Lets talk game design (holistically)...

News & General Discussion
1

I’ve noticed in the game design forum, we haven’t actually talked that much about game design as a whole. Factoring in certain decisions can invoke a domino effect and can essentially end a project before it begins, it is quite important…

Game design is actually very technically involved muddled within other descriptions like “Game Architect” for example. Of course game design duties does not include actual implementation from a low level (like programmers, artists), it’s more about how, why and a thousand foot high technical overview of how to throw it all together.

A metaphor might help, if pseudo-code helps a programmer deduce a specific feature then a game designer creates pseudo-code for the entire game.

It’s not all frolicking around the office in a toga shouting listen to the ideas in my head, again designers are (generally) very technically minded with a lot of experience across the board (respect). Especially helpful in some situations where an artist “may” propose to management lets do a “batclaw” mechanic for example, management thinks it looks cool whilst programmers and Q&A staff are nervously twitching in the corner.

As a designer, you’ve got to have the foresight and to explain what will happen to the project if said implementation enters the design document… Although like most projects, it never goes to plan and as a game designer you also have the luxury of having all fingers pointed at you.

So now I’ve set the baseline, how did you “design” your game? How did you assess how to put your game together? Did it work out as planned?

We might find some “gotcha’s” out of this topic, could be interesting…

2

I first started with what I felt was the ultimate goal of the game: to interact in a realistic way with a single NPC (me). Then I built up everything around that.

For a while that was all I had: “talk to the NPC a lot.” It was supported by my ideas for a dynamic dialog system, but I didn’t really have anything for the game itself. Between me wanting to focus on these deep and interactive conversations, and the simple reality that I’m a single person, I needed something that would be very simple. Initially I was thinking walking sim, almost as a de facto genre.

Then I hit upon the idea of a puzzle game. It was perfect, really: puzzle games (know that when I say puzzle game, I mean stuff like Myst or a point n click) have a pretty low-pressure atmosphere, so the player would be abl to relax and interact with the NPC without expecting action. Additionally, I got the idea of having the player and the NPC interact for the puzzles–puzzles would require both of them to work on different things at the same time.

Now the problem with a linear explanation like this is that I didn’t necessarily discover things linearly–the game “design” informed my story design, which informed my game design, etc. So I’m going to jump around a bit.

Anyway, my initial plans were, as I mentioned, simply to have the player and the NPC interact. Little more than that. I needed a story, though. I started to think about story ideas within the framework of that premise–of player and NPC interaction. I don’t want to give away all of my conclusions, in the slim, slim chance I ever release this, but–the player is in a virtual world. They are sent there to interact with the NPC. The two of them solve various puzzles to escape this virtual world.

Now, about this time I was reading about Yoko Taro, a Japanese game director who’s written the stories for several games set in the same universe. A lot of his stuff is really crazy, but his games tend to use the medium of gaming to maximum effect. For instance, in a game called Nier, at the very end the player is given the choice to delete all of their save data–this is at least 50 hours of time, mind you–to save a character in the story and essentially erase the player character from existence.

I enjoy linear, scripted stories, but at the same time I feel that if I’m going to make a game, I want to emphasize player agency as much as possible. And you can’t do that properly without integrating it into the plot. So I did that for my game.

There are 4 endings in my game.

The first ending occurs when the NPC and the player both enter a cavernous and suspicious warehouse. At some point the NPC separates with the player, and begins directing them to take certain actions. If the player follows the NPCs instructions, like one would expect to, they fall into a trap by the NPC, who kills them to escape.

Now, I want to point out that the way the dialog system works serves to increase the impact of the player unwittingly falling for the trap. This is because the dialog is open so that the player can walk away at any time, and if the player doesn’t interact it continues without them speaking. So at all times the game enforces the idea of YOU, the player, being a willing participant in everything that occurs, rather than merely being swept along by others orders.

The second possible ending happens after the first. If the player refuses to obey the NPC, the NPC gets enraged and begins to hunt the player. At this point the player or the NPC can kill the other to escape.

However, around the puzzles the player will have played are scattered various trinkets related to the NPC’s past. If the player gathers them and manages to show them to the NPC, the NPC will fall into a severe depression and surrender themselves to the player’s will. The player can at this point kill the NPC to escape. This is the third ending.

However, at this point there is a final puzzle which the player can undertake. Solving this puzzle allows the player to obtain the necessary McGuffin (let’s call a spade a spade) for both the player and NPC to escape.

My initial plot was built around those 4 endings. I wanted to tell a story, but offer player agency at the same time.

So I wanted to make it a puzzle game, but I really didn’t have much in the way of ideas for puzzles. I was stumped on this for a while. Eventually I started thinking, once again, about what the whole idea was–of the player interacting with me. So I decided to build puzzles based around myself.

