# How do Action RPGs make different weapon types feel unique?

I'm getting feedback on my game which has Action RPG elements (think Diablo, Torchlight, etc.) but you control a team of heroes. When you kill enemies sometimes new weapons will drop. In my game the DPS of the weapon will be randomized but stronger the further you get. Each weapon type has a hard coded range and rate of attack, but it differs between types. As an example, every Pistol will attack once per second and every Machine Gun will attack 5 times per second.

A common complaint I'm getting is:

The weapons needs more variety, there's not really many reasons to change weapons because they all have the same speed and small damage differences.

I'm struggling to achieve this. I can vary speed and range for each type, and this seems like an "ok" solution but it's not good enough. This just feels like a sliding scale that looks like this:

range <-----|-----------------> rate of attack
dps: 10


Here you can change the range and the rate of attack, but the DPS is still going to be based on the level of the enemies. In the end, it's still not going to matter which weapon you choose because they'll all have about the same DPS.

If I make it so they don't have the same DPS, then yeah, there will be reasons to change weapons, but it'll be because some weapons suck and some weapons are good. I don't think this puts me in a better place.

I want the player to feel like weapon type X has pros and cons over weapon type Y. How can I achieve this?

• Slight aside: even when the dps looks the same, there can be significant differences due to overkill. Compare a weapon that deals 5 damage twice a second to one that deals 10 damage once a second. When fighting enemies with 10 hp they are equal; but if the enemies have 15 hp, the faster one is 1.5 times better than the slow weapon. – Odalrick Jul 1 '14 at 11:12
• It sounds like you have no reload/cooldown in game, it is very useful for introducing diversity. For example, one hero can wield powerful plasma cannon to nuke enemies while rest of team covers him during reload with high rof, low alpha, guns. From gamer point of view I would want weapon system similar to Crimsonland in such game. – PTwr Jul 1 '14 at 12:30
• I really enjoy the variety of weapons in Earth Defence Force 2025. Rate-of-Fire, Range, Damage, auto-fire or not, sight scope, blast radius, targeting (3-way fire etc), protection when firing, arcing (hit enemies behind cover)... these are just some of the ways the weapons vary. – VictorySaber Jul 2 '14 at 14:34
• Side-note: self-adjusting enemies based on your level makes leveling utterly useless. Same goes for the other way around. Please rethink that aspect of your game. – o0'. Jul 6 '14 at 19:33
• BTW the answer is "do what Quake did". Quite easy. – o0'. Jul 6 '14 at 19:35

By giving them actual pros and cons.

Damage and attack speed are one way to provide a pro/con relationship -- high damage/slow rate of fire versus low damage/high rate of fire -- for example. But that relationship can be a mathematical no-op if the resulting DPS is the same. If you put weapons out of alignment, such that the DPS is not always the same, then as you said you have an obviously better weapon.

So, consider adding further attributes or effects to weapon types, or to the combat system in general. For example, if your game supports the idea of "% chance to do X on hit" effects, that can allow you to drop the overall DPS of "fast" weapons, since "fast" weapons would make up for that by affording more opportunities for the "% chance" roll to trigger. Thus, the player has a choice between a weapon that offers a higher pure DPS or one that has a lower DPS but a higher overall chance of triggering on-hit effects.

You can also apply certain kinds of effects exclusively to certain kinds of attacks. For example, in Skyrim, each flavor of magic (fire, ice, lightning) has a unique side-effect. Fire magic causes additional damage over time via burning, whereas cold magic slows the target and lightning magic reduces the target's own mana pool. This provides a choice for players, even though the overall DPS (ignoring resistances) may be the same: they get to choose a style that complements how they want to play, either a brute force approach via fire damage or a more subtle one via cold spells.

Consider allowing certain kinds of skills only with certain kinds of weapons; relate those skills to the nature of the weapon in real-life. Daggers, for example, could feasibly be thrown and might have a few range-related skills (but you're not likely to be hurling your giant two-handed sword around; perhaps it would have instead skills for rendering foes immobile by damaging or destroying their limbs).

