So I've been tossing around ideas/systems in my head for the past few months for an rpg based in the 1920's. As I was thinking about stats, and what they would do, how they would act, etc... skills inevitably began popping into my head. Then this problem....
Lets say as the player advances, they are focusing on being a great marksman(whatever skill/stat combo that may be) eventually the player would max out and be an excellent marksman.
Now we have a situation....how do you keep the player from just sniping off every single NPC in the game? More specifically, without making NPC's Invincible, how do you design a system that allows the player to feel like his work has been worth it, but there is still a chance for misses?
If the player snipes at a guy in the back of the head who is sitting on a bench, the chances of missing should be zero. But this in itself has very big balance issues. Do you incorporate a dodge skill/stat? Well that wouldnt work here, your not dodging something when your not expecting it/cant see it coming.
Do you implement some kind of max on the marksman skill/stats that say limit the chance to hit at 70%?? Thats not very believable though if that 30% miss rolls in when you just took a shot at a non moving person.
I've been tossing this issue in my head since last night, and im not finding anything that I can pinpoint and say, "Yes. thats the fix."
I will say that in designing this game, and prototyping it(im using the Unity Engine) I am trying to keep all systems open to the possibility of multiplayer at some point down the road. So while at this point the problem deals specifically with NPC's, im trying to think of something that would also transfer to other players.
Discuss. I am starving for others' ideas on this type of situation. I did some searching but couldnt find anything.
Dealing with One Hit Kills in RPG's
I would try to make the game feature situations which would benefit different builds. For example, a single guard sitting on a bench would benefit a marksman, while a group of enemies would instead benefit some form of artillery build or whatever. Perhaps there's wind, making it less useful in some places? If the player wants to snipe off every single NPC, why not let him? Give the skills dimnishing returns so that players are less likely to excel in a single area?
Author
I would try to make the game feature situations which would benefit different builds. For example, a single guard sitting on a bench would benefit a marksman, while a group of enemies would instead benefit some form of artillery build or whatever. Perhaps there's wind, making it less useful in some places? If the player wants to snipe off every single NPC, why not let him? Give the skills dimnishing returns so that players are less likely to excel in a single area?
I thought about environmental effects like wind, but that is WAY beyond my programming ability at this point...definitely a feature I'll be revisiting in the future though.
I also thought about the whole, what the hell if they want to kill every npc, let them. Its very much in the attitude that I've been designing, in that I give the player the tools, what they do with them, and how they go about progressing, even what they think of as progressing is completely up to them.
However, in also thinking about possible multiplayer down the road(wayyyy down the road) we have seen in many examples what happens when the developer gives the player the tools and lets them decide how to use them. The majority will use them for mass evil if possible. Even more so because they want to relive the adventures of those that came before them. (I.e. everyone knows about the Mass corp thefts in EVE, everyone knows about the guy who killed Lord British in UO beta, yada yada) And inevitably your game turns into an FPS w/RPG like stats(see Mortal Online, and Darkfall for examples of this)
The diminishing returns is an excellent idea, I'm going to look into that more.
And inevitably your game turns into an FPS w/RPG like stats(see Mortal Online, and Darkfall for examples of this)
Personally I think a user friendly implementation of such a game would be an ideal MMORPG. Shame I've never seen one.
Author
[quote name='eSJay Designs' timestamp='1305820393' post='4813041']And inevitably your game turns into an FPS w/RPG like stats(see Mortal Online, and Darkfall for examples of this)
Personally I think a user friendly implementation of such a game would be an ideal MMORPG. Shame I've never seen one.
[/quote]
I think the GTA series is an example of a great implementation of the RPG/FPS blend. There are multiplayer forks, but I've never played one. I have read about a couple that are pretty deep and very customized.
Lets say as the player advances, they are focusing on being a great marksman(whatever skill/stat combo that may be) eventually the player would max out and be an excellent marksman.
Now we have a situation....how do you keep the player from just sniping off every single NPC in the game? More specifically, without making NPC's Invincible, how do you design a system that allows the player to feel like his work has been worth it, but there is still a chance for misses?
Weren't the guns back then somewhat inaccurate, anyway? Especially at long distances?
I would try to make the game feature situations which would benefit different builds. For example, a single guard sitting on a bench would benefit a marksman, while a group of enemies would instead benefit some form of artillery build or whatever.
I really like this.
If the player wants to snipe off every single NPC, why not let him?
I don't necessarily like this.
The problem with "hey, if that's what they want to do, let them!" is that it takes the design out of game design. Open-ended systems like Elder Scrolls are interesting, but they end up being poorly paced, unbalanced, undirected messes. And this is coming from a guy who is counting down the days until Skyrim comes out
Do games really need more flexibility? Do players really need more control? I think we try so hard to give players a "fully realized world" that we forget that what they want is a game. They want a challenge, they want well-timed rewards. They want to be immersed as part of the world, and that usually means they aren't a god-like entity that can get away with mass genocide just because they figured out how to cheat the A.I.
A common way games get around this is your first quote (situations where different builds have different advantages). Wind is an interesting alternative for marksman specifically. Other possibilities:
'- Some games "shake" the gun to simulate the unsteadyness of holding a real gun. This could also rise based on some sort of stress factor.
- What if the Marksman stat didn't change your hit rating: instead, it changed what guns you could equip? So you see a sense of progress (I can use better weapons) but not a sense of god-like power (each weapon has its own strengths and weaknesses).
- If sniping a sitting target in the back of the head is a perfect hit every time, just don't place as many targets sitting on benches with large open spaces behind them... level design can fix alot of these issues.
