Designing a whole game from scratch

[CrazyCrab]

Hi everyone,
for a while now I have always wanted to make an entire RPG system by myself and recently I have gotten quite some time to spare. I have my ideas, systems that I want to replicate somehow and pretty much a solid base.

I have been designing videogames in my spare time for about 8 years now and DMing for around 5, so I feel like I have enough experience to take care care of it. That said, where do I start? I know how to make a video game, but I feel like the process is significantly different, in quite a number of ways...

So, any tips?

[Professor Gnoll]

If I had to give some advice?
Make it do something that another system doesn't already do. So don't make "D&D, but slightly different mechanics". Give people a reason to play this system over playing another system. Maybe it allows for incredibly fast combat. Maybe it simulates things in excruciating detail, for those who like that. Maybe it invites huge amounts of improvisation for free-form play. Just make it do something that other systems don't.

[CrazyCrab]

If I had to give some advice?
Make it do something that another system doesn't already do. So don't make "D&D, but slightly different mechanics". Give people a reason to play this system over playing another system. Maybe it allows for incredibly fast combat. Maybe it simulates things in excruciating detail, for those who like that. Maybe it invites huge amounts of improvisation for free-form play. Just make it do something that other systems don't.

Oh, don't worry, naturally I have that in mind.
While DM'ing DnD I have come to the conclusion that I sort of don't like the slow combat, its too methodical and calculated with 'I have 50 Hp left he cannot kill me the priest will cast bless before I move so I will hit with +1d4...' so I ended up with having a faster, skirmish-tabletop tactics game (ala Infinity / Mordheim) which makes everything very nasty and deadly, even the smallest enemies.

Other than that I have a pile of cool mechanics and the like. I feel like it will have enough original ideas to stand out to be honest. :)

[Final Hyena]

+1 to gnolls advice. Following the leader will usually result in something that few care for. They're the leader for a reason they know what they're doing and have been doing it for a while. Not to mention the main thing, they already have their audience/market share they wont leave for minor tweaks anyone could homebrew.

I would advice having a look around at different game systems to give yourself an appreciation for different mechanics that have their own affects on play.

Have a sit down and ask yourself, what do you want to happen during play? Is it a game focussed on combat as a main mechanic? Or is it focussed on political sleuthing? Should the systems be light simple and easy to use or very in depth with table upon table of information? You want to have the mechanics all feed into the play you want to encourage. The obvious example is dice. What dice are normally used? Is it a d100 resulting in high rng or 3d6 resulting in lower rng? Another is lethality, should players be careful and considerate or do you want an anime of explosions and over the top fights. This should lead you towards the type of challenges the players should deal with along with other aspects of survivability (HP as the obvious mechanic).

I am a fan of the saying, art is not achieved when there is nothing more to add, but when there is nothing you can take away. The systems in place in your game should have a purpose, mesh together or in short a car doesn't need 5 wheels!

Aside from that remember that the readers are interpreting the game, making it as user friendly as possible is a great boon. An unnecessarily confusing game will put people off.

Also you should give your game a turtle, people love turtles.

[qwertyu63]

I'd like to refer you to a post (http://www.giantitp.com/forums/showthread.php?328888) I made regarding this exact topic.

[GnomeWorks]

Read more games.

Read Pendragon. Read Dogs in the Vineyard. Read Nicotine Girls. Read Ryuutama. Read GURPS. Read Burning Wheel. Read Eclipse Phase. Read Amber. Read Apocalypse World. Read Gumshoe. Read Big Eyes Small Mouth. Read FATAL. Read Dread. Read The Extraordinary Adventures of Baron Munchausen. Read The One Ring. Read Mouseguard. Read Traveller. Read Fate.

As you read, don't just read, analyze. If your goal is to make a game, you need to understand what makes a game a game. You need to understand all the pieces and parts that can be used to make a game. Do you think that this mechanic works really well, or that that mechanic is terrible? Think on it, examine it from every angle, grasp why you think a thing works well or doesn't not and figure out why you think that. If you change your opinion on a mechanic later, make note of that, and try to walk through your thinking process and examine why you changed your mind.