Initially I was thinking of puzzles based around things I’m interested in–a puzzle on Chemistry, a puzzle with Biology. Eventually, however, that morphed into something more personal.

The version of me portrayed in the game is ultimately not a good person (made obvious in my second spoiler there). However, it is definitely a potential me that could exist. So I decided that the puzzles should represent my “descent into darkness.” At the same time, I didn’t necessarily want to tell the player what to feel, but I wanted the puzzles to represent certain things the player might feel towards the NPC. I came up with a list of things that might represent both the NPC’s “character development” and how the player interacts with the NPC.

NPC toward self / player relationship with NPC

Introspection / Curiosity
Conflict / Ambivalence
Realization / Acceptance
Self-loathing / Doubt
Penance / Deceit
Sacrifice / Betrayal
Absolution / Trust

Now, they don’t match up exactly. But they generally go together.

As for designing the puzzles, that’s something I’m still working on. I’m doing that by taking the idea of the theme and rolling it around in my head and seeing what things it dredges up which fall in line with the overall arc. For the first one, for example–it’s curiosity and introspection. It isn’t just curiosity about the world around oneself, but curiosity in specific places that lead to the later darkness (Introspection is the darkness creeping in, conflict is the NPC feeling the pull of the darkness without being aware of it, realization is the NPC recognizing the darkness within himself).

To represent this I’ve decided on a simple puzzle where the player needs to place three “darts” or something to completely cover the exterior of a house. They make the house, which is white, corrode to black. At the same time, the NPC is inside the house doing the same thing. Now, everything the darkness touches is corroded, so the player has to stand back to turn it on. However, because the NPC is inside the house, he can’t escape the corrosion.

That’s about as far as I’ve gotten for the design. I’ll need to design the rest of the puzzles of course. I’m thinking that each one will have some immediate, tangible benefit (like being able to jump higher or run faster) that the player needs to use for one of the later endings. I haven’t decided on those yet, though.

Whew, bet you weren’t expecting such a glut of information, huh!

2 Likes
3

@EternalAmbiguity

Nope it’s cool, I mean trying to condense your thoughts and your entire project into what’s essentially a small summary is a task within itself.

Until it’s my turn I’m going to play devil’s advocate, after a quick disclaimer… I understand projects are personal to the creator and poking holes in a project isn’t always fun for developers. On the flip side it’s always good to get a few experienced opinions, companys pay people to poke holes in procedures / projects etc. trying to find ways to improve said project etc.

As long as it’s constructive of course, “it sucks” doesn’t help… If it does happen to help then it could be the difference between actually releasing something (that may be well recieved) or it might not help at all in which it doesn’t really matter… Pinch of salt and all that.

So, basic premise sounds interesting… A few questions, when you say like Myst is it a 2D game? How are you looking to achieve the NPC challenge response? Is it a small event system (like Mass Effect for example) that adds / subtracts dependant on one of the criteria like curiosity based upon what choices said player makes?

Which of course ultimatley selects which ending you recieve. Have you considered your audience? Is there an audience to consider? How long do you believe it will take you?

Coming back to the puzzle bit, is there a way to tie negative / positive outcomes into the puzzle system? I understand that one of the primary concepts is again the challenge / response from the NPC although (without giving too much away) what exactly is the point of the NPC? Is it just to be an antagonist, is it some sort of mirror reflection you (the player) wasn’t aware of?

Does it need a little more time in the design phase to sort of get more answers? How did the prototypes go?

4

3D game, just with puzzles like Myst. The last Myst game (and its spiritual successor Obduction) was a 3D game.

I’m not sure what you mean by “challenge response.” Are you referring to the dialog system itself? If so, it’s simply set up to continue to the next line if the player doesn’t respond.

The basic idea is for the player to talk with the NPC and get to know them. Talk about such random things as chaos theory, or politics, or business. I want to have a significant amount of interaction with the NPC, so the player feels like they know the NPC by the end of the game. That’s really the whole central idea of the game, and everything’s built around that. So I guess that’s the “point” of the NPC. It’s not anything too deep. Of course, as I mentioned the NPC will be me, so I intend to use my own personal knowledge/beliefs/logic in the NPC conversations, so the player can effectively interact with me through the game.

It doesn’t really have a prototype :stuck_out_tongue: It exists in my head and in the little bit that I’ve got in the game so far. Haven’t really thought about an audience either, though I imagine the type of crowd who’s into walking sims and stuff like Firewatch would find it interesting. As for how long it’ll take me–probably longer than it should, but I don’t think it’ll be too long. I’m not planning any complex mechanics, just some simple puzzles and lots and lots of dialog and some semi-scripting of plot points.

To be honest this is totally a hobby, not something I really plan on releasing for sale, so I’m just taking it at its own pace. I definitely welcome your questions though, they’re always helpful in identifying weaknesses.

