"Hey all!
Anyone want to talk about LoL, game design, how we make champions etc? Going to be around for a couple of hours and looking forward to chatting with some of you folks.
To give you a quick bit of background I've been at Riot for about three years now. I've previously worked on Ziggs, Varus, Syndra, Lissandra, Elise and Gnar. A lot of my time these days is spent looking at the year ahead, working with other Rioters to figure out what sort of champions we want to make, what our goals are etc.
Edit: Xypherous, from the Systems Team that handles stuff like the Preseason changes, is also going to drop by to chat. If you've got questions about that sort of stuff should definitely throw them in too! He also worked on Nautilus, Lulu, Orianna, Fizz and Renekton before that as well, so if you've got questions about them he's the guy to hit up.
Edit 2: Thanks for the chat all, I need to head off and grab food. Will drop back later tonight though and add in a few more replies."
As mentioned in the first edit, Xypherous also joined in the Q&A:
"Hey guys, you might notice me popping in random comments as well. I'm Xypherous and I used to be a champion designer, having designed Nautilus, Lulu, Orianna, Fizz, Riven and Renekton. I'll be helping Meddler out fielding some questions that he might miss.
Most recently, I helped out with a large part of the Seasonal work for Season 2015.
Ask away. :3"
I've sorted the Q&A into a few categories for easy reading:
General Champion Design
As for if the design team tries to design new champions with regard to the current meta or popular strategies, Meddler explained:
"When coming up with goals for a new champion we're focused primarily on what we want the game to be like long term. If there's a strategy that's interesting/good for the game and not well enough supported by the current roster we'll look at making new champs to support it, whether or not it's a strategy that's part of the current meta. Kit design for a champion happens 3-6 months in advance of their release additionally, so designing for the exact state of the game today's not practical even if we wanted to."
Xypherous also replied to a summoner asking if they design champion abilities based on lore or if it is the other way around and the lore is based off the abilities:
"It depends on the individual champion designer - and it tends to be kind of circular.
For example, Fizz's Shark used to be throw an exploding trident at someone. Then somehow, people started saying Fizz rides a shark - and so the exploding trident became fish --> giant shark attack.
I think Fizz's 'E' came about because I simply did not know what to do with yet another pole-arm and thus had to watch every pole-arm weapon user in every video game I could think of before I stumbled across Kilik's Playful Pheonix move.
Other champions, like Orianna - had lore that was designed around the kit. We had this autonomous ball thing and a set of cool skills - so how do we skin that. It varies from champion to champion a lot of the time."
Meddler also answered a question regarding if multiple designers work on the same champion or if the separate into smaller groups, saying:
"Generally there's one designer who's the primary one responsible for the gameplay on a new champion while they're being developed. The rest of the champion design team are offering feedback, ideas, advice etc though of course, and there are also a whole bunch of people from other disciplines (engineers, animators, concept artists, QA, producers, sound designers etc) working on each new champion - it's very much a group endeavor."
Xypherous chimed in:
"You typically have on champion designer who is the 'main' on any one champion. They gather feedback from playtests and such but ultimately a lot of the gameplay decisions are made by a single designer."
Meddler also commented on the amount of time it takes to actually develop a new champion, saying:
"Making a champion takes around 6 months on average, though it can vary quite a bit depending on how quickly an idea comes together, what issues we run into etc. Most of the gameplay design (figuring out their kit) happens over the first few months), as does things like nailing down their general appearance, writing their voice lines, figuring out their background etc. After that it's mainly about making the actual model, animations, visual effects, sounds etc that'll go into the game proper (we use a bunch of placeholder stuff to test prototypes before that)."
Xypherous joined in to elaborate on how long it takes before the team starts thinking about numbers:
"Maths is odd. Almost every champion designer I know fiddles with numbers during testing and iteration to figure out whether or not a champion feels right. There's a really intensive balancing pass 3 or 4 weeks before the champion will be released done by a cooperation between champion and competitive playtest teams.
As far as maths to limit overpowered abilities or buffs to abilities, it really depends on a skill by skill basis. You can make a skill easier to land or you can increase the payoff of the skill when it does land - in development, trying to figure out which of these is better is part of the process."
As for how difficult it can be to design new champions, Meddler replied:
"We spend quite a bit of time before we start making a champion identifying what we call 'opportunity space' for them. What sort of new gameplay can we add to the game (that we want there), what sort of personality can we offer, what stories can we tell, what new appearance can we capture etc. As the size of the champion roster gets bigger that does get a bit harder in some ways. It also fleshes out the world though and that gives opportunities to do some pretty cool stuff tying champs or factions together (things like the Diana/Leona connection or Azir as a way of exploring Shurima's history are really valuable for the game to my eyes at least)."