While DM'ing DnD I have come to the conclusion that I sort of don't like the slow combat, its too methodical and calculated with 'I have 50 Hp left he cannot kill me the priest will cast bless before I move so I will hit with +1d4...'

Can you explain your thoughts on why D&D behaves this way? I'm not asking you to actually do so, I am asking you to consider whether or not you can.

The single-quoted statement you have made here is full of things that tie into the mechanics, and thus the design process, of D&D (which, lacking an edition qualifier, I'm not sure how to parse anything you said in specific). You need to be able to tear that statement apart into its atomic game design elements, examine them, understand each both on its own and in conjunction with the remainder. Until you can express, in a clear and concise manner, why that statement bothers you, in terms of mechanics and game design, you are not ready to design your own.

That said, plenty of folks - myself included - start designing games before they're ready. And what they produce is almost always crap, derivative rubbish, not worth the time it took to write. You are by no means required to take anything I've written here to heart, just bear in mind that if you just forge ahead blindly, the end result will most likely be terrible.

[Blue Ghost]

That said, plenty of folks - myself included - start designing games before they're ready. And what they produce is almost always crap, derivative rubbish, not worth the time it took to write. You are by no means required to take anything I've written here to heart, just bear in mind that if you just forge ahead blindly, the end result will most likely be terrible.

On the other hand, just because your first attempt might be bad doesn't mean it isn't worth attempting. You learn a lot by doing, after all.
But I agree, understanding the components of a game and how they fit together is crucial to creating a good game.

[GnomeWorks]

On the other hand, just because your first attempt might be bad doesn't mean it isn't worth attempting. You learn a lot by doing, after all.

Oh, for sure.

[CrazyCrab]

I'd like to refer you to a [] I made regarding this exact topic.

Thanks! It seems like a solid start. :smallsmile:




Can you explain your thoughts on why D&D behaves this way? I'm not asking you to actually do so, I am asking you to consider whether or not you can.

The single-quoted statement you have made here is full of things that tie into the mechanics, and thus the design process, of D&D (which, lacking an edition qualifier, I'm not sure how to parse anything you said in specific). You need to be able to tear that statement apart into its atomic game design elements, examine them, understand each both on its own and in conjunction with the remainder. Until you can express, in a clear and concise manner, why that statement bothers you, in terms of mechanics and game design, you are not ready to design your own.

That said, plenty of folks - myself included - start designing games before they're ready. And what they produce is almost always crap, derivative rubbish, not worth the time it took to write. You are by no means required to take anything I've written here to heart, just bear in mind that if you just forge ahead blindly, the end result will most likely be terrible.

Thanks for the feedback, I feel like you've given me some real food for thought. While some would probably see it as harsh, I appreciate brutal honesty. I feel like I do know why DnD is so methodical... but you're right, only by narrowing it down will I be able to design the exact thing that I will be happy with.

This is going to be a lot of work, but I think it's going to be worth it.

[GnomeWorks]

Thanks for the feedback, I feel like you've given me some real food for thought. While some would probably see it as harsh, I appreciate brutal honesty. I feel like I do know why DnD is so methodical... but you're right, only by narrowing it down will I be able to design the exact thing that I will be happy with.

Couching things in niceties is a waste of time. Better to face hard truth than to try to sugar-coat things.

I will also highly recommend reading char op stuff. Not because it's useful in and of itself, but because it will help you understand the thought processes that go into manipulating and breaking game mechanics. If you can learn how to look at a mechanic and see how it could be abused, that is immensely helpful in designing your own, as you can foresee potential abuses and figure out either ways around it, or that the mechanic will wind up unbalancing your game and removing it or significantly altering it.

Obviously if you read char op you should start with a system you know, so that you can follow what's going on, but really it's not the system that counts, it's the thought processes that go into char op that you would be trying to learn and figure out how to reverse-engineer and keep in mind for your own game.