I suppose you could have positive and negative outcomes with the puzzle, but I hadn’t considered something like that at all actually. Do you think it would be beneficial?

5

@EternalAmbiguity

Sounds good to me, yes I think outcomes of said mechanic could really help remove some linearity… I’m not opposed to linear structure but it’s a fine line between linear because the game demands it and boring…

In a lot of games puzzles seem nothing more than a means to an end, whilst if there were some action that caused an impact it makes you think twice about how you’d deal with said puzzle… Also it could help with replay value, you can base it around your original NPC interactivity system and the outcome variables you’ve already defined.

It just seems somewhat more interesting…

1 Like
6

Thanks! That’s really useful feedback.

I see no one else wants to share :stuck_out_tongue:

1 Like
7

I know right, I’ll have a go next… Might take a while to get it all together, trying to design a game is one thing / trying to design a good game is a whole other matter.

8

@

Well I’ll be happy to comment on anything you present. Can’t promise I’ll say anything insightful, but some feedback is usually better than none.

9

For me personally, a game is an opportunity to tell a story, it’s as simple as that. The story can be solidified (e.g. cutscene heavy), or it can be flexible (created through player behaviour, which I prefer) but in the end the idea is for the player to imagine a world outside of reality.

For me as a player, every game that I enjoy is defined by a moment when I felt that the atmosphere of the game captivated me. This atmosphere is the sum of the messages communicated by the environment, the dialogue, the choices, the reactions to my actions - in short, the sum total of the game world. If that world became very interesting to me at any point, it defined itself to me as a great game.

That is basically what informs my own game design. In fact I start a game the same way that I would start a written story - with an image in my head of a moment captured inside a certain kind of world. It might be a very concrete scene, or it might even just be mainly a particular kind of emotion at the beginning, but it sets the tone of the game and, surprisingly maybe, it helps to inform me as to what mechanics the player should use and how the game should respond to the player actions. Everything the player does should be congruent with the emotion or the atmosphere that drives the game.

That’s also why I don’t believe that it’s always useful as a rule that someone should start with a blocked out prototype. For me, a prototype without an atmosphere quickly becomes lost and full of ambiguous mechanics. So what I usually do is think of whatever object will most define the atmosphere and create it first. In the case of my space game, this was the environment itself. It doesn’t matter if it’s finished or not, the point is to anchor the game in an atmosphere as soon as possible. Then, of course, (maybe after also writing a bit of a background story, which I find useful) the main focus becomes the prototype, but always with a reference to the defined atmosphere.

That was probably the main problem with my current game, when I started. I basically decided to prototype/block out everything first, and I quickly just completely lost any artistic sense of what I wanted to convey or make the player feel. It was very hard for me to work on it without having a ‘story’ in my head that defined it. For various reasons, I couldn’t come up with a decent story and I lost motivation on it for quite a while, until I had made a few ‘game beginnings’ with different stories (and wrote a few actual background stories), which helped to break out that sense from being kind of stuck. For me at least (maybe it’s not the same for many other people) this approach is at the core of creating something naturally for me. We’ll see if it turns out to be useful for producing a fun game.

2 Likes
10

I typically start out with a basic mechanic. A spider spinning silk. A boat floating in water. Ghosts hunting PacMan. Watching tanks fill. And so on.

I’ll then go ahead and prototype the mechanic. I’ll do some basic programmer art as place holders, but most of the focus is on the code.

I’ll then spend a lot of time playing with the mechanic. Is it fun? Is there something for the player to do? Can I see a game here? Many concepts die at this stage.

Then I start to build the atmosphere for the game. This normally consists of a bunch of screenshots mocked up on paper. I’ll start to design rudimentary UI requirements here. As well as input schemes, general menu flow, and the like.

Then I’ll build a more fully fledged version of the game. I’ll get some better quality art. I’ll get the flow through the game going. Most of this just reflects what’s in paper, but frequently on the fly changes will happen as each system gets built and tested.

The next stage of the game design is building the explicit player goals and feedback loop. Basically the things that tell the player how they are playing. So life bars, score systems, high score boards, particle effects, screen shake and so on.

Then there is a round of balancing.

Finally to round it all off there is the tutorial.

2 Likes
11

  • First I ask myself what problem there is to solve, ie basically anything, might start with an impression, an emotion to recapture, etc … Which I expend into a problematic, that put the problem into context, using research.

  • Then I position a solution against that context to expend on interesting idea, generally things I found to be underdeveloped. I start to write a pitch and a treatment, which is really just deciding what direction to take for every aspects, in a way that answer the problematic.

  • During that I research existing solution and feasibility, and found area that might need new solutions to be developed. Once the pitch is done I start working on the parts that need solution as a R&D phases, I also start to accumulate documentation on the existing solution to other part.

