I was writing this for a couple of friends and realized it might make for a good blog post. Enjoy. Apologies on the poor formatting, this was cut and paste from a series of emails.
For a mechanic to be meaningful ( and meaningful is a spectrum, so I am defining meaningful as ‘strongly meaningful’ in this case), it needs to be:
- Situationally useful
- High payoff in the right situation
- The situation where it’s useful cannot be automatic
- The decision to use the tool in the situation should come at a cost that makes you weigh the situation
- Making the sub optimal decision should have a consequence
As an example of a non-meaningful decision:
Some Attack: 100 dmg
Holy Attack: 100 dmg + 200 dmg to undead
In all situations, using Holy attack is the correct choice.
As an example of a slightly more meaningful, but not meaningful enough decision:
Some Attack: 100 dmg
Holy Attack: 1 dmg + 200 dmg to undead
In all situations except undead, Some Attack is the right choice and Holy Attack is useless.
Improved slightly more, but still failing:
Some Attack: 100 dmg
Holy Attack: 50 dmg + stuns undead for 10 seconds.
Holy Attack is always the wrong choice vs non-undead. Some attack is always the right call vs anything else. The decision is made for you before the fight begins.
But at least there’s a scenario now, (the target is undead and already stunned) where ‘Some Attack’ is now a better choice. Stun, Some, Some Some, Stun again.
So let’s level up this scenario even further:
Some Attack: 100 dmg
Holy Attack: 50 dmg + stuns any target for 4 seconds. (Twice as effective against Undead)
This is now in the space of interesting. It’s not fully there, but its close enough to almost pass our meters.
Oddly enough, it fails because *now* the best decision is to *always* use this ability on the undead. Both abilities do 100 damage vs undead. There is no reason to ever use Some Attack in an undead fight.
So let’s level up this scenario again.
Some Attack: 100 dmg
Holy Attack: 50 dmg + stuns any target for 4 seconds. 30 second cooldown. (Twice as effective vs Undead)
Now, we are in the space where it looks like a meaningful decision. There’s still a flaw, but we’ll get to that.
Your decision space is now:
- Want to do max damage? Spam Some Attack
- Want to stop an enemy? Holy Attack
- Do I have two targets to choose from that threaten me equally and one is undead? Prefer the undead target if using Holy Attack.
But there’s a flaw. The tension isn’t high enough. If you *really* want to just kill the Undead, you should spam Some Attack. If you want to Stun the undead, you’d be using Holy Attack anyways. The +100% effectiveness is a red herring that makes you *feel* like it’s a better decision.
But since you are getting 100 dmg out of both scenarios, its only creating ability cadence (alternate between pattern Holy, Some, Some, Some, Some, Some, Some, Holy… repeating)
Now that’s a valuable thing we’ve achieved. Very important to MMOs and long-play combat sequences (raid bosses). However, we can do better.
Let’s wrap this up.
Some Attack: 100 dmg, no cooldown
Holy Attack: 150 dmg + stuns any target for 4 seconds. (+100% vs Undead) (30 second cooldown)
Because the cooldown exists, the improved play pattern (2, 1, 1, 1) exists and is now an optimal pattern for standard play. The ability fits into normal combat cycles – AND includes the Undead bonus pattern. Furthermore, a strong tension now exists:
“There is an Undead Target and a Non-Undead Target. I want to stun the Non-Undead… but I want to burn the Undead target down faster… what should I do?”
… and there, you now have a real decision that we, as developers, cannot answer for you. This is a decision you will make for yourself and the game will either reward or punish you in the way you decided the combat should go.
and THAT is the kind of decisions we want players to make.
Thank you for attending Ability Creation 101. You can send your Tuition checks to alex@xelnath.com via paypal.
(Bonus: +100% vs any creature class tends to make decisions automatic. It’s not a smart way to build abilities, but its a good example illustrating the ease at which we can fall into making automatic +power options which feel like good decisions, but are actually hollow)
I actually disagree with your final example being the most meaningful. As it does more damage than Some Attack, AND stuns all targets, the right decision in all situations is to use it every time it comes off cooldown. That it is more effective against undead is nice but no longer plays any role in the decision-making process.