[Knaight]

Start small, start short. Familiarizing yourself with a wide range of games has already been suggested (I'd use a different list, but generally the exact contents of the list are besides the point), and when I say familiarize I mean read, analyze, and if possible actually play. Beyond that though, trying to get something smaller in scale done first is a good idea. Maybe you'll want to make a 600 page tome at some point, but doing so for your first project is a questionable idea at best.

Playtest early, playtest often. You don't want to make a full game and only then start testing things, getting a test for individual components as they come up is a good idea. This also feeds into the design, where designing without that basis is more difficult. Eventually you'll want a playtest where you are completely uninvolved, to see how the game holds up with people who don't know it reading the rules, but early on just playing your own game in its early stages (if it's a game with a GM, be the GM) is pretty near necessary.

[Deepbluediver]

Do you read webcomics? Specifically have you heard of one called Leftover Soup?

In it, two of the character have a moment where they discuss designing game systems, and there is some good advice in there. It starts here: http://leftoversoup.com/archive.php?num=325 and runs for about a half-dozen strips.

More impressively, the comic's artist wrote up the ENTIRE SYSTEM and posted it online for people to read and use: http://leftoversoup.com/AndTheBeatGoesOn.pdf

If you're going to be designing your own system, I'd heavily advise you to give it all a read. Anything I could say is probably said there, better, except possibly for emphasizing the aspect of play-testing it all.

[CrazyCrab]

Thanks for all the suggestions! I went over all of them and I've learned some pretty useful stuff.

I've had a go at what I had in mind and so far I'm liking what I've come up with... only an early draft, of course, but it seems pretty solid.

Any tips on where I can place my drafts? I would love to hear some feedback on them (especially the mechanics, as, even though I can playtest them, you still need fellow creators analysis to see the holes in them).

[Deepbluediver]

Any tips on where I can place my drafts? I would love to hear some feedback on them (especially the mechanics, as, even though I can playtest them, you still need fellow creators analysis to see the holes in them).
For stuff that's longer than is easily format-able on the forum, a public Google-doc is best (just make sure the public can view-but-not-edit it).

[Shimeran]

It's been a while since I worked out a full system. I do recommend writing up a sample of play as well as listing the goals before you begin. That might be less critical for projects with fast turn arounds, but I found my ideas shifting as over time, so having exactly what you're going for gives you a good reference point. Not that you can't change focus later, but having a record like that helps you know when you're drifting so you can pull back into focus more easily.

[unglitteringold]

I can only speak to what worked for me!
I have been working on Legends of Cascadon for about 6 years now, and it's gone through a few different beta tests. (I'll be debuting the official LoC system at Gen Con this year!)

My first step was to construct the place, then I picked a base set of rules and what I wanted to add. From there is was a matter of trying things out and seeing what does and doesn't work by play testing with friends.

For me, I treated the world building aspects like I was learning about the world. I wanted to know the origin story, what kingdoms are in it, and what makes the kingdoms different? Who is at war with who? Why? How are the governments structured? What is the source of magic? Who has magic? How is it viewed by others?

Much of that world building helped me narrow down the kind of system I needed to build around it to support the world.

*EDIT* Also, I recommend www.obsidianportal.com for organizing your ideas. It allows you to make a wiki for your world/system and an adventure log you can use as a journal, etc.

[NichG]

I think 'get a group of players, run a serious campaign with them in the system, iterate and refine' is really the main thing. Theorycraft is all well and good, but there's nothing like seeing how players react to the mechanics and how their understanding develops. In video game terms, its like how watching someone figure out your system without actually stepping in and explaining things is painful but incredibly useful.

Even better - get someone else to DM your system and you play. If they forget to use some of the mechanics, do things wrong, make weird rulings, etc, this will all tell you what stuff you've managed to capture in the rules text and what stuff only makes sense because of things that are still just in your head and are not on paper. That's a big difference from video game design.