  • If I found enough satisfaction, that is a validation of the feasibility, I start to design proper by looking at finding the core structure (generally the progression) and developing things like theme and art direction deeper to final specification.

  • Then I start the breakdown of the structure to planned the works and evaluate the cost, from which I will do further balancing to meet reality.

Generally the r&d phases is the longer one with a lot of prototype that are obscure to anyone but me. It’s only when I fully understand what I really want to do and how (validating the feasibility step) that I think proper work and communication start.

For example

For a long time I wanted to make a planet hopping game, that’s the problem.

  • I look around (research and context) and most game with similar idea follow a very militaristic/science inspired simulation, since these are already there I decided to go to something more arcade and space op, looking at all those science fiction illustrations, like early pulp, which are barely used anymore as an aesthetics, with the idea to modernize them (ie interesting art direction solving to do).

  • I also notice current tropes don’t really talk about the radiation problem in space. I think it’s an opportunity to have an interesting angle to build an universe on. I found this curiosity that is Radiotrophic fungus, a fungus that metabolize radiation like plants do with light and photosynthesis, but using melanin.

  • Digging deeper onto that lead, I learn that earth used to be much more radioactive and most fungus have this dormant ability inside them. Given than human have melanin too immediate connection is made, it solve many problem with long space travel and surviving with little food. But since it’s not a sim but a fantasy space op, I used that for explaining superpower (the glo) because why not!

  • It also introduce gameplay opportunity because radiation is not equal everywhere in the galaxy, the closer to the center, the more intense radiation there is, which give a proper power progression to the game. It also explain why we have so few visit on earth from alien, they stay in high radioactive zone because it’s nutritionally rich lol, the lore and background write itself.

  • Also melanin have already some cultural baggage on earth, I decided to make Africa the future of mankind technological advances with the rest of world faltering behind. WHY? well opportunity to have a background that writes itself with a good dash of mystery! As in it’s very different from what we expect currently, inventing the reason, going from state A to state B, will ensure endless depth.

  • But I will use historical precedent like the rise of the western world as a template! I mean there was a golden age of civilization called the silk road, define by stable trade between massive empire, during this time Europe was in a deep dark age marked by religious fundamentalism and “tribal war”. Which is pretty much akin to current Africa, but then those civilization of the east become too trapped in religious fundamentalism, stagnant infrastructures and faltered while sciences moved from them to the western world who were freeing themselves from the religious domination.

  • Notably the works of Al-Khwarizmi (from which algorithm are named but also why we say Arab number) spur in part the renaissance through the many wave of Muslim invasion. Much like currently neo colonialism is spreading tech toward Africa unequally against the pillaging of their resources. But the western world is starting to see an increase in religious fundamentalism and secular idea, a stagnation of infrastructure which makes difficult for new competitive idea to emerge. Meanwhile Africa is leapfrogging many technology to build non existent infrastructure based on cutting edge idea like bitcoins, drones, mobile, etc …

  • Since we are fantasy and drama is more interesting, the fall of the western world and consequences is the way to go. Also Africa has all the materiel resources already (most big mineral deposit are there) and has significant landmass that cross the equator, ideal for space launch, they also have the biggest genetic diversity, which is good for the necessary research to develop the melanin augmentation! Without the power flowing from the western world, balance of power will be challenge and the continent will be able to radically reinvent itself. It writes itself and is not overcomplicated, all concepts flow from each over.

  • So using the positioning with problematic, I think I’m having an original world for free. I still have to look for research, it turns out planet generation is nothing new and there is wealth of documentation of how to do it. Most tend to point toward quadtree, but I find it unsatisfying for gameplay metric reason. Turns out No man’s sky is released and do stuff we can analyze from glitch.

  • I do a number of prototype to identify the failing point, right now I haven’t solved stitching from a quadsphere, so feasibility isn’t validated yet, but I’m confident it’s possible because it’s been done. I also want local city generation (which have been done but I want to expend to flowing city) that hasn’t been fully tested, and some advance new ide for travelling NPC generated using local generation (I know nothing that is similar so it’s full problem to solve). So the final design and breakdown isn’t done yet.

1 Like
12

Ok, let’s start.

Well, idea guys / gals get a lot of flak but good a good concept idea for a game is hard to come by. Concepts / visions get warped when they come to implementation, it’s somewhat inevitable and in reality time / skill / budget will place restrictions.

So I’ve tried to prototype / or tried to be involved in quite a few types of games at this point, FPS, RPG’s, MMO’s, side scrollers. Mainly to try and understand what I want out of a game, a few concepts actually went past “framework” stage but were scrapped like the rest due to them being rather generic.

I’ll just note it’s not always the large games that cost the most amount of time, besides technical limitations open world games have padding which smaller more intricate games don’t and requires more effort to balance out. So with what I’ve learnt here’s my latest proposal…

The concept:

A first person RPG which isn’t new or significant by itself, the major difference being this time around I won’t be fighting the engine. It may sound a bit defeatist, but it’s nothing more than a circumstance of what your “vision” or story is telling and what compromises you’re willing to make.