Meddler also replied to a question asking what his favorite part of champion design is:
"The initial brainstorming, particularly working with a small group that includes a writer and concept artist, is a lot of fun. I also really enjoy the end of the process, where all the assets come together in game and the character you've been working on from months goes from a recoloured version of another champion with placeholder effects and no voice to a real champion. A champion's release is also great too of course - finally get to share what you've been working on with a bunch of people and (hopefully!) watch them enjoy playing it.
On the design side trying to nail down the final ability on the kit can be pretty tough sometimes. When you've got 3 of the 4 abilities figured out you're trying to figure out what the missing piece you need is, in a way that'll be healthy for the game, satisfying to play with, balanced, won't overlap too much with another champion etc. Sometimes those constraints give you a really clear picture of what's needed, sometimes they leave you feeling like you're really restricted in what you can do."
When asked if there are any champions he looks back on and wishes were not released, Meddler replied:
"There aren't any champions I don't think should be in the game at all, I think they've all got a solid core to them. In some cases though that's not very well executed on though. Urgot and Yorick for example have cool underlying ideas (undead cyborg war machine and gravedigger with pet ghouls) that aren't well realized.
Oh, yeah, then there's Teemo. Teemo should burn."
As for his favorite champion that he created, Meddler noted:
"Hard to say, I find it pretty hard to compare and enjoyed working on all of them. I do have a bit of a soft spot for Ziggs though since he was the first major project at Riot that I did the primary design work on."
When asked why several champions released after Yasuo has a knock up, Xypherous commented:
"We tend to use knockups just because they feel really impactful without actually adding a whole ton of power. They're visually satisfying without being too disruptive to gameplay compared to the alternative..."
Xypherous also commented on how the team decides who needs a rework next:
"There are a lot of competing factors here but generally speaking, these are the things to think about:
State of their kit / Design Health
Visual Quality of the Character
How 'iconic' is the character to League of Legends
Has there been previous pre-production in the past that we should really finish
Does this character thematically fit a larger initiative that we want to do
It's a large set of considerations to be made - kit and power are only one among a lot of others."
Meddler also commented on their recent approach of adding variety to female champions design:
"Something we've been aiming for is to get a greater variety of appearances/tones for champions of both genders. There were a couple of years there where many of our female champions especially ended up looking pretty similar and we've been making an effort to try and avoid that. That's not to say we won't continue to make attractive champions, male and female, where appropriate for their character, but that we also want a wide range that includes champs like Kalista and Rek'Sai as well.
I can't speak with personal experience about the intent around the game's initial release, that was well before my time at Riot. Certainly there seems to be a lot of support internally for the greater variety we've been exploring recently though."
Specific Champion Comments
When asked if there are any plans for Annie changes, Meddler shared something they are experimenting with::
"What we're currently looking at (note this is experimental, so could well change tomorrow) is slightly lower resistances on Molten Shield, with Tibbers also getting the resistance bonuses as well plus a brief movement speed boost. One of the big goals there, in addition to balance changes, is to make casting Molten Shield a bit more interesting and add a bit of extra depth to how you micro Tibbers."
Riot PhRoXzOn jumped into the Q&A to mention upcoming plans for Fizz's live balance:
"There will be some tentative changes on the PBE for Fizz within a few days looking to alleviate some of the concerns, especially the up front burst from his Q and giving him more gameplay in general while still keeping the trickster element to him.
We definitely look at off builds. Take the Tristana rework for example, where we wanted to keep the AP off-build AND also give it more gameplay (tension between jumping in first, or stacking up the bomb and then jumping). We also look at and test off builds than can potentially arise as a result of the changes (where an off-build previously didn't exist)."
On the topic of Fizz, Xypherous also commented on the original intention behind Fizz's E:
"I'm totally responsible for Fizz's 'E' - and the intent was to make a frustrating skill that conveyed that aspect of Fizz's personality. I think my goal was something like 'Let's aim for something that feels like Shaco in terms of trickery/frustration without going too far.' In retrospect - it's probably worked too well in a lot of regards with that.
The idea for Fizz's 'E' was primarily to allow him to evade opposing damage without actually being durable - so that every now and then his opponents can choke Fizz and obliterate him and feel good about that.
Fizz was really meant to be a character that felt slippery enough so that you experience a moment of catharsis when you finally catch the jerk and annihilate him.
