Wednesday, June 1, 2011

The Combat Minigame

Moving away from Rift, I would like to write a little bit about that combat minigame in general. It is the most important minigame in every popular MMORPG and I am sometimes amazed how little attention is gets in the blogosphere.

In yesterday's post I described my experience with Rift and my warrior. Rift supplies you with many, many abilities of which most are absolutely clear when to use, irrespective of the situation. Many are just on a cooldown and are to be used as soon as they are available again. Knowing exactly what ability to use when is not bad per se. But knowing it irrespective of the situation is .. at least debatable. The situation can be improved (and abused) with extensive use of macros, but this doesn't seem to be the best solution from a game design PoV.

In my opinion, a good combat minigame has at least these characteristics
1) The player focuses on what is happening simulation-wise, not on his user interface.
2) Every action the player performs has a clear, albeit not necessarily instant, impact/feedback. Abilities are iconic, memorable.
3) The player feels like he has a lot of abilities and freedom to choose.
4) The player thinks that he is performing well, although he could perform much better. Combat is easy and not frustrating to learn, but hard to master.

Do you agree? Can you come up with another abstract characteristic? Are there games that you think do this exceptionally well?


  1. I would interactivity in some form or another to the mix. Combat in which both sides just do their thing and whoever deals lethal damage first wins isn't very interesting.

  2. Hi Scrusi,

    I agree, but I think (1) takes care of this. If you need to care about what is happening simulation-wise (who is in your vicinity, what do they do?) that means that you need to react, and thus there is interaction. But perhaps I should emphazise it.

    One problem with interaction I see is the speed of combat. I can understand that players want fast-paced combat, even though I am not convinced that it is optimal.

    But making fast-paced combat focus on interactivity on a ~200ms ping connection and still be fun is very hard.

  3. Seems I missed an "add" in that comment ;)

    Interactivity doesn't necessarily have to include an immediate reaction to every action an opponent takes. (Though that would help to increase the degree of interactivity. But I digress.)

    Taking Starcraft as an example, we can see quite a high degree of interactivity that still works under normal network conditions. Players react to each others strategic decisions on a global level but also to tactical decisions down to the movement of individual units.

    I don't see why an RPG couldn't work the same way given a certain agency of your enemies (if they aren't players themselves.)

    One possible issue with agency in NPCs is that agency takes processing power and simulating somewhat-intelligent enemies in an MMORPG might put a rather large strain on servers.

  4. The problem that I keep running across is that many gamers *want* to be able to faceroll a button in order to win. Some players prefer combat to be a way of comparing stats from the that gear they grind for. Other players want there to be skill involved, but they don't want the execution to be too important, or at least they don't want reflexes to be too important. Other gamers want combat to be as responsive as possible, require timing, precision, and skill through execution.

    Thinking of how to incorporate the spectrum of players, there should be different aspects of the combat minigame. The players who want to be skilled but are not all about physical execution would probably be like tacticians of some sort. The players who want to be completely stat-based should have the option; to be like the mechanist dependent on the equipment, and as such the progression for them would come from making, managing, and maintaining optimal equipment. The players based purely on execution would fight based on ability and responsiveness.

  5. All very good points also I d add another characteristic : "feel and flow" .That is typically such metrics

    - average combat time
    - actions per sec
    - quality of visual feedback and animation fluidity

    Regarding combat mechanics itself I really like that idea with defence/atack pools

    Let say melee would be split in 4 moves and they would act like tempos in fencing. Decision frame for each move would be 1 GCD ( 1 sec) which is fast enough to turn maintain a flow and not be turn based

    Picking right move countering enemy chain would deplete their defence pool.

    For even slower pace of combat I would have for casters similiar system of moves and counter moves , but at scale of 5 GCDs . Would employ such things like resistances buffs/debuffs, mirror and counter casting, precasting ,etc

  6. Thanks Max. I agree with the feel and flow. Problem is that this is almost impossible to design in the abstract. You need to iterate this until it feels right. It is very important - in some way it is the most important aspect of combat as it 'enveloped' all other aspects.