The story never really dictated whether it had to be sci-fi or old world, it just specifies it’s on another planet in a galaxy far, far away (wait that sounds familiar). It never even dictated whether said planet and / or scenes had to be full of life or barren, a lot of what I was doing was based upon concepts of other games and that was my first mistake.

Especially in a first person game, RPG mechanics can do a lot in the way of linearity and it’s a necessity… I’m not going to give too much away story wise, because one of my previous concepts I posted years back ended up in a game. It’s a shame really, because my original concept sucked ;)…

I will say it is along the lines of Silent Hill meets Mass Effect / Bioshock that also encounters games like Seiken Densetsu (Secret of Mana). This time around, there ain’t restrictions based on what a game should be… So hopefully I won’t get bored of the concept and dump it half way through.

The technical bit:

So, how to approach it? This time around I’ll probably not need a massive SQL based event management system; any data curve / table import method system will do just nicely. I don’t have to worry too much about foliage (the rendering systems nightmare).

I might even go down the path of static lighting and at that point I’m well and truly covered. Although I’m fully aware I’m trading some issues for another, for e.g. basic terrain and foliage can be set up in a day or so, whereas I’ll have to create a lot of art for a mainly mesh based solution.

A lot of this is down to compromise, if I was to go down the proper terrain path I’d want it to be filled with foliage / have a decent draw distance. Then one thing leads to another and before I know it I’ve got a massive multi-async tiled terrain with tons of custom systems / shaders / rendering techniques to try and shave off the odd 0.5ms here and there.

The other issue is filling the terrain and then having to QA it all, a primarily mesh based game is a lot more work up front but in the long run (as I’ve found out a few times now) it always ends up simpler.

I of course want it to look as good as possible, so decent animation blended IK / blendshapes / ragdoll systems will be needed / lit particles / good looking UI etc. etc… Ultimately I’m looking to remove a couple of years off a basic prototype I have going on which will allow me to spend more time on the important stuff (like gameplay)…

I’ve run into a lot of engine issues in the past, I doubt I’ll have a fraction of the issues encountered in other projects… But just to cover the basis I will develop it in both Unreal and Unity, concepts are the same and it’s not a whole load more effort TBH so why not?

I could go on for a long time here… So I’ll leave this and add more detail if there’s any questions or suggestions.

13

I might have misunderstood the goal of the thread, but I was expecting more talk about the non-technical gamedesign aspects. Like how is the player supposed to feel over the course of the game, and how do you intend to accomplish that? What decisions are there to be made, what kinds of gameplay- and feedback-loops will there be, and how will they relate to each other? How high do you intend the skill ceiling to be, relative to established RPG franchises, and how does that relate to the narrative of the game? How big of a role will combat play in your game and how will you go about making it feel satisfying? You’ve talked a lot about the technical side again, which is always interesting, but I was under the impression this subforum is more suited for discussions in which non-programmers can easily participate.

14

It’s about everything game design (holistically) and as I said originally it usually does include at least a high level topology of all systems, trying to answer loosely such questions as how long is this going to take?

With larger projects you’ll probably have risk assesment managers, technical / artistic / sound / UI leads and other game designers vetting the architectural / design document. When you’re an indie, well? Whatever goes, but I don’t belive it’s a bad idea to understand what impact your design decisions are making; not just at a player interaction level but a resource level also.

As to game vs. player impact, some of the questions you are going to help me understand the point of them, as I understand it (I could be misunderstanding by a country mile) they seem to be somewhat arbitrary.

“How is the player supposed to feel over the course of the game?”

Unless they give feedback you will never know, everyone interprets games differently based on their own preferences. I remember reading the Bioware forums before ME3 came out and from story to graphics all the way to combat systems etc. everyone had their own take on how they “felt” about it and that’s a good thing.

Your job is to correctly articulate the games message without holding their hands too much, also trying to strike that balance so they can somewhat “fill in the gaps” and make it personal to them without the developer imposing what their POV should be.

What decisions are there to be made, what kinds of gameplay- and feedback-loops will there be, and how will they relate to each other?

You’re going to have to be way more specific there, I think as an example there’s a “mechanic” loop where the exo-skeleton / power devices and wep / addon system are all tied into each other. If you run out of power you won’t be charging in like a bull… But of course there will be upgrades.

How high do you intend the skill ceiling to be, relative to established RPG franchises, and how does that relate to the narrative of the game?

As I briefly mentioned about “peaking over the fence” I’ve tried it and honestly I believe it’s a bad idea to relate it to other RPG franchises. Because ultimatley it has to work in the context of your own game and usually it won’t be like for like in any sense.