As to lane functionality, it was meant much more for the 'lower the poke he takes' rather than the 'instantly annihilate you with Q' that it's become. In hindsight, I wish there was way more incentive to stay on the pole rather than instantly double-tap to move faster."
When asked about the status of Gangplank's long discussed rework, Meddler explained:
"Bit early to say much alas. We had a crack at a GP rework last year that had some interesting stuff in it, but didn't quite hit the mark so didn't get shipped (or talked about after a while, our bad, we should have mentioned it was a project that had been put on hold). We're going to give him another look this year, wouldn't expect anything for a while though, a number of other reworks will come out beforehand that are further along already."
When asked what role Lissandra was expected to fill during her design and any comments on Ahri's current PBE changes, Meddler replied:
"Our expectation during Lissandra's development was that she'd be played primarily as a solo laner, with which lane being defined by where players found her most useful. Currently pretty happy seeing her used as a flex pick who could be mid or top lane, I feel that results an interesting variety of lane match ups and team comps. One possible concern is whether she's crowding out melee champs too much, along with a few other ranged champs with strong anti melee capability. Don't think we've got any current plans to change her, though she's on our 'keep an eye on in case this champ's too dominant' list.
Not much news on Ahri besides the stuff that's already been reported from the PBE. One nice bug fix I don't think's been discussed though is a bug we've found and fixed with her W where, if your passive was up, the third Foxfire charge wouldn't deal damage at all under certain circumstances. That fix will be going out in patch 5.2 as far as I know"
Xypherous also commented:
"Quote
I believe Lissandra is a great champion who is under-appreciated, she is hardly ever seen in games, do you believe her AP ratios and skills are where they need to be or does she need a buff to be considered a top-tier mage?The pro players have picked up Lissandra as a flexible top-laner - so hopefully that period of under-appreciated will fade as people learn to realize her awesome.
Additionally, removing DFG should allow for Lissandra tweaks without immediately turning her into unstoppable assassin but we'll have to wait for that to go out first."
As for the design of Syndra and her current ultimate, Meddler explained:
"Quote
Hey Meddler! I really love playing Syndra and have a lot of fun with her, my questions: 1. How difficult was it designing a champion who relied on her q ability for every other spell. 2. How did you guys come up with the idea of her incredibly bursty ult?Took a little while to hit upon the Q as the foundation of the kit, though then felt obvious in retrospect since her original concept art featured her spheres so prominently. Once the basic idea of 'Q generates objects, other spells manipulate them' came together things were relatively smooth.
The intent with her ult was to give her a reliable tool (given her other abilities were all skillshots) and allow her to punish someone caught out of position or diving into her team (theory being that as a non mobile mage she'd have limited target selection). Ended up being a bit more of a 'burst down a key target after stunning them' ability than originally intended. It's counterplay's a bit lower than I'd like in that regard, though having said that champs do often need tools like that if they're to compete with other highly mobile options."
Meddler also briefly commented on the scope of Taric's future rework, saying:
"Taric we'll be doing a substantial rework on (not Sion levels change, but bigger than most, is the expectation). Not aware of any current plans for Olaf, though agree he could use a little something, given his CC immunity though I'd be cautious about adding mobility to him though."
When asked about the intent behind Varus' design, Meddler shared:
"The intent with Varus was to make a champion with strong poke, a lot of utility by ADC standards and reasonable auto attack damage, in exchange for low burst and mobility. Fitting into the gap between Ashe at one end of the immobile spectrum (heaps of utility, lowest ADC damage) and Kog'maw at the other end (heaps of damage and poke).
Been thinking recently we might want to strengthen his poke a bit (possibly by tying his passive into more than just his auto attacks). Haven't had a chance to bounce any thoughts about that off the live balance team yet though, see what their current thinking is."
Meddler also briefly commented on Ao Shin, a elemental dragon concept that was previewed but was constantly delayed afterwards::
"Quote:
is Ao Shin going to be released at any time in the near future?
Someday, yes. Soon, no. One of the big problems we hit and need to solve was how we'd animate his long, serpentine body and how that would be treated in gameplay (how does it act when he turns, which bits are targetable etc)."
As for why certain champion concepts, such as Omen or Sand Mage, were either discontinued or delayed significantly before being released, Meddler noted:
"Sometimes champions don't work out and we'll cancel them. That was the case with Blind Monk, Omen and Sand mage. In the case of Blind Monk and Sand Mage we decided the core idea was worth trying again, and went back to them and gave it another shot. The versions of those champions that got released are really different to the originals that were stopped at the time though."
