Some background...

I'm making a turn based RPG/strategy game. It involves a girl who gets trapped in a mansion-maze with some creepy dolls. In every level, she finds a key somewhere and needs it to defeat the big doll boss.

Combat is similar to D&D 3.5. Square grid, 8 dir movement. Mostly melee attacks, with a few projectiles, like some dolls throwing rocks/knives/needles. Main damage type is psychological/sanity where the girl loses the will to keep going and game over. There's still physical damage, dolls will try to break her limbs, stab her, etc, but to prevent the game from getting too gory, physical damage is mostly used to increase the helpless feel.

The feel of the game is to make the player feel like a trapped rat. Unlike in most RPGs, the default stance is to escape, while fighting is used when running is not an option. Many enemies will be sort of undefeatable, like cloth puppets, which can be knocked around with a baseball bat or thrown, but won't be neutralized.

It is a turn based game.. I want to have it feel like X-Com. End turn, pray that no enemies are around the corner, pray they get unlucky and miss you. Start turn, keep running or fighting and try to end turn in a location where they can't get ganked.

Surprises are a core part of horror, but the player should be able to anticipate a surprise. A seemingly dead doll might rise and grab her, might jump out of a box, might break down a door, or run from around a corner when you hear footprints earlier. But shouldn't pop out of nowhere and the players would learn possible hot spots early in the game.

Difficulty is mostly so that the player can get through intuitively, without having to repeat and memorize the level. Deaths should feel deserved, as in the player did something stupid or made wrong assumptions on the level/enemy behavior. The player should die very rarely, but the threat of dying/losing should feel real enough. I don't like a savescumming level of difficulty as it ruins the atmosphere, and goes against the "you can still survive" feel of the game.


So, on to the main point...
It looks like a large segment of the game depends on the level design, so I want to take it quite seriously. There isn't a lot of manuals and examples on the TBS horror genre, so looks like I'll have to make something up. I've brainstormed sort of the following elements in a level.

Damaging trash: An encounter designed to weaken, but not kill the character. Not really for any tactical purpose, but it's sort of to get the character at a moderate health level to make them worried. And to discourage people from being perfectionist savescummers.

Mobs: Weak encounters, mostly harmless, but they keep her from getting through. Might be combined with a tough encounter as some sort of artificial wall, might just be there to give the player something to hack and slash and burn through.

Chase: I like chase encounters in horror games. It would probably involve a highly damaging or undefeatable creature. The player would be running into a place they can't follow, desperately trying to open doors or crawl into things. But not certain how it could be done in turn based games without being annoying.. run, end turn, run, end turn.

Entrapment: Enemies try to trap the character in a position where she can't fight and would be whittled down.

Treasure: Often combined with entrapment encounters, would be the key, save point, or something like a weapon. Usually well guarded, the player is made aware of the risks and a possible treasure.

Decoy: Acts like a Treasure but turns out to be nothing, or something very minor. Less damaging than an actual treasure encounter.

Surprise: Often used to spark Entrapment encounters. Or to ruin some kind of plans, like make a Treasure encounter tougher than the player expected. Can't really think of a good use for them, aside from giving an "Oh shit" feel.


I haven't really figured how and where I'd use them. If anyone has any extra ideas, suggestions, tips, or a good article/book on it, I'd be very interested.

Personally, I like achievements as a sort of guideline. It's nice, since I normally play open ended strategy games. Something like "become emperor of the holy roman empire" or "established a thriving colony". IMO, achievements are best as a sort of checklist.

Yeah, I'm aware of this syndrome after a few years of making games. I really have no defenses for it. The best I can do is make excuses - "It was rushed, it was the most technically complicated so far, STFU". And that art game defense worked great for my last game. And "I'll fix the gameplay in the sequel".

The funny thing is that there's always someone who thinks that my baby is cute, no matter what. I could call them blind, but they still won't budge and will cheer on my game despite the broken bugs and beg me to hurry with the non-existent sequel. I've had a guy follow an ugly, broken, vaporware game for 4 years, before he sodded off when I stopped replying his emails.

On the other hand, there will be a lot of people who'll hate your kid for no reason. I guess they're easy to tell in that they tend to be shrill and illogical, but the careful trolls make me question whether my game would really suck or not. But if say, someone who loves Halo criticizes my Civilization style game, I guess I could brush it off as simply not being his kind of game.

I love the team aspect of making games. I'm great at macromanaging, but I suck at micromanagement. Basically meaning that I can't manage myself when doing it solo. I keep flip flopping between different tasks and just end up accomplishing nothing.

