I find that introducing small amounts of randomness in damage or other affects is useful in turn-based or other slow-playing games to prevent certainty of any particular outcome. If you keep the range of randomness small, "bad luck" is unlikely to actually decide any particular situation. The player with the larger or better army will still win - but by how much? WIth what exact losses? Over time, the small variations snowball into larger amounts of uncertainty, keeping thing exciting.
Strategy Game - Unit Damage
When you're fighting that strong boss and you're low on health and items, you so intently watch how much damage you're going to deal - you wish for a high damage.
The amount of damage you deal in a situation like this makes you feel one of several ways: satisfied, frustrated, amused, incredulous etc. In any case, it's something that made you get involved in the game, get immersed. Damage-variance is responsible for this, for this "roulette" flavor of entertainment.
Like others have said, using damage-variance doesn't greatly affect the outcome of the battle or take away any strategies a player might have: a stronger unit will always deal more damage than a weaker unit, even within a damage-variance model.
You address this matter with the concept of base damage, or minimum damage; The lowest damage a unit can possibly deal in an attack. What is determined by chance is the 'bonus damage', a value ranging from zero to a defined amount that is added to the base damage.
If you make a graphic of the base damages and bonus damages in your game, you'll see that these ranges follow the "skill" of each unit.
Skilled units have higher base damages than inexperienced units.
To make things more interesting you can play around with variances: the more experienced a unit is, the more "reliable" it becomes - it deals a lot of base damage and has little variance. Inexperienced units deal less base damage, but have a bigger variance.
This can also be implemented with attack types: different attacks have different base damages and variances.
Mind you can still make some types of attack like magical or weapon-based still be deterministic, even within this damage-variance model. An attack type with a fixed amount of damage is a weapon with 'zero bonus damage' (zero variance), dealing only base damage.
It depends on the game.
- Randomized Damage: This tends to work better for games that are designed to be quick, casual, and/or have little to no death penalty. In addition, Randomized Damage can work well (as someone said before) in situations where a player has ample time, resources, and options to re-evaluate his/her strategy in light of some good or bad rolls.
- Deterministic Damage: Tends to work better for ultra competitive games (like starcraft), and games where the player has enough on the line that losing on a failed die roll will piss him/her off enough that he/she might not want to keep playing.
In both cases, its important to balance appropriately. A poorly balanced game with randomized damage can come off as completely random such that player choices have no value. Conversely, deterministic damage can end up boring because everything ends up working out the same way every time. As the game designer its your job to address those concerns in some alternate way.
EDIT: I should add... all good damage models have variance... its just a matter of where you put it and how much direct input the player has on it. In starcraft, the variance comes into play in terms of damage types and the ability of a player to micro-manage his units such that they perform optimally. When two armies clash in starcraft, you can have radically different outcomes based on positioning and micromanagement from one battle to the next. That is not to say that system is better or worse its a matter of preference, and the type of game you're trying to create.
What I'm saying is there's always a way to make something random not as random.
For example, in my current game, accuracy% is actually not random.
The way it works is it will readjust itself based on the shots fired in the encounter:
Here would be a sequence assuming the unit has a 70% accuracy:
Hit (100%) 1/1 (too high, next is a miss)
Miss (50%) 1/2 (too low, next is a hit)
Hit (66%) 2/3 (too low, next is a hit)
Hit (75%) 3/4 (too high, next is a miss)
Miss (60%) 3/5 (too low, next is a hit)
Hit (66%) 4/6 (too low, next is a hit)
Hit (71%) 5/7 (too high, next is a miss)
Miss (63%) 5/8 (too low, next is a hit)
Hit (66%) 6/9 (too low, next is a hit)
Hit (70%) 7/10 (pitch perfect, above 50% accuracy, therefore, start with a hit)
Hit (73%) 8/11 (too high, next is a miss)
Thus, you'd be able to expect this unit to behave exactly like this everytime its a new encounter. This works quite well in my game because fights are triggered at the end of a turn, and last for a set number of rounds. As you can see, accuracy here would play a role, just not a random one. And it would be hard to say that a 10 damage 70% accuracy unit is worst than a 9 damage 80% accuracy until checking the exact firing sequence, and the opposing unit's hp (in some cases, it could kill it more quickly, or later).
Essentially, there's probably a way for you to do the same with your damage: keeping a range doesn't mean it needs to be random.