And then there's damage type you could model: piercing versus slashing versus bludgeoning damage. In D&D, certain creatures could only be hurt with certain kinds of damage. While that might be too draconian for modern games, you could incorporate some amount of that flavor.

What you really need to do is force players to make an interesting choice; games are all about interesting choices.

• Weapons can also have side-effects to the character wielding them. For example, some weapons may have substantial recoil which would be no problem when the player is standing on firm ground, but could make things difficult when firing while standing on a tightrope. – supercat Jun 30 '14 at 23:01
• This. The tension is between too much tactical variety, which leads to everyone keeping a golf bag of weapons about their person because they cannot know what might be useful, and not enough tactical variety, which leads to everyone immediately discarding all but their highest-DPS weapon since the others are strictly worse in all curcumstances. – user44630 Jul 1 '14 at 2:01
• +1's all around. Another interesting aspect to recoil if you make it move the shooter backwards: if your target is stationary has a longer range than you, recoil is bad (shoot, knocked back, move forward = less dps). If your target is moving and has a shorter range than you, recoil is excellent. Reverse the situation for a gun that can knock-back the target. – Patrick M Jul 1 '14 at 19:36
• Another very common weapon variation is area-of-effect (AOE). A grenade launcher may fire slower and deal less damage than a sniper rifle, but it can hit several enemies at once and can be much more useful in certain situations. – Nathan Jul 1 '14 at 20:30
• Another thing you could add is "rarity" modifiers. A lot of ARPG items have rarities (common, rare, epic, legendary etc). The rarer the item, the more likely it is to have multiple secondary effects, as mentioned above. This creates interesting decisions: "This lvl 10 common has higher base DPS, but my old lvl 7 epic has all these cool effects, which one is better?" There will come a point where no number of effects can compete with a straight stat increase, but rare items should be viable against items a few levels higher than them, the rarer, the longer it stays viable. – CyanAngel Jul 2 '14 at 10:30

## A problem that all game designers face

Games need to engage their players.

In general, a game can be broken down into two dimensions. Depth, and required knowledge. There is generally a positive relationship between depth and engagement with the players, however depth is usually accompanied by additional required knowledge. There is a mostly negative relationship between burden of knowledge and player engagement and retention.

Requiring that the player learn some things in order to play your game is fine, however things can get out of hand quite fast. Just look at the genre of fighting games. One must learn a large number of button combos just to be considered competent at the game. It can get quite overwhelming for a new player.

Depth is often provided by adding things to the game. Adding mechanics, adding equipment, and adding hazards to the environment are all examples of methods for adding depth to a game. In any scenario where two options are not identical (such as a non-symmetrical battleground, or different weapons), a lot of care must be put into the balance of the game. More options means a more dire need for balance.

Not only is balance necessary to make PvP a fair and fun experience for all players involved, but it's also necessary to make a single player game a fun and engaging experience for the player. If you can progress in a game by simply mashing a single key, or if you can use any equipment without regard for its stats, then that aspect of the game is not engaging at all.

## How does this apply to your game?

Your players have submitted complaints about bland and unvaried weapons in your game, which is a symptom of the lack of depth. I gave the prior exposition so that you might have a better idea of how to fix this issue.

The best way to figure out why your game is lacking depth, and how to fix it, is to look at other games in the genre. See what that game has that engages their players as well as it does. Try to extract the essence of it so that when you apply the same concepts to your own game, it doesn't seem like a rip-off.

## Let's look at an example

I'm aware that your game uses guns as weapons, but let's take a look at From Soft's Dark Souls series as the concepts applied there are valid for almost any game to some extent. Each weapon falls into a general category that more or less dictates the style of weapon. There are swords, greatswords, katanas, clubs, giant clubs, bows, crossbows, and more. Each major category dictates how the user wields and uses the weapon by changing the animations for each weapon. This makes each weapon category feel significantly different.

However, it gets more deep than that. Individual weapons within the same category might use a varied version of the category's standard move set. In Dark Souls 1, the Uchigatana and Iaito have almost identical stats, but they have different movesets. Even though the Iaito has two or three less damage than the Uchigatana, some prefer the different moveset.