Xypherous also added:
"For Omen and Sand Mage at least, a large part of this was that we didn't really have the tech infrastructure to really bring their fantasies to life. There's a lot of under the hood work that both those champions needed that wasn't exactly available at the time.
At some point, Orianna was 'cutting-edge' tech wise for us. The things that these guys required make her look primitive in comparison."
Gameplay
When asked about the preseason changes and the goals heading into the 2015 season, Xypherous explained:
"The goal of Season 5 was to figure out a way to make each area of the map feel different and to put lots of different power scattered across the map that could help for different purposes.
We wanted to make each reward on the map and each objective have very different purposes to them - and so that's why the rewards on the map range from small permanent power (gold/xp) to delayed power (Dragon) to temporary ridiculous Siege (Baron).
We also added a ton of inherent base defense (base gates, tower mechanics) but the hope is that we also balanced that out by making siege power more accessible or not as necessarily (Bonus Siege Power: Elixirs, Baron. Map Control: 5-Dragon reward)He continued:
We'll still be tuning if the game feels too stalematey - but hopefully that helps clarify things."
Quote:
I feel that the sieging part of the game is still somewhat awkward due to the fact that the new inhibitor tower murders minions waves far too quickly. In the game mentioned there was a nidalee who simply spammed spears while vayne was condemning us while simply farming up until full build. Only problem I have so far is the fact that the inhib towers are waaaaaaay to strong for sieging, other than that, good job with the update, the game feels likes a fresh new game nowAah. I see - that makes sense.
The natural counter to that should have been Baron (since one of the primary benefits to Baron is effectively making your Melee Minions invulnerable.) However, right now Baron is a little too difficult and risky for that to be true but I believe RiotAxes has been trying to tune Baron so that case gets resolved by a Baron.
I'm glad to see you liked the update though. :3"
As for which season 2015 change is his favorite, Meddler shared:
"I really like how dragon's now an objective that matters throughout the game, that teams plan around and fight over, rather than something that, late game, someone just does on the way past at times."
When asked about his thoughts on Essence Reaver specifically, Xypherous explained:
"We could fork Essence Reaver off Brutalizer - and the item would be immensely more popular and bought more - you're absolutely right on that.
Unfortunately, making Essence Reaver's build path and build up good is something that would most likely send a certain class of AD assassins (whose gates are early/mid-game Mana) into somewhat crazy levels. Pantheon/Talon, etc - would get the lion's share of benefit until ER was nerfed to be suitable for them.
Ultimately, ER will always kind of be bad until Assassin's have their own itemization and have their inherent power reduced.
The weird part about items is always that we can always make an item good or powerful - but making that item good and powerful on the people we want it on and not cause side effects elsewhere is the tricky bit."
Xypherous also provided more context for why DFG is being removed in patch 5.2, explaining:
"So the main thing that DFG was keeping down was the ability to use base damages on 2/3 damage skill kits with damage ults.
Basically, whenever we tried to add enough base damage for the character to be able to freely itemize something other than raw AP - DFG basically made that approach invalid as the amount of base damage + % amp + % health damage (another form of base) tipped it over into assassinating squishies.
With DFG removed - it allows us to basically use base damages on mages much more freely than in the past. It's true that we'll have to worry about some flat penetration builds - but those can be much more easily itemized around than the DFG build.
The con is that we've removed one of the keystone AP items in the game - as well as a key fantasy piece for players - Blow the hell out of that dude because he needs to die (as well as the ability to buy new spells in the game that feel magey.) It's something that we'll have to replace without falling into the same pitfalls.
The other thing to figure out is how to actually support the fantasy of assassination as a mage without limiting our ability to do really cool secondary mage patterns like tank mage or mobility kiter - it's a tricky problem."
As for a new AP item to replace DFG, he noted:
"I don't know about 120 AP and cooldown reduction.
I do know they're looking for something to fulfill the 120 Ability Power space - although what that is exactly is uncertain.)"
When asked about the current state of ADC itemization, Xypherous shared:
"For the short term, there's a wide range of ADC itemization paths depending on the ADC. However, there is little variation on the items that an individual ADC buys.
There isn't a whole lot of range on a specific ADC for the items they can buy - but at least they're buying different things (I see 3 or 4 main builds that vary across type - and then there's Hurricane on Kalista - so there's variance across ADCs but not in an ADC.)
That's in an okay spot for now - but in an ideal world lots of things would be different in terms of reactive purchases and such but I think the other classes' itemization deserve a touch up before we delve back into ADCs.
