience throughout the course of the game and the monsters'' abilities shouldn''t vary that much.

This topic is closed to new replies.

Advertisement