In my opinion, the most meaningful example you gave is actually the “Improved slightly more, but still failing” example. It’s situationally useful (stuns undead), has a high payoff in the right situation (the aforementioned 10-second stun), is not automatically better in that situation (using Some Attack instead will kill it faster), the decision to use it comes at a cost (less damage, for a longer fight), and using the sub-optimal decision has a consequence (less damage with no benefit against non-undead). The stun is probably way too powerful at 10 seconds, but drop it to 4 against undead (and no stun at all to non-undead) and it would be fairly well-balanced against the reduced damage.
“Improved slightly more, but still failing:
Some Attack: 100 dmg
Holy Attack: 50 dmg + stuns undead for 10 seconds.
Again, Holy Attack is clearly the right decision.
But at least there’s a scenario now, (the target is already stunned) where ‘Some Attack’ is now a better choice. Stun, Some, Some Some, Stun again.”
I’m sorry, I’m just a little lost on this one. If it ONLY stuns Undead, then why is the “right” choice ALWAYS Holy Attack in this scenario? It wasn’t stated the attack was being used soley against undead for this scenario, and vs. anything-but-undead, it appears Some Attack is now the “right” choice, since it deals more damage and won’t stun anything-but-undead.
Still this was a very good basic primer on ability creation. Well written, and to the point. I would love to see some deeper dives into this topic, perhaps using some real-game examples, if possible. Cheers!
Christopher,
While I understand your perspective, “pushing this button is now always the right choice” – that is not actually true. However, I glossed over explaining why – so let’s explain:
This opens up the beautiful question of ‘What are you optimizing for?’
There is actually a significant cost to pushing the button. You lose access to that stun. This point was probably missed, because this blog post was long and I didn’t draw attention to it:
But for the final scenario, the reason the latter is an improvement is because of the paired situation presented – you have both a monster you want to kill quickly and another monster who has some other dangerous effect you want to stun (to interrupt or prevent).
Therefore, the question is not ‘What does the most damage?’ – in which case your reply would be correct.
Instead, the question is ‘Which is more important to me? Killing target A or stunning target B’.
That creates the tension. IF you choose ‘High DPS on A’ – you invoke the opportunity cost of losing the stun on B. Therefore, you are paying a price (B gets to act freely) in exchange for that extra damage.
By comparison, if you hold off on pushing ‘Holy Attack’ – you can save that stun for the time when it is most needed!
Thanks for calling this out. Let me know if you understand it.
-Alex
CJ – Thank you for pointing out my typo. I inverted my logic. Fixed inside the post!
Hi Alexander Brazie Team,
My name is Anuj Agarwal. I’m Founder of Feedspot.
I would like to personally congratulate you as your blogAlexander Brazie has been selected by our panelist as one of the Top 50 Game Development Blogs on the web.
http://blog.feedspot.com/game_development_blogs/
I personally give you a high-five and want to thank you for your contribution to this world. This is the most comprehensive list of Top 50 Game Development Blogs on the internet and I’m honored to have you as part of this!
Also, you have the honor of displaying the badge on your blog.
Best,
Anuj
“Ability cadence” is not important for an MMO. It might be important to keep players awake, but it still ends up becoming automatic.
The final example is a worse example of an interesting decision, and is only superior in “ability cadence”. In most situations, there is no decision to be made.
Compare Fire Blast in WoW for mages in classic WoW. A mage had their main ranged nuke; the alternate nuke that wasn’t improved by talents; an expensive, channeled arcane spell; and Fire Blast.
Often, a mob would be killed entirely with the main nuke, opening at maximum range. Cases where you used other spells:
1) specialization wasn’t too high, so arcane missiles did respectable dps and damage-per-mana. Finish with arcane missiles and mana would begin to regen faster due to the five-second rule.
2) you were drinking water frequently and the mob was close. Finish with fire blast to avoid having to kite or use frost nova.
3) maximum dps, such as against multiple mobs. Fire blast increased dps. (At many points in WoW’s life, it probably no longer did.) Also allowed for 1.5 seconds of kiting.
4) Clearcasting procs. Depending on the situation, you could finish the current cast, use fireblast or cone of cold, or use arcane missiles if it wouldn’t get interrupted.
Especially on a PvP server, you could kill a mob simply by casting Frostbolt four times in a row, while spending all your time spinning your camera around to check behind you and in all directions. There was no need for the most efficient cast sequence to include multiple abilities for the game to remain engaging.
Thank you, Anuj!
It’s quite flattering to be recognized for this little site.