endstep

behold my masterpiece

The meeting where we made the calls on which concepts we would present for voting was pretty tame, but there have since been some heated debates around what exactly the Eerie Invitation is inviting you to and what yordle bagpipes sound like

Mandatory pro changes are part of the balance framework (I think this is the most recent dev blog about it, but it's a few years old and I don't work on the Live pod so it's possible there's a more recent post or a more recent version). There are other scenarios where champions are changed for pro purposes (e.g. the framework doesn't suggest champions to buff for pro, only champs to nerf), but the must-nerfs come from there.

Using K'Sante as an example case, looking at top 5 regions, on 13.12 he was nerfed and fell to 70% presence (from 80% on the previous patch). 13.13 saw him drop again to 56% presence, and limite...

This is not true in any real sense - I imagine this line of thinking probably originates from people talking about Elementalist Lux requiring the loading of a large number of different models (which is true), but from a gameplay perspective things like different skins, a champion in different modes, etc., are all the same, as it should be. There are of course some minor exceptions (e.g. in the past there was a logic fork for Zz'rot that would modify its functionality if the gamemode was Urf to prevent people from spamming infinite portals and pushing), but for the case that's being discussed here it's simply not true - if you want to modify, say, a champion's base damage on a spell, that's literally just changing a number in one file. All the tooltips will pull from that value, and you change it in one place and you're good to go (there are exceptions of course, some calculations are hard to show for tooltips for example so in some cases it will be two values you need to change, bu...

Magic reference, to the End Step, the last part of a turn. I haven't really been much into Magic in years though.

holy sh*t is that NIBIRIA I am a HUGE fan

oatmeal raisin perma F tier

hell yeah

np

I am also a pretty big AP Voli fan - enough of a fan to play it for ~75 games straight at one point and have a silly montage made of some plays.

Lutzberg (the designer on the Voli rework) did explore a lot of different ways to scale AP Voli and different avenues to let AP provide power, and the one part of those that ended up shipping was the AP ratio on the passive AS. However, from my recollections of conversations back then (which may not be perfect) as well as my own thoughts here, anything that "cheats" P stacks to max or even grants easier access to max stacks comes at a cost to the power of P, particularly for AP builds. There are very few spells that scale multiplicatively with a single stat, but Voli P scales both the damage and AS with AP, and since AS sc...

This was something Voli had in early testing. It's been a while so I may not be remembering everything perfectly, but I believe what we found back then was that letting R stack P meant that the max-stack P payout had to be lower than we wanted and R's damage also had to be meaningfully lower, so we made the choice to unbind P max stack payout from R even though it was a cool fantasy win to grant full stacks on R.

I don't work on PC League anymore so I couldn't tell you the current thinking there but I suspect the same concerns would probably exist for this mechanic today. The bigger concern is around the P max payout, where it should be a really powerful and satisfying moment, and if you can "cheat" your way there instantly with R then there's much less gating on it and necessarily it can't carry as much power or satisfaction.

You do not. But maybe one day.

yes

view 20 match histories and get one free

Hey, that's me!

GATGATGAT

It's not that complicated. Lifesteal is represented internally as a decimal - 8% lifesteal is .08, so naturally if you miss a decimal place or typo and put in .80 instead you end up with 80% lifesteal instead of the intended 8%. I've worked as a professional software dev before (though it's not what I do at Riot) and I can tell you that this type of error is quite common, particularly when trying to do fast iteration and push things into a testable environment (which is exactly what happened here).

In League, most of this stuff doesn't get seen by a wide audience though as it never goes to PBE and is fixed before it leaves internal environments. This is one small case where the list was shipped to PBE immediately and so was seen (and was certainly disruptive to PBE games). A parallel here to non-game dev is that you wouldn't see this type of error in other software like web pages, mobile apps, etc., because they don't have publicly-facing non-prod environments of any kind ...

I'm not an expert on the topic, but I don't think there's much (if any) policy around "similarity of Rioter names." There are at least two Riot Penguins, and probably a variety of other similar and/or exact match Riot names. Situations where that matters just don't really come up (the only situation I can think of was some humorous confusion around which Riot Penguin was being referenced in a random normal game I was playing).

It depends on many factors - in many cases this is true, though not always (e.g. when Darius' mana costs were raised in 10.6, his drop in winrate was relatively flat across different MMR brackets). In particular mages tend to exhibit this most, though of course it depends a lot on which champion, which spell's mana cost, and so on.

This particular case is interesting because Illaoi is one of the most (if not the most) mana-constrained fighter, so the expectation is that mana likely isn't skewed towards newer players/lower MMR players given that unlike on less mana-constrained champions, more skilled Illaoi players still have to opt into mana itemization/rune choices and still run oom with better skill usage. It is a scaling buff which tends to favor low MMR (level scaling is more impactful later in the game, games in low MMR last longer on average) so that might mean there's a light low MMR skew here for a separate reason but typically mana/level buffs don't exhibit this v...

It's true that part of why we didn't go forward with one of the various "thread-attachment" ults (even though they were very cool and a slam dunk on the thematic) is because they were very heavy on cc and pushed Gwen away from the skirmisher space, but there were also other reasons.

The coolest version in my opinion, which is the one shown in the gif in this article, also had an issue where it needed Gwen to input multiple different directional commands back-to-back which didn't fit well with the base kit which already asks you to do that (both Q and E are high-frequency directional casts that are often cast on different vectors). So while it was super fun to use in isolation, it was very demanding of the player to manage casting Q + E + R1 + R2 all in a short timeframe given they were typically all in different directions and some required high precision (mostly the R casts). One of the big advantages of the final ultimate, even though it has three casts instead of two, is...