Hello. Today marks sorta a new beginning for this blog, since Games Workshop has effectively killed my interest in tabletop wargaming. Thanks GW, you saved me a lot of money! In all seriousness, though, I still enjoy creating content and writing. There are a lot of concepts I would love to share on my YouTube or Twitch but due to needing a lot of explanation or math the best place to share them is in written form. Instead of making a forum post on my guild's public theorycraft forum, I'll post them here. I'll reserve post on the guild forum for finished products and fully fleshed out concepts for the greater community. My plan is to use this blog to spitball ideas and innovate new concepts that might not actually go anywhere. Additionally, I will use it to share my thoughts on community issues that are a bit more longform than would fit a snarky Twitter post.
|Warning possibly bath math ahead|
My impetus for today's post was to deal with the many people who criticize our published DPS numbers on the basis that "DPS numbers calculated on an inert target in a vacuum mean nothing." I find this criticism personally baffling.
For starters, the goal in any PvE encounter is to figure out how to counter or negate the boss mechanics so you can focus as much as possible on DPS. Since your goal is to literally make the fight as close to fighting an inert target in a vacuum as possible it stands to reason that you'd want to know what your optimal DPS rotation actually looked like so you could endeavor to remain as close to it as possible.
Secondly, in an action combat system like Guild Wars 2 a concept like "real world DPS" is meaningless. Every boss, every fight is different. The DPS you could get against an easy boss is different than a super challenging boss. So if your DPS changes so much from encounter to encounter, doesn't that mean "real world DPS comparisons" are simply cherry picking data from whatever boss fight or encounter justifies your position? A lot of stupid people pick the Veteran Giants in Cursed Shore as a "real world DPS comparison." These giants can be killed in under 3 seconds or less with a build tailor made for that purpose. What does Veteran Giant kill times prove when comparing two to each other to say which one has generally higher DPS? Almost nothing, since the only thing that comparison is good for is evaluating which one if better at killing Veteran Giants. The best build for solo killing a Veteran Giant is substantially different than the best build for solo killing a challenging boss like Lupicus.
Since "real world" comparisons are essentially meaningless except when you have an actual goal (aka we want to set a speed run record for a particular dungeon or boss fight) since builds designed for actual specific goals are purpose built. When it comes to making a general comparison for the purpose of comparing the relative DPS among classes, the only methodology that makes sense is against a static target using optimal rotations.
So what about all those dead Veteran Giants? Do we learn nothing? Actually we learn quite a bit... about Veteran Giants! Which leads me to the purpose of this post (glad you stuck with me!), namely, using the ratio of optimal DPS achievable against the test golem in a vacuum vs the actual DPS achieved in game. Guess what that ratio tells us: how easy (or hard) a boss is. If one boss has challenging enough mechanics to limit your DPS to one-half of what it would be in an optimal rotation, we could say that boss has a .50 difficulty coefficient. If your meta berserker warrior build achieves 7,500 DPS against the test golem, but manages to achieve 3,750 against a Champion Broodmother in Malchor's Leap, she has a .50 difficulty coefficient. Simple enough right?
Now, there are some really cool things you can do with this framework. First, interestingly, some mobs will have a difficulty ratio of greater than 1.0! This is because when we calculate optimal DPS rotations against the test golem, we estimate 30 second fights. There are many reasons for this, but for today the important thing is that if you can kill a Veteran Giant in 8 seconds you can be reasonably sure that your DPS will be greater than our 30 second optimal number, since you're bursting the target using high damage, long cool down skills that get averaged out when calculating the sustained DPS you're capable of in long fights, but in short fights will give you spectacular DPS numbers. If you get 7,500 DPS over 30 seconds against the golem, but manage to get 12,500 DPS in 8 seconds against the Veteran Giant, we can see his difficulty ratio is 1.66. So a pretty easy fight!
Another thing, a fight that is easy for a warrior, might be hard for a Mesmer. Not because the warrior is capable of higher DPS necessarily, but simply because the warrior might be able to bring more defensive utility or CC or evasions than the Mesmer can without breaking from his optimal rotation. So it's entirely possible for there to exist bosses that might be a .75 difficulty for a Warrior but a .35 for another class. This, again, explains why "real world DPS comparisons" are worthless: the situation could totally be reversed for a different boss with different mechanics! Then how are you supposed to decide which class has better DPS?
A cool feature is that as you add more players to your party, fights get easier. Duh. Well that means your DPS will get closer and closer to the optimal even against challenging bosses as you add more players to your party. So you can use this framework to measure the difficulty of bosses for 5 man teams too. If raiding is introduced in the expansion, it will be very valuable to know your party's effective DPS vs optimal.
What else can this be used for besides cataloging bosses? Glad you asked, because cataloging boss difficulty is probably the least valuable information you can get from this. Why? What does YOUR difficulty ratio have to do with MINE? Simply put two players at different skill levels will have different ratios. And the ratios might not be consistent between bosses. A boss I've practiced a lot I might have a 1.0 against but you might only have a .75, but you could be a better player in every objective regard.
The ultimate use of this system is tracking your development as a player. If you get a .60 against a boss in May, and you practice and get a .80 in June, you can see a measurable improvement. If you play a lot and practice and you are getting improvements on nearly every boss even without putting in specific practice, you are objectively improving as a player.