It varies. I think design documents are essential. But treating them as well, some bible is just pushing it too far.

In reality, your game will change greatly from its design. People will find exploits you didn't expect. Something you suggested may turn out to be a bad idea. You could add in some things that screw with the balance of everything else.

But.. design documents help a LOT in the game development process. If you have a complex enough idea, you'll have a huge headache keeping it in your head while trying to code it. That's what design docs are for. They give you a plan to follow and so you don't have to think about 4 different things at once.

Personally, I think that formatting etc doesn't matter. What you really want is a full description of what to do. The big difference between documentations for a game made by 200 people and a game made by 1 person is that how easily understood they are. With a single person document, you'll still describe things from start to finish. But the one for less people will have minimal formatting and minimal descriptions.

Anyone who thinks that it takes too long and you'll accomplish nothing.. well, you just need practice! Not practice writing down random junk; practice in trying to create a document while making the game. Make lots of small games until documentation becomes second nature.


My best design documents are normally just a bunch of notes scribbled together. I'll get a heap of ideas, I toss them onto a txt doc. If I'm working with someone else on it, I paste logs of conversations with her, copy images/vids/links for possible concept art.

There's a part on a short description of the game - what it is, what it intends to achieve, the big picture. This helps in making sure you don't go off track from your original idea when you'll have to make changes.

Then I work on the core design of one "demo" level, which is supposed to capture the heart and soul of the game. This is what attracts the other people I want to collaborate with as well as the players.

After that, I branch out to working on the loose details of it. How many hit points? How much speed.. don't give a number, just something like Monster A can outrun Hero, but can't jump as high. Fine tune it to numbers later.

Then with the demo level complete, I use it as a prototype. Get the game and the feel right, and work outwards from there, until I can describe everything I want to do.

I don't think there's any games that's truly endless. I get bored of most of those games eventually. It's just seeing how many different things you can get out of a roguelike, before it becomes all the same thing.

Spelunky did a great job with this, though. By the time you can spot the sameness in a level (and get bored with it), you'd have a shortcut to the next group of levels.

Personally, I do like an end to an 'endless game'. Dungeon Crawl and Nethack had that orb stuff to "end the game" when your character was becoming too powerful to be stopped. I'd rather it be something like that than Tetris, that tries to 'kill' you with the game mechanics when you get too good to enjoy it.

I'd agree with what people in the thread say about game design being an art rather than a science. I've worked with hardcore engineering, artistic engineering (emotions in voice), and just plain art.

To me, a science is something simple enough that can be accurately explained by laws. Even the most advanced sciences are still simple enough to be broken down into parts.

Art is something that's so complicated that you can't explain them with rules. You can only explain patterns, which patterns work and which don't. Creativity is simply having so much experience with the patterns that you can spot on sight what a good pattern is.


Trying to turn game design into an art would give it similar success to other entertainment medium. Fully analyzed game design would produce something like classical music. You can synthesize a work of beauty from it, but highly theoretical. And the results would be the same too - it would appeal to a group of people, but not everyone. A classical musician may shun garage rock bands, pop, and auto-tuners, but the fact is that the ones played by instinct have larger audiences.

However, theory is not all useless. The indie guitarists who were trained classically can create some amazing metal, since he knows some musical patterns that work. Same goes for an indie game designer with theoretical knowledge.

On the other hand, the movie industry has a formula, sticks to it, and churns out good movies often enough that there's little risk. It's not impossible to get a good formula for art.

I'm not sure where game design lies compared to other entertainment mediums but it's somewhere around there.

I'd like to chip in that the reason I hate RTS games is because the units are so stupid. The game that gets the most right that I've played is the original Rise of Nations. The units are smart generally enough to not get killed pointlessly (they react to attack somewhat intelligently) and they're stupid enough to do what you tell them.

For me, the balance is having priorities right. Following my orders shouldn't be a unit's top priority; that's unrealistic and idiotic. No, my orders should be second and preservation of self should be first. (Then after orders, defending structures and other units should be third.)

There are a lot of RTS that run when badly injured. But they didn't catch on for some reason.

Personally, I don't think of a RTS as a strategy game. They're games where you memorize build orders and 'tactics'. People play them as twitch sports, just like people would play basketball or ping pong. They're pretty much games where winning and losing is the whole point.

If I wanted a strategy game, turn based strategy works a lot better. Something with a much slower pace like Neptune's Pride or Supremacy strikes a great balance between TBS and RTS at the downside of being something you check all day.

Yeah, I'd echo what everyone is saying about this not sounding right. A good theory should be like an epiphany, shouldn't give you a "Huh? What?" feeling. Unless it's sufficiently complicated or poorly explained. A Theory of Fun doesn't sound like my kind of book, though I'll check out the other two.