What I do believe is that you can learn from others mistakes, like most agreed that an MMO style combat system does not work very well in a single player game. Then again you don’t need to be a game designer to understand that a combat system with the depth of a puddle won’t be that interesting, just a bit of common sense.

I’ve seen some games use the skill tree as a narrative “decider”, Pillars of Eternity comes to mind… Whilst I can understand its premise in that type of game. I don’t believe it would of been worth implementing in games such as Mass Effect. It is project dependant…

How big of a role will combat play in your game and how will you go about making it feel satisfying?

In an FPS combat will always play a big role, or you’re just essentially wandering about talking to people. The issue with FPS games is they are somewhat band limited… I mean you can’t really do much with them, I’m going to try and make the system as interactive as possible where different means can cause different results and you may have to “think on your feet”.

Stealth systems whilst fun have been covered one too many times.

“But I was under the impression this subforum is more suited for discussions in which non-programmers can easily participate.”

End of the day, we’re all here to learn… It’s not just learning about what decisions you should make, it’s also about learning what impact these decisions have on your project and they will have a major impact.

Designs can impact what engine you use, what tools you require, how to approach a project, the length of the project and player interaction / fun.

One of the main reasons I made this thread is because I’ve made “design” decisions that have been costly… Where a core mechanic wouldn’t scale right or it just stopped being interesting in a greater context.

2 Likes
15

Hm, I still feel like I don’t “get it”. I intially thought you wanted to talk about some specifics of your next project, put them up for discussion and elaborate on some choices you’ve made, but that doesn’t seem to be the case?
My questions were merely examples of some topics I find interesting. With “relative to other franchises” I don’t mean it in the sense of “copy from”, I mean “shared experiences we have that can be used to talk about things more efficiently”. E.g. Dark Souls and The Division are arguably both RPGs, but imho they live on opposing ends of the skill ceiling spectrum and that’s huge for how the games feel and to what kind of players they appeal.

16

Hehe! I don’t get what you “don’t” get…

Sure I of course would like to share but not just about a pin point laser selection of game design, I want to discuss the whole thing from top to bottom… I’m trying to avoid talking about my project too much as you can learn a lot from others methodology and way of thinking (even if I don’t seem to be paying attention, I do read and process it)…

Here’s the bottom line when it comes to my stuff, as I said I’ve prototyped a fair few projects now but nothing has really grabbed me… I’ve made design decisions that’s had a plethora of ripples, like making openworld games.

I’ve somewhat had enough of fighting the engine also, but I still want to remain firmly down the “RPG” path… I don’t have an issue persay fighting with an engine, it’s just the combined results just equate to another “generic” game (due to time)… So it’s become somewhat obvious that I need to focus more on the game and less on “tech”…

But this will affect design decisions and I’m embarking on something I’m not quite sure what to do with… Like would a more linear / conserved game require a character creation system? Is a TOD a necessity to current “semi open world” games? I mean I’m wondering what I can get away with?

These are important questions from a design perspective, if the player base doesn’t expect certain features then it saves me a lot of headaches. Although I am not willing to compromise on what I want to achieve (which ultimatley is the source of a lot of issues)…

Scale of the game I can tweak, settings / art I can tweak… Again nothing about the story or mechanics requires massive open world forests, I was just doing it because I kind of like it… Although it proved not to be worth the effort, not that I couldn’t of finished it, it’s just the level of quality would of declined.

Biggest question is, what now?

17

I think it’s finally starting to click. I somehow assumed you must be long past the phase where such decisions are still open. I thought you had a super clear vision fleshed out already and the remaining questions are about strengthening that vision through gamedesign.

That’s good advice which I need to take to heart myself.

I think you can get away with a lot as long as it makes sense in the context of the game. In Assassins Creed Unity I could have sworn they don’t have real TOD and just blend between few prebaked light settings, but I could be wrong. Arx Fatalis (really liked that one) was set entirely underground. A setting like that can solve many problems. If you are open to making traversal based abilities a core aspect you could go for a very vertical gameworld where you start at the bottom and with unlocking new abilities will be able to further and further ascend towards the top. Verticality in Shooters generally seems to be viewed rather favorably, although from a strategic point of view starting above and going downwards would make more sense to maintain highground. I’d make as many time saving compromises as you can bear, to free up time for the things you care most about. E.g. I’d go for barren wastelands or underground caves over lush forests any day, if it means you can have higher quality gameplay or character assets. In a first person game I think you can easily skip character customization and protagonist voiceacting (a lot of work for a feature that people ultimately don’t seem to like, if you look at Fallout 4). Players still like to express themselves though, and in FPS style games that’s often done through weapon customization. If you wanna do something with exo-suits, that might even be more interesting to get creative with than weapon customization.