    When it comes to the speed, I would like a system that allows occasional rapid pressing of buttons, but generally favours pressing the right buttons in contrast to any buttons.

    In Wow-like systems it is often more important to not miss a GCD than to perfrom the correct ability.

  7. 3 is trouble. Too much 3 can cause UI problems in 1. Also, too much 3 can cause overlap in abilities, causing problems with 2: abilities may become less memorable if they are interchangeable.

  8. I agree with Kleps a bit on this one.

    How can you have #3, without messing up #1?

    Just as you said in your last post Nils, on your Champion you have so many abilities to use, it's best to just macro them and spam the macro - voiding #2.

    I guess it depends on how many abilities one considers "a lot". I have a feeling that number tops out in the single digits for most players.

  9. Klepsacovic, Masterlooter, did you notice the "feels like" in the sentence. It's written that way for a reason ;).
    Also, it's not unusual that you have contradicting ends in a design process. Of course, every player wants to have many abilities. But then, yeah, too many abilities are no fun at all.

    Having eight abilities can feel like alot if they, and the rest of the game, are well designed. And 20 abilities can feel totally boring if the game is badly designed.

  10. I think it would help for clarity to replace "abilities" in #3 with "options." Options do not have to be abilities (they do in WoW because of the way the game works, but they don't in the abstract).

    I think my question is, what combat systems do you like? And having gathered a few answers to that, what are the challenges in using those types of combat systems in an MMORPG?

    I think that in order to get a system where most people feel like they are doing well but could be doing a lot better, you would almost necessarily end up with a system where most people perform acceptably but the people who are good are very, very good. If you took almost everyone I know and put them in a first person shooter deathmatch against a lone world champion, the champion would easily take us all out. Are you concerned about that kind of power discrepancy appearing?

  11. I would add the following to the list

    * Every action has a negative impact (the prize).

    You should not only think which action would be the correct to use but you should also decide if you could afford it.

    The negative impact could be
    - Short term resource consumption (Mana, Energie, ...)
    - Long term resource consumption (farmed or bought items like Soul Stone or Powders or stuff.)
    - Nagative impact on your character (health loss or a short or long term debuff (exhausted)).

    This would also add the option the the correct next action would be the NOP action.

    You would have to decide between letting your bleed/DoT tick or do another instant damage attack at their cost.

    The removal of the NOP action was what made healing boring in raids.

  12. I think it depends on the type of game. For example, some games take a very simulationist approach and some take a very gamey approach. And in terms of combat speed, some games rely on strategy and some on reflexes.

    So I'd say:

    -If you're making an immersive simulation, then combat should feel natural and the UI should be unobtrusive. Mount and Blade does an excellent job of this.

    -If you're making a strategy game, then combat should involve interesting decisions. Players should have to think about which skill to use rather than using a predefined rotation. (Or they should think about positioning, terrain, etc.). Wizard 101 and Guild Wars aren't too bad, but I haven't seen a really good strategy-oriented MMO yet.

    -If you're making an action game, combat should be fast and reponsive. I'm not an action gamer, so I can't really comment here. ;)

    -And in any game, combat should be viscerally exciting. City of Heroes did a great job of making abilities feel very fun and powerful. My character could fly, turn invisible, throw people through the air, etc., and I was a healer!

  13. Thanks everybody for the insightful comments. :)

  14. I've always been a fan of 'less is more.' The combat mini-game should be comprised of a small but diverse set of abilities- options to be weighed in any scenario. These options should be variants of 'power per second' (PPS), 'power per resource' (PPR) and weighed against burst vs. sustained output. No need to have 30 abilities when 80% of all tasks could be covered with a generic 4.
    Furthermore, I don't think every mini-game should always have the ability to react to a given scenario. Shortcomings in classes, for example, should be allowed in order to force decisions as to the approach. These shortcomings, obviously, should not be the conduit, in and of themselves, to a loss.
    Lastly, abilities with cooldowns vs. spammables. Small cooldowns that aren't multiples of each other (3sec ability vs. 6sec) mean that refreshing each upon availability is not longer the job of a bot. The immediate one coming off cooldown must be weighed against the one that is due just a second later. This piggy-backs the thought of designing abilities that go for PPS/PPR and gives another layer whereby a decision must be made by the player.