For an example of more long term stuff - critical strike is eventually going to have to be looked at to prevent AA based carries from forever being the norm, for example - but that exploration will take a while."He continued offering a wild ( and not likely to be implemented) idea:
"(Warning - the following comment is entirely Pie in the Sky and not going to happen in the near future.)When asked if he had server considered just removing lower tier crit items, Xypherous replied:
I'd kind of like to envision a world where every ADC has a special 'Crit Damage' skill like:
Vayne - Silver Bolts (Every 3 hits, deal X + Y% of the target's maximum Health as true damage. Every critical strike does Z% of the target's maximum Health)
-or-
Varus - Blighted Quiver (Critical Strikes automatically apply full Blight stacks to the target).
That way, every carry would look at crit a little differently - based on their pattern and what they think they could accomplish etc."
"I tend to agree that the lower crit items tend to be silly as they don't really add a ton of power and both feel bad and are dangerous in lane.
As to 'more different crit modifiers' - my personal take is that it'd feel cooler if criticals amplified the ADC effects rather than being standalone - Something cool was lost when Draven's Bleed was removed in terms of ADCs and their relationship to crit that I'd want to figure out how to recapture on their kits.
As far as Flat AR/MR pen - I tend to agree with you that their damage amplifier case is kind of odd. At the same time, I can't deny that there's something interesting about Flat Pen being a dedicated build path to explode squishies.
As far as 'Reactive' countering goes - I'm not sure that countering someone's armor and magic resistance will ever really be reactive - mostly because, they know your damage type. You know your damage type. You know that you will eventually be itemized against and thus you will eventually itemize against that resistance.
It's true that you could make the items more of a hard counter to resistance stacking - but I'm not sure that's what people want to have (or fight against, really.)"
Other
When asked how separate the Systems and Balance team are, Xypherous explained:
"They're two different teams. We both have a strong focus on the state of the live game but systems tends to take a longer view on game flow and game health while Balance tends to look at the immediate things affecting champions.
For example, systems might think about whether or not we want more or less snowballing in the game and adjust things like jungle gold or laner experience to accommodate while Live might look at a specific champion and whether or not he should be changed in the upcoming patch because he's too snowbally."
When asked what some of his favorite scrapped champion abilities were, Meddler shared:
"Varus used to have the ability to set brush on fire. Was very cool, though only moderately useful and at least at the time we didn't have the engineering support to do the ability properly either (make it look good, temporarily remove brush etc). Hope we go back to that someday though.
One of Gnar's first ults allowed him to devour an enemy champion, removing their vision and carrying them around in his belly while he dealt damage to them for a few seconds as they got digested, before spitting them out. Really liked the skill overall, didn't fit well with his limited control transformation however (was only accessible in Mega form of course), so we pulled it off. We're trying it again on another character in development right now, might not work out but very interested to experiment with it."
"A ton of moves were scrapped in the making of him - he started out as being very light in CC because he had the hook but... yeah.
Off the top of my head, here's the things I remember trying:
A punch which knocked back anyone but the primary target (Xin Zhao's current ult).
Passive X-Ray eyes that could see through walls in a cone.
Cage of Pain, which brought up a giant circular shock/slow 'fence' - like a super big wall of pain.
Bullrush, where you got MS and flung anyone you moved through (multiple times, charging bull kind of deal)
Throw - where you impaled someone on the anchor and then flung them and the anchor to a point of your choice
'Enrage' when a nearby ally got attacked (yeah...)
I tend to cycle through skills a lot - I'm sure there was more that I can't remember. I just remember a long phase of Nautilus being 'A bull who sets up cages that he then chases people in.'"
When asked how he got the job as a champion designer at Riot, Meddler explained:
"About four years ago I was working in New Zealand as a transportation engineer. I'd been playing LoL for a little while when I noticed a post from Zileas (our head of design at Riot) looking for applicants for game design. One of the categories of people he was interested in hearing from was people with professional experience in an analytical field with an interest in game design. Sounded interesting, figured I had nothing to lose so might as well give it a shot and, eventually (after many interviews and some visa delays) wound up here :)."He continued, commenting on his university background and advice on starting a career in game design:
"Riot tends to hire a lot of designers with analytical backgrounds (programming, engineering, hard sciences, maths etc), though that's because people with those backgrounds are more likely to have the sort of approach we're after, not that those are essential. I studied Geography and Film, TV, Media Studies initially then, after working for a few years, went back and picked up a Masters in Transportation Engineering.
In terms of advice biggest thing I'd recommend is working on games directly, whether mods, flash games, board game prototypes or whatever. That both teaches you a lot about what works/doesn't and gives you some great examples to talk through/add to a Resume."
No comments
Post a Comment