Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Deon]

I would like some feedback from the community. There are many people understanding what a "good game" is. I am starting a new project as a hobby, and I would like some insights and ideas from the members of my beloved Bay12.

Short introduction:
I do not develop games as a profession. I am used to making mods for various games, and gaming was always a hobby. I think that I also enjoyed modding games more than actually playing them. A while back I learned about Unity and how useful it can be when you want to pick up an engine for your game. I always had problems with graphical representation of my games, they were mostly text/ASCII based attempts before.

At first I practiced with basic games like Arcanoid, Tetris and the like. Then I immediately tried to make a complex RPG/arcade game, it took me a few attempts and I dabbled with Pathfinding, AI and inventory management. After a while I realised that without a strong game design background and experience in making games any game I develop is going to be a crap or at least inferior to thousands of games out there. That's why I decided to focus on a game with simple game mechanics and interesting feedback loop ("fun") instead. That's where the idea of the current game was born.


====== Theme ======

What is it about:
I want it to be a single player player-versus-environment game. I also want to have a rewarding playing loop, so people actually enjoy the game pace rather than consider it grindy/overly boring. The idea: have it play like a turn-based board/card game, with story elements to make the gameplay diversified.

Theme:
Anti-utopian futuristic city where each person is suppressed by the government. A big influence: LCS (Liberal Crime Squad). I don't want to get into US politics because I am not educated enough on the topic to make a believable environment, so l decided to make it a "futuristic scifi antiutopia", like some episodes of Black Mirror.

The player would play a role of an insurgent character with actions that can influence the environment. The companions would be "cards" which you can recruit which would give you extra resource-tokens to deal with the environment.

The city would consist of multiple layers, think of a multi-story futuristic city with government/elite on top and garbage/industry/outcasts in the bottom. You start on a residental level and only have access to actions there. You have to open new safehouses on different levels, which opens access to actions there.

Win/Lose conditions:
The game would feature a "perma death" of sort, but it would be not detrimental to the play. When your character is killed, your previous actions leave an impact on the city.

When you "win" the game, you lose: there's a huge federal raid which arrests all the members of your team, and rewards you with points for your actions. You can use these points to influence the state of the next "cycle", making replayability a tool. If you win and your main founder is arrested, you can rescue him from a difficult "prison" area in the next game to get bonuses for your playthrough.

There should also be a global "win" condition between multiple playthroughs, which you would influence by multiple games, and which would have different endings based on your choices in the previous games.

Level design:
There would be the main playing field: the city. Each level would have separate difficulty/security ratings which you would want to approach to be able to operate there.
If you lose multiple challenges on one level, you lose a safehouse there. You can reopen it after, but it becomes harder.

The current city plan is the following:
City: 1st level Town hall/police/courthouse; 2nd level: Media center/hospital/Education; 3rd level: Apartments/leisure/Park; 4th level: Industrial zone/Utility/Fire station; 5th level: Machinery/power station; 6th level: Garbage dump/slums.
Outskirts: Prison, military base.

Agency:
There should also be a game system which reacts to your actions and makes the game harder to prevent a snowball mechanic, which would make the game too easy.
When you become very successful, it means that you also become a pain in the arse for the city, and a special gameplay AI element "agency" should take counter-actions and make your life harder.

====== Gameplay ======

Player loop:
   1. Event phase (random events affecting the state of the city).
   2. Opportunity phase (take actions on the city map).
   3. Feedback phase (result of your actions).
   4. Deployment phase (Try to counterweigh the result of feedback phase and prepare for the next event phase).

Hitpoints - "secrecy":
There will be a single "secrecy" meter. When you fail challenges or cause a big impact on the environment, it should lower the secrecy meter.
Establishing safehouses in different levels increases secrecy. When the secrecy drops below special thresholds you lose safehouses.
When it reaches 0, it's a game over - you are raided and are arrested.
There's another utility to secrecy: when it's lower, your are more notorious and are more likely to succeed in some challenges. That should help you to achieve goals in an "end game" scenario when you are about to lose.

When you lose your secrecy, it is due to certain factors: social disclosure/tech meddling or hacking/rogue activity/display of force. The game will remember HOW you lost your secrecy.
When the game is finished (for this character), the next game situation will be affected on how you lost your secrecy. If you were all gun blazing, the next game will make it harder for you to display force. If you hacked the agency computers, the next game will deploy increased tech security measures. If you persuaded the populace, the next city mayor will employ prominent media personalities to make influencing opinion harder. It should somewhat force you into different play tactics, or provide interesting challenge ("can you win a game three times in a row with an aggressive approach?").

City state gauges:
Like in LCS, the city is supposed to have specific min/max gauges which start on max ("controlled by state") and can be lowered through your actions.
Based on their state, you will have different options in companions/equipment/actions possible. I.e. you are unlikely to meet a journalist if the "speech" is fully state-controlled. And you will more likely meet a burglar or a gangster if the crime is not state-controlled enough.

Resources/companions:
Recruiting members into your cause will give you four resource-tokens: social, tech, force and rogue.
Based on the situation in the city, you can find different member classes to join. Each class would provide you with a certain number of "tokens"

Challenges:
The challenges would be chance-based, depending on their difficulty. To avoid frustration and total randomness in the game, you would use the resource-tokens described above to raise your chances of success in specific situations.

For example, to counter a government speaker on a public meeting, you would want to have a lot of "social" tokens. The original state of the challenge is "difficult". The game would show you 0% + your leader's skill % as a chance to succeed, and three empty green token spaces. If you allocate three "social" tokens into them, they would add 25% chance to succeed each, bringing it up to 75% + your leader's skill as a chance to succeed. Please note that the numbers are very arbitrary now.