You could have a 8-11 dmg attack that always cycles through its loop (8-11-9-10) or start each loop with a random number from the list, and then proceed with the others. A lot of RPGs actually do this so that the player doesn't feel overly cheated by bad streaks, or gets away too easily with a bossfight by a sudden streak of luck. The upside here is that you insure that the odds of chances are diminished: the player will receive a short-term streak that is within the normal curve.
If you DO want to go with randomness, you need to have a good reason for it. Playing with the surprise must thrump the strategic aspect of your combat. I'd say that, more often than not, its better to include an external factor (morale for example, that could affect accuracy as per listed above) which adds an out-of-combat element you need to take care of to acquire an in-combat advantage (keeping troops well fed will give you a better chance in the fight).
In warfare, there isn't much luck. There's things you did not account for (the sunlight reflecting on a nearby rock giving you a tough time to target your opponent from this angle) but none of this should be entirely unpredictable (well perhaps aside from a sudden current of wind, but this really only affects long-ranged rifle shots).
An idea: Have random damage, but pick from a constrained set of values instead of from a range. A unit with 20 base damage would pick from [20, 20, 10, 30]; deal regular damage half the time, while having a chance to do bonus or reduced damage.
Another idea: Have random damage, but also add gameplay elements that lets you affect the randomness. The units in the King's Bounty remake has random damage, but you also have spells that makes a unit always do its maximum or minimum damage.
If you go with fixed damage, you can still add variance through other effects. Terrain can be a big factor: the old standby where being in mountains makes you take less damage than being on grasslands.
I think it’s fine to have some amount of randomness as long as the player won’t feel like the random values are actually changing the outcome of the game. Ideally, your random calculations should have a low variance and a large sample size over the course of any battle, so they will average out to relatively constant values and the player won’t be able to blame a win/loss on high variation in the random damage. Generally, you don’t want the outcome of the game to come down to a small number of battles that are decided with a few rolls of unweighted dice where the result of the battle can very possibly be drastically different from the expected result.
I have to wonder though, if the randomness is used such that you wouldn't notice it, is it really adding anything? Is doing a range of 20-26 damage much different from doing a flat 23 damage?
I can understand the arguments for/against both types, but I vastly prefer random. I'm coming from a background of Iron Crown Enterprise's Rolemaster RPG system though, so.... I've always liked the idea that a lvl 1 nub with standard weapons/armor could take on and potentially get a critical hit on a vastly superior foe and take them out (i.e., a 'you strike foe through the eye killing him instantly' kind of thing). It also preserves a lot of tension (and yeah, presents a huge potential source of frustration) when your high level unit can be taken out by vastly inferior units. I think it also makes for a much more exciting and memorable game when there is at least some degree of randomness present. Otherwise no matter how many more strategic/tactical elements you introduce into the game you're still going to have something akin to chess.
I've always wondered about systems like Rolemaster. Since there's the possibility for anyone to be taken out by a nasty blow, what tactics does the game allow you to take to prevent and escape such scenarios? Or are you at the mercy of the all mighty Random Number Gods? ;)
I imagine such systems would out more for roleplaying and storytelling than for strategy.
I have to wonder though, if the randomness is used such that you wouldn't notice it, is it really adding anything? Is doing a range of 20-26 damage much different from doing a flat 23 damage?
I think what I mean to say is that randomness can be something you notice on the short term, but over the full course of the game, it definitely shouldn’t be something that you can point to and say “I only lost because of bad luck.” Civilization for example is a game where randomness can play a big part in small battles, but since the games can last for many hours and include thousands of individual battles, the result should ultimately feel fair to the player.
I also think there’s ways that randomness itself can add to the experience, even without causing a noticeable amount of uncertainty. For example, in the game I’m making, all projectile weapons have a random targeting vector within a base accuracy range, which makes them less likely to hit units at max range or physically smaller units. (the counter system is largely based on accuracy, rate of fire, and unit size) The same effect could have been accomplished by simply decreasing a flat damage depending on the size or distance to the target, but with the random vectors you also get the emergent effect of discouraging clumped up units, since clumping tends to negate the penalty for low accuracy; even if it misses the intended target, a projectile will probably hit one of the units next to it.