If you don’t want to fight the engine, consider if there is a setting or artstyle you can make work well entirely without light baking. E.g. in space the light/dark contrast can be so harsh that shadow areas are pure black for the most part, and if you don’t have atmosphere and happen to face the sun at a constant angle, you don’t need TOD either.
You could also go for an architectural style that has many repeating basic shapes that can share the same collision meshes and only have variance in the visuals, to make QA easier.

How are you gonna solve balancing? And how broad a spectrum of different character builds do you want to accomodate? That seems like a huge QA/testing concern to me as well.

2 Likes
18

I wonder why this thread fell asleep. I think it’s an interesting topic.

I don’t really have the time or energy right now to work much on my own game, but one of the future questions to consider would be what kind of metagame for a mission based action/tactics game would be the easiest to implement, test, and balance, but still help make the overall experience more compelling.

@ : did you get any further with your planning / exploration?

1 Like
19

Ough, tough discussion as it is always in hindsight and the gotcha’s are often inevitable! The process is about making mistakes or doing it wrong to get it right later on. But, I’ll share you mine.

  1. The first thing I decided was not to follow traditional practices. I needed a simple description of the game that would set the outline of the story:

It is XXXX. XXXXXX years have passed since the XXXX outposts above the planet have been taken down, remembered as the first Interstellar War. For a brief period, mankind is united. The Super Nations begin a joint venture: the XXXXX to protect the people against a possible retaliation little than a century away. But underneath the sky, Teracities, like XXXXXX, deteriorated quickly, shifting the power from municipalities to the megacorps.
Driven by furiosity, XXXXX fights her way through a cruel existence against the megacorps who are responsible for the injustice against her family, friends and the citizens of XXXXXX, unaware that her actions will determine the faith of the entire planet.

My biggest inspiration - how movies are summarized at IMDB. I didn’t go for a full fledged design document or really made the typical protagonist/antagonist distinction at the start. That was a good choice, because the villains developed during the process (I got a genuine evil Lex Luthor now…). My focus was on the scenery and the main character, the two biggest impact makers in the game. Everything revolves around this short piece of text. Notice it doesn’t have a title.

  1. No title, only work titles. Marketing is boss, and the title becomes sometimes unfitting for weird reasons. Sometimes the mechanics change a lot and the title does not reflect the game anymore. I striked the entire space and interstellar stuff. Another one was because a horrible game came out with a similar title. Didn’t want to ride the wave of stinky turds in the ocean. Changing the title was very late-project and made the project easy to adapt to it. It might not look like a “game design” thing, but I have to collaborate with people and I need to name things. Theming in my opinion should be around the scenery and the character.

  2. Research. Lots and lots of research. The devil is in the detail. Asking myself simple questions. Why did this come to existence? Who was responsible for this event? What would be a plausible explanation? The homework always pay off.

  3. Composition and orchestration instead of technical approach.
    Every piece and part in the game should be in my opinion a self-sustained unit. Sometimes, things need to be scrapped, sometimes things enhanced. Control per element allows me to make small adjustments easily or tune them for performance later. They are also helpful as many elements are required outside the game process - marketing like branding posts or worse… steam packs.

  4. Tools, tools and tools. I’ve picked a strategy most people wouldn’t. I’ve avoided as much as possible to use “easy asset store picks”, and had some custom made. One specific tool allows me to design levels like you would play in SimCity. This was possible by doing #4 where I could replace my fancy cubes with detailed models later on. Especially when you want to test navigation and camera angles long before the rest of the things are mashed in. Especially because models and art just takes a lot of time. Another set of tools abide gamification. Doing grunt work to design characters and shops for example are all about repetitive work. I’ve got tools to build entire objects and their icon sets based on ScriptableObjects. This approach is mainly good as I prefer composition; but Unity doesn’t handle nested prefabs as the serialization system is just incapable of doing what you can in Unreal Engine. But, I must say, I prefer Unity developed versus others over any day. Love is a strange thing.

  5. Ecosystem of mechanics.

6A.
Often on Slack, there are discussions about which AI are used and which one is better. I hate those discussions, because they are all inferior as single entities as I can make them collaborate. Using shared resources and designing a technical subsystem as an information bus allows that. I often dumbed down very complex AI to simple AI. Like using FSM for the “sensoric” intelligence (do something based on what you see or hear), utility AI to decide how the action should be executed versus stats. A simple node tree for dialogues to remember choices and change the data for that actor. And something I call a Personality - a set of AnimationCurves (they look fancy in the inspector) to make the actor react and respond in certain environments. A small list that acts as short term memory works in conjunction with having advertising objects for AI planning; it is just plain easier for me to use with composite objects than GOAP (and I bet some people will disagree with me, but I wouldn’t care less).