The other answers touched on this, but further distinction is made between weapons by factors beyond raw DPS and attack speed. In Dark Souls, weapons can apply status effects such as bleed or poison. While most katanas have very low DPS when compared to other weapons, the Uchigatana builds up your opponents bleed meter. When the meter fills (which happens in 4-6 hits), the opponent loses a large chunk of health. Bleed still builds up even if your opponent blocks the hit too. This makes the Uchigatana very useful against shield turtles. The main reason why I use this example is that you can make weapons that require specific counter-play. For example, you could make low damage bullets or projectiles that ignore or remove armor from their targets. Even though they have objectively low DPS, if your opponent builds armor then these penetrating bullets will have higher DPS than normal bullets. To make matters even more extreme, you could make bullets that not only ignore armor, but deal more damage based on how much armor your opponent has. Just food for thought.

Additionally, there are multiple methods of striking your opponent in Dark Souls. You can hit people with basic attacks like you would expect, however there are also things like backstabs and ripostes. These are considered to be "critical" attacks, in which the damage you deal is multiplied by your weapon's critical modifier. A dagger has very low base damage, but the critical multiplier is so high that you can oftentimes kill people in a single critical with a dagger.

Now obviously, your game employs the use of firearms. You can find very similar qualities in the weapons found in the Unreal Tournament series. Outside of the basic machine gun, all of the weapons are perfectly viable for use in a fight. Each weapon fills a specific niche, which may make you better off with a given weapon depending on the situation.

## What does this mean for your game?

There's a large number of dimensions that you can manipulate to change the feeling of different firearms. You can change the firing speed and mode of course, but you can also change things like how the weapon affects your player when fired, such as recoil and knock back. You can also modify the projectiles too. Not only can you change the damage they do, but you can also make the projectiles do special things. You can make slow projectiles, projectiles that are affected by the environment (wind, gravity), projectiles that you can charge (like the Unreal Tournament bio-rifle), projectiles that explode on contact (like a rocket launcher), or projectiles that erupt into many smaller projectiles upon landing (the Unreal Tournament flak cannon and bio-rifle again). There are plenty more that you can play with, but these are the first ones to come to mind.

## Watch out

A very difficult problem with building depth of choice into your gameplay is that you risk creating first order optimal strategies (FOOS). Put simply, a FOOS is an easy to execute, highly effective, and difficult to counter technique or style of play. Dark Souls is a prime example of what issues arise when severe imbalances in the equipment system lead to FOOSs. Since Dark Souls is predominantly single player, it's not a severe issue. However in a multiplayer setting, FOOSs can kill a game. The only real solution to the problem of FOOS in a deep gameplay architecture is extensive play testing. Simple number crunching can't match the ingenuity of players.

## Conclusion

The answer to the complaint of "these choices aren't unique enough" is usually to make the choices you offer more unique. Look at other games that are similar to the one you're working on and see how they provide depth to their gameplay. If your current toolset is insufficient, you might want to consider improving the capabilities of your toolset.

If you can balance your gameplay by merely running your different weapons through DPS calculators and making sure they are close to equal, then you are likely doing something wrong. You aren't providing room for more skilled players to flourish by providing more difficult or risky weapons that have higher damage potential.

Look at the channel Extra Credits on Youtube. They have a lot of useful information about this and many other topics relating to game design.

• most online games just require you to equipe items of next level, leaving absolutely no thinking to the player. => 0 depth, but they have quite success. :( why? – CoffeDeveloper Jul 2 '14 at 19:32
• While the equipment system in games like Vindictus and Tera add almost no depth whatsoever to the game, they still manage to add depth to the game through other means. A very good set of gameplay mechanics (such as combat) can take a game really far where depth is concerned. Even though equipment selection is braindead, players have fun with other aspects of the game. Or, the game is just a glorified Skinner box that abuses psychology to get players to play more, rather than making it an actually engaging experience. It's one of the two. – Kaslai Jul 3 '14 at 12:48