The function of random damage itself, I think is mainly in either adding some uncertainty to the result of battles where none would otherwise exist (especially in a turn based game) or to get the player to pay closer attention to individual units in a slow paced tactical game. (e.g. the way you have to react to critical hits in an RPG with extra healing rather than simply saying something like “the boss does 50 dps and my potions heal 500 hp, so I’ll just click a healing pot once every 10 seconds.”)
I have to wonder though, if the randomness is used such that you wouldn't notice it, is it really adding anything? Is doing a range of 20-26 damage much different from doing a flat 23 damage?
It very well could, depending on the game. Imagine A small-unit count tactical game, where each unit actually means something to the battle. A player is retreating his units, the rear of which has 22 health. This unit takes a shot before it can get under cover/out of range.
Does it die?
In the flat deterministic system (23 damage per shot) it does. In the more randomized system there is the chance that it does not. Provided that the game itself rewards players who respond well to such situations, it could be a very fortunate thing for a player to salvage one more unit out of a battle.
Really, it all depends on the rest of the game. A bit of variability can provide short-term advantages or disadvantages, thus giving players the chance to capitalize on the former or attempt to mitigate the affects of the latter. Thus, a dash of randomness can reward a player's ability to adjust on the fly. Conversely, a completely deterministic system may be more rewarding to those who can plan far in advance.
I think if you're going to have a range of damage it should represent a few types of damage and not just the luck but the visually obvious variables that alter "luck" the types of damage could be: hit, graze, puncture, armor impact, knock-back and miss. The range of damage could represent each of these and use visual indication of each type of damage. As the character attacks, the attack could also show off the type of damage its going to do. The miss could have the attacker turn his head and blindly attack while the puncture attack would clearly represent a slow controlled attack with thoughtful intent with each one in between having a clear indication of what is going to occur. As for the damage types 0-5 damage could be 0 miss (distract), 1 armor impact (weakens armor maybe), 2 graze (slow for X time), 3 knock-back, 4 hit and 5 could puncture (possibly hitting another enemy). If you aren't going to take the time to make the range of damage visible and meaningful to the gameplay then just do fixed damage.
A good point, randomness can regulate the amount of useful planning: if every turn is so random that "anything" is possible the game is boring, if randomness is so small that the whole battle can be planned reliably the game is boring, but there must be at least one sweet spot with the right amount of randomness to plan in detail at a small scale, plan how to remain open to contingency at a medium scale, and trust the central limit theorem at the largest scale.
Thus, a dash of randomness can reward a player's ability to adjust on the fly. Conversely, a completely deterministic system may be more rewarding to those who can plan far in advance.
A good point, randomness can regulate the amount of useful planning: if every turn is so random that "anything" is possible the game is boring, if randomness is so small that the whole battle can be planned reliably the game is boring, but there must be at least one sweet spot with the right amount of randomness to plan in detail at a small scale, plan how to remain open to contingency at a medium scale, and trust the central limit theorem at the largest scale.
Depends on what do you want for your game. Is the combat a core element or is it not?
If the combat is not a core element - just leave the fixed amount of damage, the player should already have enough optimizations/problems/goals in the game to concentrate on.
On the other hand if the combat is a core element, go with the varying damage. Some explanation is following.
First of all - Random usually means a line. Aka - the chances you will get 40% of the damage are the same as the chances to get 60% of the damage. This sucks.
*Curves* - Use curves - they give random a whole new aspect, take per say normal distribution ( http://www.gmlscripts.com/script/gauss ) - it implies that most of the attacks would do moderate damage, but in the minority of cases they will do very little or a lot of damage respectively.
You can use all kinds of curves to do your goal - want a lighting spell to have insanely high damage, but most of the time just throws some sparks at the enemy - create a curve for it ( exponent per se ).
You can go even step further and add armor curves. That adds another layer - your spell might have a cool normal distribution for its damage, but if the enemy has the same one - you won't do much. On the other hand the exponent spell will deal hefty of damage on the normally distributed armor curve ( 0.9-1 for the exponent are very high, but 0.9-1 for a normal distribution are below average - apply that on an archer vs pikeman scenario ).
In summary - it all depends on the goal of your game.
If you want the player to carefully pick combat units that will overpower the enemy because of the way they attack and to work on the finding the most optimal unit configuration to get the most optimal damage and survivability against and depending on the opponent - curves are the way to go.
If you want the player to carefully optimize the process of his expansion, economics and technological advancement - pick the fixed damage, he already has his hands full with that.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