6B.
I picked fixed controls. Porting is a process of turds and game input is about tasting and touching massive turds. Locking the player at certain points is necessary for cutscenes or adding actions that I do not want the user to control. One major win is that I can quickly tie this technically to record input and replay. I have a large game with a lot of time and choices that matter, so beta testing can be horrible. Sometimes I just want to make an actor run from A to B and kill mobs on the way as soon as possible. Little things I do not want to spend time to generate a build for and let someone play for ten minutes, just to discover bug A, B, C and F to reappear because of crunch time gave me typo’s. I ended up building a test framework and automated replay tool just to save me the time. You might think now this is about development, but it is a win-win. Sometimes I want a scenario to work with a little bit of extra duck sauce. It can be an explosion, or a particle effect that would be excruciating to test every time again when I do a build. Mechanics can and will break, and I don’t want to waste hours every week just because I had an epiphany - that could’ve been a memorable moment but I have to scrap because of that possible breaking.

6C. User interaction requires a good theme. Sometimes the UX and UI look like they are from two worlds. It is better to do vintage right than designer wrong is my personal mantra. Controls need to be known for me in advance and a styling guide is a necessity to keep everything matched.

  1. Actors

I’ve divided Actors (NPC’s) based on their role, but also on their technical capability. With over 200 actors in place, I didn’t want to work out typical behaviour and manage them independently. So, I divided them over heroes, bosses, mobs, funnies and crowds. A big fat Google Sheet that has their id, their name and a one liner description to never exceed an amount of characters. Easier to import, even easier to maintain. With a special tool, Actors are composed based on this information. I had the need for supporting actors to give the scenery some ambience. Integration is king.

  1. Mockups

Everything needs a mock up in my perspective. I don’t need a perfect model or even the correct shape. Using mockups create a visual representation of what you want have envisioned. Why waste 2 hours on copy pasting and sharing PSD files, when you can create a mockup scene? It is incredible handy for me as I have (here again… tools) a mockup scene, where all my mocks are under a group and I’ve locked them using HideFlags. Such a scene can be loaded to an existing scene in Unity, so I know exactly what is done or still needs to be done. You also see a game become slowly alive.

  1. Avoiding discussions and meetings

Ideas are great. But often, like 90% of the time, it will be about grunt work. It will be crunching and crying. I can discuss for hours what can be done. I just make a plan and stick with it. Finish tasks and reflect if it’s acceptable or not. I’ve wasted to much time to get people up to speed personally, so I’ve just written background information. You have a question? Read the docs. You have an idea? Park it and save it for evaluation. Also, it is a two way street. I avoid the questions if “can it be done”, because always a lie is born from that question. My question would be “would it take more than two days?”. Every question I ask must lead to a decision, else it would be a waste of time and tissues. If someone can’t make it work, they have to take ownership for it - make it work. Bite off more you can chew or do something less challenging, like working in Wordpress.

  1. Change is inevitable.

Everything can be scrapped and will be a candidate for scrapping. As a designer, I want that amount of control. Somethings - like 40% - won’t make the release. There are too many factors involved; porting, filesize, gameplay, performance, unfinished. Sometimes I need more than to switch gears. Sometimes I just need to get out of the car and go for a stroll.

  1. Talk with other game designers.

Not so much to learn about their approach, but when hitting those walls. Some have solved or scrapped a mechanic for reasons I didn’t think of. Knowing everything ahead and the “common” pitfalls are lies. I stick with the basics and expect chaos. I take the heat and when the going gets tough, the tough gets going.

  1. Mental state.

Never lose your cool. It is better to ignore someone than to go all Rambo. Planning is easy if you understand that the balance is made up out of two columns - time and feature. If you want more time, scrap features. If you want more features add time. Every feature has a task which I call 30%-nonsense-and-ludacris time, which I think is crucial to get there. And I make no illusions; everyone is always lacking time. There will be no exclusions in that area. I pick 6 hours a week to “revitalise” myself - AFK.

Hope this gives some gotcha’s, but probably won’t be anything near useful for most people as I’m batshit crazy anyway.

3 Likes
20

Yes, although I am somewhat procrastinating at the moment… It’s hard to not just copy a game like fallout, but I think I have a solution.

I’ll be using holotech weps / upgrades centered around the exosuit as a core to add variations of difficulty. As for enemy scaling, personally I think you should use your noggin a bit, so doing generic everyone levels up at the same rate isn’t where I want to go.

Not to say I might not use cohesive minion scaling, but on the whole I will customize paramters for certain enemys. The layout design of the first too levels will be a small stint on a ship followed by a small city cut into four sections. Where the game really kicks off.

As I am a graphics nut I won’t be doing TOD and keep it baked, to save major headaches the scene’s won’t be big enough to justify massive amounts of light changes. Things like character creation suites can come later, it’s only a case of morph targets and sliders anyway.

That’s the core of it…

@samnarain

I don’t think I can give you any advice, it seems you have a solid foundation… But it was a great read, thanks for sharing.

2 Likes