There are many factors that can differ between weapons. In your case, with guns and and sci-fi elements, here are a few possibilities:

• Area of effect. A grenade launcher, flamethrower, or machine gun can threaten multiple opponents. This isn't a good thing in enclosed spaces or where friendlies are crowded in with enemies, of course.
• Strength required. Heavy weapons may not be an option for your small spy character (though StarCraft Ghosts are genetically engineered and much stronger than they look).
• Accuracy. High precision is obviously critical at long range, while most games operate at spitting distance. Still, you might have super-tech aiming enhancement, like the smart guns in Aliens vs Predator, which is useful at medium range.
• Agility. One reason that SWAT teams and special forces use SMGs is because they're compact and can quickly swivel to face an attacker. A large heavy rifle takes a fraction of a second to shift, which may be too long.
• Concealability. When infiltrating an enemy compound, only very small weapons may be possible. Perhaps the players have to scavenge guns from their enemies, or fight to an armory or airdropped cache; forcing players to get by without their first choice of weapons for a while will make them appreciate them more!
• Electronic warfare. In a futuristic setting, maybe there's a gun which fires small rockets which track their targets, but it's ineffective against high-tech enemies with jammers. An ordinary slug-thrower won't be affected.
• Space tolerance. With small modifications, ordinary guns can operate in vacuum, but it's hard to use a gun with a trigger if your spacesuit has mittens or thick gloves. Low-recoil lasers or rocket pistols may be best in zero gravity, though a smart fighter will be behind cover and therefore probably braced. An armor-piercing gun might make holes in your space ship (or the cooling lines for colony LV-426's fusion reactor).
• Target interactions. Maybe electric weapons are most effective against robots, but microwave guns just scatter sparks; low-powered lasers can easily blind animals but most human soldiers have protective eye covering; plasma weapons are intimidating against rabble but too short-ranged to be effective against disciplined troops; only gauss rifles can pierce the heavy hide of a really big beast, but other weapons can still be useful if carefully aimed at the eyes.
• Repairability. Fancy guns need a machine shop, at least; a simple bow can be replaced anywhere there's springy wood. If your players operate in the field, far from home base, this may be important.
• Availability. If the players might lose their weapons, perhaps ditching them in space because they're about to be boarded by the police, the player who prefers hum-drum pistols which you can buy in any shady spaceport will be at an advantage over the player who really only mastered one obscure weapon.
• Style. Ideally, all of the above should mostly net out, in the end, such that I can choose to use a flashy machinegun, a stealthy pistol, or whatever, and be roughly equal to other characters. Let your players make a choice which expresses who they want to be, without it crippling them, and they'll "buy in" to your world more.

Others have already mentioned magazine capacity, reload speed, ease of acquiring ammo, shotgun spread, procs (DoT, stun), and armor-piercing.

The game Borderlands tried to stir up the usual level-based guns by having different brands with random modifications - Bob's guns tend to have high base damage, Shur-Fine guns have larger magazines, etc. - though IMHO the differences tended to be minimal such that one almost always traded up when a higher-level or higher-color weapon came along.

For an extreme example, games like Worms and Battle Block Theater have dozens of weapons which are each completely different. Exploding frogs, Ming vases, grenades, cluster bombs, and fireballs, why not? Obviously that works best with a comedic or over-the-top space fantasy setting. If your game is meant to be relatively realistic, this may not fly - but it's worth perusing for inspiration.

• A fantastically written first answer. Good job. – Tom 'Blue' Piddock Jul 1 '14 at 13:54

Beyond statistical differences, consider modifying the actual use of the different weapon types to better reflect some different playstyles. The "Feel" of a set of choices will frequently be completely tangential to the actual stats the weapons bring to the table, but can result in a much greater sense of variety.

Even within choices with the exact same DPS there's a big difference between something like an axe that sweeps in a wide arc and a spear with longer range but a narrow thrusting path. The former could be useful for dealing with fast enemies that may try to flank you while the later could allow you to carefully keep an enemy at bay (out of their own attack range). Similarly you could differentiate your pistol from your machine gun by things like weapon spread and whether you need to keep hitting the fire button for every shoot or can hold it down (not to mention the possibility of reload cycles, even if you have unlimited ammo available for simplicity)

Note that obviously these kinds of changes take far more coding work than tweaking the numbers, but if you can give a few points of variety here you can mix them with the numbers changes to make a multiplicative amount of available different weapons.