    Make a smaller combat minigame to allow for more creative solutions. The key, I believe, is to involve the player, and force decisions.

  15. Two points I'd add to this.

    Too much info is bad:
    Having started in WoW I found LoTRO combat in group content, at least initially, to be more engaging. This is the opposite of many others views I know but in LoTRO there are no DBM or Omen add-ons to smack you in the face with lurid alerts. So tanks have to watch what each mob doing to keep aggro and anyone with an interrupt has to watch constantly for the cast animation on enemies to know when to interrupt. Keep it simple and do not allow players to over-automate it!

    Combos and abilities:
    Regarding 'choice of abilities' I think choice of tactics could suffice. Endless abilities with no real tactical choices isn't necessarily that good. I like the *sound* of what GW2 will be doing with environmental combos - the firewall + multishot = flaming multishot maneuver. We'll have to see if it actually works in practice but this could give group-play an extra layer to the combat minigame beyond hitting abilities on cooldown.

  16. ... er.

    Close? I think the most interesting and exciting combat comes from:

    1) Varied and impactful abilities.

    (Rift and its "here's another way to do exactly the same thing with different flavor text!" abilities is not adding fun by doing so. Everything I do should have a point - WoW sorta got this mostly kinda right, Original EQ's magic classes did it better!)

    2) An 'optimum build' that builds around my skill and chosen playstyle, not theoretical maximums.

    (Rift! I want to play this way, so what do I pick to get there? Best thing about this game. WoW is the worst: 'here's the best talent build, there's only one gameplay style of significance with each tree, and once you're setup, you have two points to do whatever you want!)

    3) For most MMOs: a slower pace but a better strategy game.

    (Give me a game a bit more like 'speed chess' - it's not just about hitting abilities, it's about positioning, terrain, thoughtful use of fewer abilities, preferably with environmental concerns. Planetside wins, hands down. Nothing else really even breaks the horizon - I don't find WoW or Rift terribly challenging. There is nothing really interesting about 'target, push buttons, move a little'.)

    4) Less gear dependency - at least in PVP.

    (PVP should be about skill and strategy, not stuff. Stuff-as-a-gateway-to-PVE works well enough, but that model in PVP is an abjectly painful experience.

    In this, I think GuildWars is the best design: your levelcap is low, your gear is essentially fixed, but designable to fit your playstyle, and the interesting thing is building out abilities that let you play as you want within your initial build.)

    WoT is a damned fun game that has ONE BUTTON to hit (for the most part) for abilities. It fits all of the items above, however.

  17. And offtopic comment:
    Nils it would be really interesting if you play World of Tanks and write what you think about the experiences!

    Even the carebear Tobold liked it.

  18. I put up a quick blog post with a few more thoughts about making abilities more situational. Feel free to comment if you like.

  19. Max, it's not that I wouldn't like playing WoT. I love BGs in MMORPGs and WoT is just a game about BGs. I might like it.

    It's just that tanks aren't really all that faszinating, in my opinion. What I do consider interesting is how much attention the game gets in MMORPG blogs, however.

    And then there's f2p and I hate to constantly think about paying.

    Lastly, please send suggestions for blog posts or anything off topic to: .
    This email is also at the bottom of this blog ;)

  20. I'd like to add a factor that is closely related to your 1) and Max's feel and flow, but not quite the same thing: "drama".

    I want combat to be emotionally enthralling - either the glory of mowing down hundreds of minions kung-fu movie style, or fear of losing to the big boss man.

    Without that, you may as well be playing a turn-based game, which will have more tactical depth.