My favorite book so far is Chris Crawford's On Game Design. There's a lot inside which I disagree with, but overall, it's been quite insightful.

I do believe there are some things you could learn about game design, but so far formal game design's been counter-productive. I'm not fond of any of the games since the game design revolution. I've stopped having up to date computers because modern games are too shallow. I'm mostly just playing old games and indie games. Modern games have too much of this focus on "don't have stuff that's not important". Well, I like a lot of that stuff. I play games for the same experience that people try to get from books.

One thing I noticed is that there's a lot of theory for casual or competitive games, but very little for things like roleplaying and simulation. There's a reason that people who love Fallout 3 hate Fallout 2 and vice versa, it's because the things that make either one great fails horribly for the other. All the unnecessary, unbalanced stuff makes a roleplaying or simulation game feel a lot more fun and challenging.

In game penalties are good. Out of game penalties are not.

Out of game penalties

What?

Mainly things that you value IRL like time and money. Something like, I put $15 in to make my char have higher strength, then someone puts a spectre with a weaken spell that permanently reduces strength. Or something like spending 15 minutes in the game without a save, then getting killed and having to start over, even though the player knows to avoid whatever killed him later.

On the other hand, killing someone 15 times, when each death costs only 1 minute isn't as bad. As long as you give an out of game reward, like an achievement, or letting them explore a new area.

More suggestions, then we need a compo based on this thread.

Personally, I find knowledge to be a tricky thing to play with. It either improves the exploration or gets really frustrating. I wouldn't enjoy fighting "cute bunny creature" only to find that it (nearly) kills me in one hit. Or similarly fighting a creature who looks tough, but turns out to have low HP.

Eventually, I'd just give up and look for a wiki on the thing. The likelyhood of frustration is higher, not worth it unless you're targeting it with a group of people who enjoy not knowing.

This is amplified both ways if it's an online game, where people compete with/against each other. Exploration gets more fun, frustration gets much higher.

I start all my games with a solid game design.

But all the games that I've actually finished, I just said screw it, and started coding the game with a half assed design. Had to remake stuff halfway through, but in the end I finished making the game.

OK, let's see how this goes. Since there's usually a lot of designers, I'm hoping someone has a bit of extra time.

I'd like to build a RPG system. Right now there's two parts that I want to work on:

1. I'd like to make a non-linear stats system. This should allow to calculate stats ranging from fairies to giant robots. The system is broad enough to be kept simple if needed (like just a "mental" stat for a combat heavy campaign instead of intelligence/wisdom/etc). It would be detailed enough to say, make a sports management game by refining certain stats, but you could collapse those detailed stats back into a "physical" stat if needed, like to transfer one character to another campaign or system.

It'll be completely computer run (no dice), so allows for much more complexity.

2. Body part damage system. Fairly simple.. guy hits, arm gets scarred, chopped off, etc.


For the first, the system itself is already designed, but getting the stats needs a bit of work and research. Testing it out needs even more work.

For the second, also similar, getting the numbers and testing it out with different conditions. I want to keep it as simple as possible, which is a challenge.

It's not really hard stuff. I just want someone to talk to with about it and get some ideas. No pay, except for some credit in any game that I make with it. And it's a good system too, so anyone who's making a RPG should enjoy and get something from it.


Drop me a PM if interested, I might forget to check this subforum

In game penalties are good. Out of game penalties are not.

Roguelikes and a lot of online games are stuffed to the brim with penalties. Drink from the wrong pool and you get eternally cursed (or die).

The reason that people don't mind them is because they're not repetitive. There's actually something different to unlock each time you get penalized. And in some cases, you get closer to the Big Goal even if you get penalized. Spelunky does this brilliantly. If you die from one thing, you're given a different level so you don't have to memorize the same button presses.

The wrong way to do it is to penalize the player for things he didn't do. It makes the player frustrated. Games are also about discovery, and too much punishment will slow the player from discovering new things.

And here's a good blog article about how Super Meat Boy gets it right:
http://jeff-vogel.blogspot.com/2010/11/review-super-meat-boy.html

I didn't read the whole thread yet

What I do first is identify the kind of players who want to play my game. I write out a bunch of things that they'd like. I make RPGs, so I boil it down to people who want to blow up/kill stuff, people who want to explore, people who want to have a cool character, people who want to change the world. Effectively writing down the kind of gameplay I want.