On the other hand, allocating tokens of specific types will drop your secrecy accordingly.
As I said above, you may want to balance or pick specific secrecy loss type, to influence your next playthrough. Also the feedback phase would depend on the types of "secrecy loss" during the current turn.

Some challenges could be improved by different tokens at once: stealing an agency van could be improved by 2 force, 1 rogue and 1 tech tokens.
You could choose to spend only specific types of tokens to drop your secrecy accordingly.

It will also be a valid gameplay decision to try to reinforce your main character's skills, fail some challenges, but avoid dropping secrecy.


TL/DR. This is a raw concept. As you can imagine, I haven't started coding anything but basic classes for the game in C# yet. For a good reason too: before I would have a fun idea, start coding it into a game, make art, but in the end realise that systems don't work together or there's not much of a gameplay. This time I want to take a careful approach and actually "design" the idea before implementation.

[Gornova]

Well done! I strongly advise to keep scope small, be focused and produce ASAP a playable demo. See system working is better way to design them, even if require a lot of time!
I like cards and permadeath solutions, seems to be a game that I'd like to play !
I suggest to focus on "spy" a little bit on design on similar games: for example Xcom (old and new ones), for increasing difficulty and events and also hearthstone/magic for cards mechanics and resources management.
One key point of player vs environment game, but also many games, is explain to player how an action interact into environment, one example of good solution here is Spinmortality  (there is a free demo) with nice graphic and clean UI and a incredible focus on UX too!

For game design I've found incredibly useful and easy to design using suggestion from Designing Games: A Guide to Engineering Experiences, by Tynan Sylvester., I'm talking about this book and some ideas here

And also take a look to Europa Universalis IV series: a mix of random events and scripted events, something similar to your concept

I agreen on not start to code right now, do a little research, but don't wait too much! Motivation an momentum are really powerful allies

PTW of course

[Deon]

Thank you Gornova.

Right now I am struggling somewhat with the idea of the "Location visit" event.

That's what I currently have: you arrive in the area, you can do something on the spot, or go deeper. In the park it's just down the path to the center of it. In specific districts it would be into specific buildings.

That's the design:


The framework is already in place, so that's not difficult :



Now the hardest part: create the "event generated" in a NOT retarded manner. Preferrably with content loadable from XML files, so I could have "triggers" and "requirements" already pre-defined, and then just make XML events for that system. That would make the game both moddable and easily fixable if I make typos/want to change something, without recompiling the game.

On the bright side, I figured out how to make "game preferences" saved in a special PlayerPrefs.ini file. Now before i make it overlycomplicated, I gotta figure out how to serialize/deserialize a save file, so you don't have to sit this game through in one go...

[Deon]

Another hard choice in front of me, besides of the implementation of the above (which technically just takes time and testing to avoid bugs in XML loading/class references), is to figure out if I want the combat to be also Roll-based, or create a separate screen where your characters actually fight the generated opponent, like in old JRPG, our actions vs their actions, hmmm... Any opinions on this?

[Deon]

I just realized how amateurish and probably bad the idea above would look. I mean, there's definitely no need to have "next phase" button for every phase. I would rather have a "approve" button after an action screen, and each screen could be different to add life and variety to the game. And I would have to make a shit ton of good art to make "event popups" for EVERY action be interesting.

1) Event phase: TV news screen/weather report/top down view of the city.
2) Opportunity phase: a nice screen with up/down arrows which would switch you between levels and show you extra information (mockup without extra information):

3) Feedback - again, either news screen about enemy activity, or short cutscenes (which would probably be a no because people would get tired of the same cutscenes), or a "report" window on a computer/tablet showing the result of your actions.
4) Deployment - your rebel base with "positions"/jobs to allocate people to.

[Deon]

So after some hard thinking, I decided that "roll to hit/dodge" system without visual representation will likely be sad.
I would rather have a party vs party combat scene, where you take turns to attack/cover/try to escape/use skill situation. It will take me a while to make the system proper, but the idea and the code is there...

[Gornova]

Hi again!
Well the main loop seems to be clear, I wonder if the player has an avatar to identify or the player is the avatar themself ?

For the event generator, I like the approach to separate data from logic, but be aware that will require a lot of effort if you change the way you store data, for example: add/remove something from an encounter, will force you to revisit all encounters. In his blog on Six ages development ( http://sixages.blogspot.com/2015/07/unit-testing.html ), the author explain how he used unit testing and the following posts also automated testing for scenarios. If you take this approach (with some mechanics done) you could be sure that past encounters will not broken!

And about generation in general: do you think to design all encounters, or go for some kind of random generation ?

For fights.. I don't like fight scenes: I mean, in a game like yours I appreciate how my actions will influence game world (you said that will be a player-versus-environment, right ?), instead of repetitive fights scenes. And because I'm also involved in hobby development.. this will require a lot of work, graphics and testing and in the end players will not play the game because of your fights mechanics, or I'm missing the point here ?

For player feedback (something I've to work with Infiltration), I suggest to put more focus on when an enemy do something. An example that I have in mind is Heartstone. You play some cards/some events are triggered and as player you understand immediately what is going on and WHERE. Also some big moves (play a legendary cards, destroy everything) will have special animations, because special actions need full attention by the player. If you search for some heartstone videos, you will get immediately what I'm saying.

[Deon]

You have very good points, sir.

For the character - that will be the main party member, which can get arrested/killed. Other party members are useful assets, but also cannon fodder.

For the fight scene, just animation of what's happening in numbers may be a better idea.
I am afraid to take too much control from the player though. Like, it would be nice to decide when to bail out etc. But I do agree that if the fights are repetitive, it will make it boring.

I will take it in mind for sure, thanks a lot!