- You're prototyping in Unity? Simple wind is easy
Create two variables somewhere: a public float windSpeed that is clamped between 0-1 and has a public Vector3 called windDirection that must be normalized. Then for each rigidbody bullet that you want to affect, just apply AddForce(windSpeed * windDirection * Time.deltaTime) each frame and tune those two variables until it feels right. If your bullets aren't rigidbodies, you will just do the equivalent with whatever physics you are using. The side effect is that players might find it frustrating that the bullet doesn't go where they're aiming, so be careful... but it's worth a try. Maybe adding an on-screen indicator for wind direction and a TrailRenderer to the bullet so the player can see the path it took will help people get used to it.- Have a skill cap: if "maxing your character" is 1500 skill points, only allow the player to reach 750 skill points so he has to choose his proficiencies. If he wants to spend 100 points on Marksmanship, he's going to be very weak in other areas. You can hide something like this quite well, to where the player doesn't even feel like they're being manipulated (see Deus Ex).
- Place targets in groups: you might be able to snipe the guy sitting on the bench, but that's going to alert his sidekick that is pacing on the sidewalk practicing a speech...
"Weren't the guns back then somewhat inaccurate, anyway? Especially at long distances?"
No. The weapons made before and during WW1 were some of the finest sniping weapons ever made. Given a good sight, the British SMLE (for example) will quite easily do first-shot kills out to 1000 yards.
The idea that infantry engagements happened at shorter ranges and that weapons really only needed to reach out 3-400 yards (and hence could use smaller calibre ammunition) is a post-WW2 concept.
No. The weapons made before and during WW1 were some of the finest sniping weapons ever made. Given a good sight, the British SMLE (for example) will quite easily do first-shot kills out to 1000 yards.
The idea that infantry engagements happened at shorter ranges and that weapons really only needed to reach out 3-400 yards (and hence could use smaller calibre ammunition) is a post-WW2 concept.
You don't have to simulate wind perfectly to have a wind effect. You could vary a wind variable and adjust it over time to increase and decrease. You could take this effect and make it's effect on the bullet more pronounced the longer the shot. Also, keep in mind a bullet drops as well. You can't put an enemy square in your cross-hairs when you're a mile away the same you would just 100 yards away and expect to hit said enemy, especially in the head. You can also swivel the barrel a few degrees at a time to simulate breathing and such and just tie in your marksman stat to gradually decrease this effect, but NEVER eliminate it entirely.
My quick two cents on the issue.
My quick two cents on the issue.
I don't have a lot to add to the excellent replies you've already gotten, but I can imagine a couple of other balancing mechanisms.
First, as has been mentioned, reducing the number of stationary targets will help to balance out excellent marksmanship. Hitting a stationary target isn't too complicated, but hitting a moving target is much more so. So if an enemy is moving, it would be more believable that the shooter's ability to hit would be less, leading to fewer one-hit kills. This is easily modelled by adjusting down the player's chance to hit stat against some combination of the target's movement and use of cover, plus factors like wind should you choose to include them.
You could even break this up by adding seperate sub-skills to address moving targets, which would let you both start the player off as a poorer shot at moving targets but also let the player hyper-specialize, if they so choose.
Grouping enemies will also help to lessen the imbalance of good sniping. Sure, you can take one out without much trouble, but then the others will scramble and take cover, not to mention hunt you down. So you can't just blow them all away, but strategizing and employing other skills will still let taking out a single high-value target in one shot a valid approach.
Good level design will include obstructions to view/line of fire, which will affect success a priori to the character's skills. And finally, while I strongly encourage the above suggestion to set up play situations in which other builds will have special advantages, I'll also suggest looking in the other direction as well: situations where taking out one target is especially desirable. It can be a mission objective, or a particular type of enemy that might otherwise be difficult to overcome, or whatever you can think up. But whatever you decide, long range rifles aren't meant to be rapid fire. The player gets few shots relative to other approaches, but the tradeoff is that those shots can be highly accurate and so particularly effective.
Sniping is not generally a strategy that a single person can use to take down a lot of enemies quickly or easily. If the player wants to do that, there should be opportunities to do so, but that doesn't mean that every build (including a sniper) needs to be able to.
First, as has been mentioned, reducing the number of stationary targets will help to balance out excellent marksmanship. Hitting a stationary target isn't too complicated, but hitting a moving target is much more so. So if an enemy is moving, it would be more believable that the shooter's ability to hit would be less, leading to fewer one-hit kills. This is easily modelled by adjusting down the player's chance to hit stat against some combination of the target's movement and use of cover, plus factors like wind should you choose to include them.
You could even break this up by adding seperate sub-skills to address moving targets, which would let you both start the player off as a poorer shot at moving targets but also let the player hyper-specialize, if they so choose.
Grouping enemies will also help to lessen the imbalance of good sniping. Sure, you can take one out without much trouble, but then the others will scramble and take cover, not to mention hunt you down. So you can't just blow them all away, but strategizing and employing other skills will still let taking out a single high-value target in one shot a valid approach.
Good level design will include obstructions to view/line of fire, which will affect success a priori to the character's skills. And finally, while I strongly encourage the above suggestion to set up play situations in which other builds will have special advantages, I'll also suggest looking in the other direction as well: situations where taking out one target is especially desirable. It can be a mission objective, or a particular type of enemy that might otherwise be difficult to overcome, or whatever you can think up. But whatever you decide, long range rifles aren't meant to be rapid fire. The player gets few shots relative to other approaches, but the tradeoff is that those shots can be highly accurate and so particularly effective.
Sniping is not generally a strategy that a single person can use to take down a lot of enemies quickly or easily. If the player wants to do that, there should be opportunities to do so, but that doesn't mean that every build (including a sniper) needs to be able to.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement