Stealth Mechanics for tactical turn-based type of game

Hey!
I work on my game for fun during free time. I slowly yet steadily make progress towards very basic prototype. There was a vague feeling of what it should look and feel like. I wanted these core mechanics features to be there:

• Hexagonal grid - I find square field with its diagonal movement too unfair with 1.4142etc times difference with straight one. Or even more unfair if it is restricted
• Turn-based combat - Every school kid beats me in games with reaction required to win
• Stealth elements - As a fan of nuclear strikes from cloak and dark templars StarCraft series I find it awesome if game allows but not forces you to play stealthy, appear and hide unexpectedly, play mind games. Thus obstacles are not only cover from ranged weapons, but a visual obstacle too. Vision is not 360 degree, but 120. Other side of stealth that I'd like to have is hearing sounds. So that you can shoot through walls if noticed movement. Or turn around to put down stalking guy with knife.
• Randomization is limited as much as possible so that people can somewhat be sure about results of actions. Except firearms hit chance given the scatter... I guess. There will be no way for save scumming as it is online PvP arena. But I don't want people to explode once they got critical hit or unexpected melee miss.
• "Fair" ballistics - bullets are really fired and they can damage obstacles and penetrate them going along the trajectory with less damage.
• Destructible environment - Not sure if required but server side logic is almost ready anyway, so I'll give it a try.
• Stats and squad units specialization and play style decisions. Variety of ways you can approach fight or even start game must be greater than none)))

But I'm more a programmer than game designer to be honest You can notice how vague are my key concepts.

So here's the actual look. Few "completely unseen" cells are a bug of ray tracing. For now I have movement and basic shooting somewhat working.


What do you guys think about this whole thing? I want it be fun for real people opposing each other.

Also I can't decide what action model to choose: Fallout-ish or X-Com-ish? I cant get it what would work better here...

Without them it is going to be pretty much classic stuff, nothing special.

 

Want to become a game designer? Here's three steps. First, read this chapter. Second, hold up your hand and say out loud, "I am a game designer'. Third, build, finish, and release a game. While you work on that, here's some random thoughts about your current design:

• Consider building the game in just single player mode. Even without dead reckoning, multiplayer-networking is time consuming.
• I like the hexagon look with the rigid columns - Might be fun if the characters were simpler than the models you have - they could be sphere's with textures, or even column'ish (though I don't think that's a word)
• What's the core gameplay? You description is torn between a variety of game styles, which is one of the reasons for, "how vague are my key concepts". Finding-the-fun is not a random process.
• Consider that you can build MULTIPLE GAMEs. Since this does not have to be your last and ONLY game, you don't need ALL these elements in one game: stealth, turn-based, multiplayer, ballistics, stats, ballistics, etc... is a LOT to balance.
• Consider creating a series of single-player puzzles, for the player to solve, using just a few of your concepts.

There's a lot to work with there and there's a lot of ways you could go with just what you have. Consider simplifying.

Gigi

 

I appreciate your input

• Well, there is already fully functioning multiplayer. To be honest, all core logic is written on SFS2X server side in java, so I'd have more problems rewriting it to single player now)))
• Every model here is just a placeholder and I assume visual design as separate issue. Lets put it aside for now, but thanks for kind words about these stubs
• You are correct, there might be too many styles there for a one man game. I'm tempted to leave turn-based and limited field of view just because they're already there...
• Agreed
• I promise to consider this but I'd really like to stick to PvP as a personal preference

Simplifying is okay. The problem is just exactly as you said: there are MANY ways I could go. Decision is the problem. I guess I have to go and read the article, draw and imagine gameplay until I have crystal clear understanding of what it should be.

 

With all due respect, I disagree with Gigi. While taking small steps help, it doesn't necessarily mean that you have to simplify your game. You can take as much time as you want to add in the details bit by bit, balancing and adjusting as you go.

While simple games are easier to make, do leave out the ideas you want just for the sake of simplicity(Unless you aim to attract gamers that goes for simpler games)

 

This looks pretty doable. Turn based is a great way to go if you are new to multi player. Each client can basically wait for the others move. It's also easy for clients to verify each other to prevent cheating.

I like the idea of stealth in a turn based game. In my head I'm seeing this as a table top miniatures game with a few enhanced features that can't be done on a table. Consider making stats and chance of success totally transparent and visible for each action. Table top miniatures are also worth investigating for rules and mechanics handling.

 

Interesting idea BoredMormon - I'd like to add this could also be a direction for an easier (more) simplistic aesthetic/style. Characters who are supposed to represent static miniatures would be pretty cool imo. Or army men 'type' static characters with the feet platforms. Just a random idea piggy backing off of BoredMormon's initial direction.

 

Thanks. I think there is a lot that can be gained in terms of game design from studying board games. Graphics engines and CPUs are pretty limited in the tabletop domain, meaning a games design is all it had to stand on.

 

You've basically just described the original X-Com games, but with a stealth element added. It's absolutely doable, and certainly not too much to balance.

This is actually similar to how I wanted to do combat in one of my games (currently on the backburner, as I suck at network programming). It's also quite similar to how I'm doing combat in my current JRPG (minus the stealth). You should lend a hand!