Then I work on the story. I make sure that the story has all those gameplay elements. There'd be a story path for killers, there'd be a story path for heroes/villains, there'd be one for explorers, there'd be one for a shiny guy.

And I try to merge them around a bit. Someone who's playing a social/killer kind of character might be allowed to duel or assassinate to gain political capital or lead armies against other nations. Someone playing an explorer/social might be allowed to make deals with tribes, or marry a tribal girl and retire in the jungle.

Once I have all the story I want, I build the levels around these stories.

I'm usually a competitive gamer, so I play without cheats. "Easy/Beginner mode" is pretty much a good way to make the game casual, without letting anyone cheat. But some people still love cheating anyway, they get fun out of it, so putting in cheat codes is a decent idea. I doubt a lot of people here are casual gamers, but a lot of my uncles and my parents play games only if they have cheat codes. They play games as sort of an interactive movie.

Still, there's one reason to cheat - getting to the ending. Some games are just unfairly brutal, or long. Mainly games where the storyline is far better than the gameplay itself.


Multiplayer cheating on the other hand should be wiped out. It ruins the game for everyone, including the cheaters. At worst, make money out of it - if someone is so determined to get ahead of everyone, make them pay for an in-game advantage.

Cheaters usually want the fastest and easiest way to get ahead and will only use multiple accounts, proxies, etc as a last resort if it's possible to pay. Doesn't apply to everyone, though.

There is a bit of dumbing down. It's not as bad as everyone makes it out to be, though. If it was really that bad, then I'd be playing only the old games. Fallout may have been dumbed down, but you still get good quality out of say, the Paradox games. The Dominions games continue to get more complex, and so has games like Roller Coaster Tycoon. And on top of it all, you have games like Aurora and Dwarf Fortress, which didn't exist 10 years ago.

A close relative of mine has a MBA, is on the board of a few businesses, and I asked him for an opinion on how to build a game company. His idea is to simply "do what has proven to succeed", because that's how businesses work. He does play games, not too hardcore, more like things like Starcraft, Age of Empires, SimCity, and Superpower 2. But the businessman's job is to sell a product to as many people as possible.

There's no real formula to predict success, only using something that has worked. Blizzard has the closest to a success formula as any other company.. none of their games are too original, but they're the biggest in their genre. They're a newcomer to the MMO field, and now WoW is synonymous with "mainstream MMORPG".

Games are an art entertainment medium. People love something different, but that goes in direct opposition to ways to make money. The designer's job is to make it as beautiful/artistic/fun as possible. The producer's job is try to stretch it out to as wide an audience as possible, and those two will butt heads. Where risk is high, the producer will win. That's why we have indie games.

Personally, I think the best approach is to just take a good concept and tone down the learning curve as much as possible. The "casual gamer" can really get hooked on a game like Dwarf Fortress.. as long as they can love the concept from seeing a trailer/screenshots and understand how to play it in half an hour.

But hey, I failed at X-Com when I was 12 and still had fun. I don't see why games need to be dumb if it takes out from their fun. I've got fairly dumb friends, they loved Baldur's Gate, Stronghold, Age of Wonders, Civilization. By contrast, if you dumb down a game too much (like into Farmville level), you'll lose the college educated audience, and that's the audience that actually has the money to buy your games.

Hmm.. the problem seems to be that you're documenting answers to problems that nobody experiences. I think one better way to approach this would be to document answers to problems that people do have trouble with. For example, just about every thread here is a problem. If you could get a summary of the approaches to those problems, then you have something much more useful.

How someone can argue for grinding is beyond me.

Because thousands (millions?) of players play games with heavy grinding involved. Some enjoy it, some see it as a "the lesser of the evils", while the creators just see it as the easiest way to make money. Charge $30 for every player or convince a single player to pay $300 to skip the grind... it's a win/win situation for a lot of people.

The point is, grinding is a lazy mechanic because it says "You are not Number enough to do this action. Play our game for hours (without any real challenge) until you are higher Number".

To me this is mostly the reason I consider it bad. It serves as a lock for further content. The player is no longer playing for fun, but rather, they're weighing the cost of unlocking that content with the entertainment they will gain from that content. You could let the player move on to interesting content now and they would lose that boring time-sink. To me that is grind executed poorly.

The purpose of grind is always a content lock/dilutor. It's not always a bad thing, only when it's overdone or noticable. The reason most people like the grind is that it lets them unlock unique content, but doesn't allow others. You could be the only person in the whole game world to get the +300 Crystal Sword of Norath after clicking in a certain location for 3000 hours. That's the kind of achievement people go for it. Others may be able to buy 1000 hours of clicking for $20 (which may just be an hour of RL work for them).