So first, I would say don't have all pistols have the same RoF etc, there probably isn't any need to keep that consistent. Have them vary slightly around an average for that weapon type.

But in terms of parameters you need to make a shotgun different from an uzi and from a pistol and a sniper rifle etc.

The main two parameters are obviously RoF and Damage, which give you theoretical dps.

But then you have the factors which limit your dps: Range, bullets/clip, time to insert new clip, accuracy. You could use some sort of critical hit system and give sniper rifles high crit chance.

Then you have the factors that don't directly affect damage - scarcity of ammo, weight, size, attachments (grenades? Light?).

Then you have factors based on specific enemies. Thick skin/armour? Reduce damage/shot by a constant amount, which will affect high RoF weapons low damage more than high damage low RoF weapons. A large number of weak enemies will be hard to kill with a sniper rifle simply due to the overkill damage the characters will deal - just make sure that your 'weak' enemies have less health than your high damage low RoF guns do per shot.

Then you also have some special factors; Shotgun damage should decrease with range, whereas other weapons should not. Some machine guns could get hot, so they have great RoF for a short time and then reduce.

Based on the initial complaints you mentioned, about a lack of variance and your subsequent assumption that a slider takes care of the variance you think would work, I've just got a few things to suggest that might help, without being able to draw on a thesis of game design which really wouldnt apply here and tbh sounds like a bunch of unnessesary chaff.

TL;DR; - Stop restricting your weapon design to just DPS and add enough diversity to make it more of what feels right to a player. Explosions, flashing lights, dancing monkey.

Basically, you have a slider moving from range to rate of fire. You're missing a modifying bend on that slider, where a high rate of fire doesnt necessarily have the accuracy desired, as others have suggested an area of effect could come into play. In most games of the genre I've played, I'll usually trade off rate of fire and damage to get a medium kill zone, so while I'm not threatended the weapon isnt useless when in close quarters. Usually means I'm a mage, but the occasional weaponsmith comes along sometimes (thanks torchlight) which makes guns a valid option. And shotguns are just pistols for those who suck at aiming.

So your slider secretly needs to invoke a usefulness rule. Long range slow firing weapons suck when ambushed, and high rate of fire weapons generally suck when taking out a sniper or grenadier.

People will always have their favourites, you arent going to change that. However, adding twists into the mix such as specific weapon classes/models required to access bonuses as has been mentioned previously, you can create the desire for your array. Like picking up a rifle extension that makes a shotgun long range capable.

Another thing to consider is having modifiers for weapons, strip out the concept of 70 different pre fab weapons giving nods to mp4's and ak's, and instead have like 5 or 6 weapons that they can mod the hell out of. Have ammo/battery pickups be more of a straight ammunition increment, and make the player do their own interesting things. have different mags that modify the ammunition properties, say using this green mag I add acid splash to my shots, or this red mag could have molten metal in it. Like, the guns could modify the vanilla ammo picked up to suit. you can still then restrict the pickups at that stage too, so shotgun/pistol/machinegun/sniper ammo, and then the players control what happens next.

Further, by using acid splash you might be adding aoe but its useless against say, energy shields or maybe when using molten metal ammo the capacity in a mag is reduced. A certain battery pack ( if it was an energy weapon ) could provide a stronger shot, but have a reduced capacity of energy before you need to pick up kmore of the vanilla input. You have so many avenues here I admit I dont envy the choices you have to make. When I sat down and started trying to make games the immediate flood of ideas about what I wanted to make was overwhelming, even trying to lock down a genre was hard let alone a subgenre.