 

Thanks again for all your thoughts. There are few things I struggle now with simple combat. For instance, how to make things transparent to the user? Let's take his expectations about chance to hit a target. Basically it is a trivial task to precalculate this. But there is a catch in my case. All calculation logic and real world model is on the server. So getting numbers might not be the solution. Other way would be to draw the approximation of a gun scatter area on the client, that would be pure client-side on one hand. On the other hand it might not be exactly right in some edge cases. It would give player an idea of what the chances are, but not a promise.

Anyway, while I try to make combat better now with all your words taken ito consideration, there's going to be a meta level of the game. If I use this term correctly.

I thought that visual design doesn't matter now. But recently imagining my game in a Noir or "Sin City" style I've got few thoughts about meta game mechanics. What would you say if:

• You are a gang leader and also a business man, so resources spent do matter
• As a leader you assign resources to a certain mission. Resources are usually: people (not personalities, but mostly a set of skills), maybe equipment
• Each time there's a mission that can be achieved by killing opponents or other conditions. These other conditions are gives another chance for stealth to shine
• Also, assigning people to the mission is a balancing act. There's a limit that you can spend, that would be enough for, lets say 3 unskilled characters or single versitale one. OR! you could send less and any spare resources would be converted to some kind of additional revenue to spend in meta game. But you never know what opponent send against you. You know that he might want those spare resources too, though.
• Evil deeds often take place in a dim light or even darkness so that range of sight is much shorter than even pistol fire range.

 

I like it.

Also, you could always just show a percentage chance to hit the intended target.

There was an RPG (can't remember at the moment... Persona?) in which you would choose how much health to take into combat. The less you took, the more Experience you would gain.

 

This would require server request to know percentage. Is it a good idea to trigger server request on each mouse-over-target event?

Ah, everything seems to be invented already) Never played persona, at least I'm glad that this idea is kind of proved.

 

transparency doesn't have to be an exact number though, it can be an approximation. This gives the player the decision to attack or not, based on the percentage.
Chance of success 75%
Chance of critical hit 20%.
etc.

 

Yeah, they'll never know that it wasn't true value ok, agree

 

Fun for dev's - Hate filled rants delivered from players!

Chance of hit - 95%
You miss!
@#$%!

 

Fun fact - Rand() creates undesirable streaks. It can make players mad and give feedback which confuses the link between action and outcome. So, games sometimes replace Rand() with a human-generated, psuedo-random list. The list can be short or long, controlled on the client game or on a server, and persisted as part of the save state, or not. Whichever way you do it, it's likely to result in a better player experience.

For example: 1, 7, 3, 9, 5, 2, 8, 4, 10, 6. As opposed to For (int i = 0; i < 10; i++) calling Rand().

Gigi

 

The client could always generate the percentages, so long as it's the server that does the final calculation.

 

In my opinion, a turn-based game is all 'bout dat strategy. Stealth should be more subterfuge than tactics. I think something along the lines of hiding a unit or blocking the enemy's line of sight would have great strategic value.

 

I agree, but subterfuge is a tactic. A largely underused tactic in video games.

 

Oh, I haven't thought of it that way. In my mind, tactics have always been flanks and whatnot in SC2 while strategy was the long-term hiding your Dark Shrine and getting them up in their business. I guess in a turn-based game, the definitions shift around since the long-term in turn-based games is so much longer.

 

For turn based both clients could run the calculation and compare. As long as the results match you are good to go. No need to involve a server at all.

 

What to do if they don't match? Some of the answers involve things that feel a lot like a server .
Gigi

 

Here's the update BTW. I was too busy with client-server issues to work on mechanics actually, but you may notice covers working for stealth and being destructible.

 

See if you can find a copy of missionforce:cyberstorm as it was a turn based mech game from the mid 90's. It used fog of war dependent on each mechs view range so you could move around the board & detect the enemy then send the heavies in. I know there are plenty of other turn based games but it was hex based as well & the hills etc weren't smoothed, they just stepped up in increments.

You could also paper prototype your stuff to see how it feels before coding it?

 

Thanks for the input, I'll check this game out.

I'm not sure how to paper prototype these kind of games, and it was fun for me to program it anyway. How can I know if it is fun while I can't get real feel of insufficient information and taking risks stepping into the unknown area.? I might be narrow-minded :[

 

It shouldn't be too hard to paper prototype. For all the combat stuff you just need to figure out some simple mechanics (Roll dice to hit whatever, how far can you move, etc).

Then you simply do the stealth thing manually. Rather than having pieces on the board you can see, you just write it down on a piece of paper (the board you can label each hex). Probably including facing. Then you calculate vision and if your guy is spotted you simply mark it onto the board (or you can just say where he is) etc. This could include mid-turn LOS rules or not depending on your design.

As an aside, I believe stealth works best in WEGO format. Simultaneous turn based combat will result in a more fast paced agressive play, whereas IGOUGO the winning strategy will almost certainly be to lie in wait for your opponent to screw up so you can take the first "attack" action against them... or you could make it let you pause mid-turn when you spot something of interest and take a partial action. That'd also work well to reduce that issue.