To me, futuristic sci fi allows a range from a pistol which can have heavy slugs, dual weild ( basically double dps ) and energy based projectiles; shotties allow again the use of slugs but have a lot more fore to them. and reduced mag. Machine guns have increased mag sizes but are generally lower accuracy and of course firing rate can eat through ammunition which one might not be able to spare. And then rifles ( I still dont see the point in anything other than fps or the old "commandos"/"cannon fodder" games of old where you got a little reticle as a mouse cursor.

I think focussing on DPS alone is just going to be boring, yes players want to get the job done but if you can get an emotional aspect to come into the mix then its less of a calculated puzzle and more of a fun romp killing and looting.

So while your slider is now more of a graphic equaliser patch panel with logarythmic curves and assorted motorised dollies associated to it, I hope that some of my ramblings might have provided some ideas you could apply to your game to hush the moaners. Oh and there will ALWAYS be moaners.

As an example, one game I'm designing is a space game. In that game, players can choose to have their ship made of materials which are particularly reflective or absorbent of electromagnetic radiation. In a simple game, there would be a clear progression from better to worse. In my game though, I made one element in which you want as much radiation reflected as possible (enemy lasers are EM emissions) and I made another element where you want as much radiation absorbed as possible (radar relies on your ship reflecting enemy EM pulses). So now it's no longer clear what the "best" state is, but it is clear that any decision you make will have meaningful results. The choice the player makes here might not make them the strongest, but it will change the kind of play style that is optimal for them. In that way it changes the game itself. If a player has a super-reflective hull, they might get tired of being completely incapable of hiding so they might even get excited to find a "normal" one. It's less about finding "better" things and more about finding different things or finding things that enable a game mechanic to emerge.

You should incorporate that level of pros and cons to weapons in your games. Maybe a longbow has great range and power, but is ineffective inside a certain range or in windy areas (like open fields). So now a player has to think, "Am I going to be going in open fields? Are my enemies going to charge too close too fast?" Maybe a two-handed sword is very powerful but less effective in long fights due to the fatigue caused by its weight. So now the player thinks, "Will I be in short quick fights or long epic battles?" In this way choosing a weapon is less about getting stronger and more about gradually polarizing your optimal play style.

I just want to reinforce the idea of allowing certain skills to work just on an specific weapon type.

For instance take Vagrant Story and its Break Arts. Break Arts are skills which are tied to each specific weapon type (swords, daggers, axes, etc.), and are obtained as you use any weapon of that type. That way, Break Arts also play a role while deciding how to gear yourself.

Another idea that comes to mind is how weapons interact with other pieces of gear. Do you have the concept of one handed vs two handed weapons? Can characters dual wield one handers? Can certain types of weapon combine with specific pieces of gear? Most probably you already implemented such a system, but I mention it just in case.

Also, does your game include the concept of character class? If so, you can limit certain character classes to equip only certain type of weapons. That way, the overall diversity will feel higher, even tough players will have less alternatives for each character.

• Only one weapon per character. There are character classes, and I've already limited certain character classes to certain weapons. – Daniel Kaplan Jul 3 '14 at 18:43

Not sure if this works in your game, but you could add accuracy/spread as an attribute to your guns: The farther you are removed from your target, the higher the penalty to accuracy becomes.
The spread of a gun modifies this penalty (i.e. low spread weapons like rifles work effectively on long distance, while high spread weapons like shotguns are almost useless at long distance.)
I would advice to calculate the hit-chance for each independent bullet, since that would allow you to occasionally hit an opponent from a distance with a high spread weapon, at the cost of lower damage. (A shotgun fires multiple pellets per round - say 7 pellets - and only X of those hit the opponent if you fire from a distance, X being lower the farther you are removed from your opponent.)

By adding or subtracting predefined attributes to individual items in the game that determine their usefulness. Most RPGs now let characters craft their own weapons and come up with personalized weaponry to a limited extent. The ability to craft ones own weapons can sometimes lead to hacks and unsupported actions by the players who find these "glitches" in the crafting process.

• Hello and welcome to GDSE. While your answer provides good information, there are high scored answers here, so you should try to expand your idea with examples and more details about your ideas. It's hard to get attention and upvotes with "one line" answers. – Katu Jul 5 '14 at 10:49