Aion Analysis: Character Creation

Character creation in Aion was one of the most involving aspects of the game itself.  It starts off by presenting two faction options: Elyos or Asmodean.  In vanilla Aion, this had a large impact on your gameplay, as your faction would determine some of the abilities you would get access to later.  While this was eventually changed, there remained a factional imbalance with how the maps were set up.  That aspect will NOT be covered in this article.

After that, you are presented with 12 options for classes to play.  In vanilla, there were only 8 options.  There aren’t 12 classes to play, though.  The game presents male/female as different versions of the classes, although there is no difference in gameplay between the two.  The differences are purely cosmetic.

One of the more confusing aspects of this window is that you are presented with 6 stats in the lower left: power, health, agility, accuracy, knowledge, and will.  There is no explanation what these stats are for, and it is entirely up to the player to learn.  There is no equipment within the game itself that has these stats on them either, as HP is the stat that appears on gear NOT health.

These stats cannot be changed in any way, shape, or form until 5.0.  In 5.0, players are given a talent-point style system (called Essence Allocation), and I will cover this system in great detail in the future.  Before the Essence Allocation System, though, these stats were completely immutable, and they also did not indicate what they were for.

After a lot of experimentation by the player base, it was discovered that these stats indicated how well this specific class scaled with stats.  If a character had 100 knowledge, they gained 100% of the value of every point of magic boost.  If a character had 95 knowledge, they only gained 95% of the value of every point of magic boost.  If a character had 130 knowledge, they gained a whopping 130% of the value of every point of magic boost.

While I do view more knobs to help balance a game as a valuable tool, these particular stats always puzzled me.  They were presented to the player at class selection, and seemed to be of vital importance.  While they were very important, the way they were presented always made it feel like they were a stat that could be changed, or a stat that would come into direct play later.

But neither of these things are true.  These stats were always hidden in the background, and it was never something that factored into gameplay decisions later.  In fact, at 10, when you ascended to become your true class, these stats changed again.

This gave character creation in Aion a confusing feel, as it would sometimes present you with information that was neither important nor relevant.

Then comes the cosmetic portion.  I am sure virtually everyone has seen pictures of hideous and beautiful Aion characters, so I will not cover those here.  Instead, I am going to talk about the armor and ability showcase mode.  In this mode, you can view 4 types of armor:

  • “Naked”
  • Low level armor
  • Mid-high level armor
  • High level armor

You can also view 6 different abilities.  Of these 6 abilities, some may not actually be in the game, though.  This was one of the odder realizations I had when I bought a plastic surgery ticket (a character appearance modification item).

While this may not seem major, one of the cooler abilities that it showed for sorcerers was an ability that didn’t even exist within the game, at least not for sorcerers.

In conclusion, the game’s creator has a lot of cosmetic options, but it also has a bunch of information that only serves to confuse players (or simply be forgotten).  The stats, along with the sometimes missing abilities, are a confusing combination of things to present the player with that seems to serve no actual purpose in the game.

Click Here To Go Back To The Article List

Aion Analysis: Introduction

“Aion Analysis” is a series of articles I’m going to be doing on the game Aion.  This article only introduces what the series is going to be about, and it will not do any actual analysis.

Aion is a game developed by the Korean developer and publisher NCsoft.  It was originally released in South Korea in November 2008 and quickly became NCsoft’s flagship product.  It released in Japan in July 2009, and was subsequently released in North America, Australia, and Europe in September 2009.  The game was advertised as a PVPVE game.  One of the core focuses of the game was for PVP and PVE to often intermingle instead of being separate subgames.

In South Korea, the game experienced an incredible amount of success.  It surpassed Lineage 1 in terms of financial success, and it was the most popular MMORPG in South Korea.  Despite being wildly successful in South Korea, the game was not immensely popular elsewhere.  It saw massive initial success, but the success tapered off and, as the game was westernized more and more, the game continued to do worse and worse financially.

In spite of Aion’s tapering popularity globally, it is still rather popular within South Korea, often being ahead of World of Warcraft in LAN cafes in terms of popularity.  Outside of South Korea,

Before I go into analyzing the game, I feel that I should discuss my own background in Aion so that you may understand the context I am viewing the game through.  I am from North America.  I hit level 50 as an Elyos Sorcerer in patch 1.5.  I played sparingly during 2.0 and 3.0, but I came back and played again during 4.x and 5.x.  During 4.x, I got an Elyos Songweaver to 65 as well.  Most of my hours played are in 1.5-1.9 and then 4.3-4.5.

This is going to be a long series of articles and sometimes I will discuss very specific aspects of Aion.  The first article should be up on May 29th.

List of all articles to date:

Character Creation [May 29th, 2017]

Master X Master Master Idea – Chief Gunner Koakoa

All numbers on here are just meant to be there to give an idea of the range, power, or reusability of a specific skill.  This is not meant to represent finalized numbers.

Chief Gunner Koakoa

Core Design Philosophy:

Chief Gunner Koakoa is designed to be more a fight-controller and a zoner than he is a true damage dealer.  Koakoa’s abilities are very mana intensive, and he is heavy reliant on his passive to restore mana for sustained presence.

Weapon:

Koakoa uses a long-ranged charged weapon that fires an explosive projectile.  The projectile deals more damage the more the weapon is charged.

Ability 1:

Molotov Cocktail

Cast Range: 15 meters
MP Cost: 225
Cooldown: 18 seconds

Throws a Molotov Cocktail at the target location, dealing minor damage on explosion and placing a patch of fire on the ground.  The fire deals constant damage.  The fire starts as a 1 meter radius AOE.  The radius grows by 0.1 meters every quarter second.  Lasts 3 seconds.

Design Insight:

The purpose of this ability is to give Koakoa an option for a sustained team fight ability.  The growth rate may be a bit large, but I was just trying to spitball numbers.

This ability also provides very powerful wave clear and PVE power for stationary (or semi-stationary) targets.

Level Traits:

  1. Increases skill power by 5%
  2. Reduces cooldown by 1 seconds
  3. Applies a 10% snare to individuals within the fire pitch
  4. Increases skill power by 5%
  5. Increases duration by 0.5 seconds
  6. Reduces the delay between growth spurts from 0.25 to 0.2.

Ability 2:

Spike Bomb

Cast Range: 15 meters
MP Cost: 200
Cooldown: 15 seconds

Throws a Spike Bomb at the target location.  0.75 seconds after the bomb lands, it explodes in a + formation.  One of the lines for the bomb travels parallel to Koakoa’s forward vector at the time of casting, and the other line is perpendicular to the same forward vector.

Deals very high damage and applies a 30% snare to all targets that are hit.  A player standing in the center of the + where the lines overlap will take double damage.  Each “line” out of the bomb extends 4 meters from the center.

Design Insight:

A tricky ability to land, Spike Bomb is meant to provide substantial team fight burst and enable follow up.    Its unique targeting style makes it an ability that can be used to cut off corridors and deny space in addition to potentially dealing damage.

An alternative implementation I had considered was to have it pulse 3 times instead of just one, but the snare would be reduced to 10% per hit, stacking up to 3 times.  The damage would be reduced to roughly 35% of its current damage per burst.  The damage would be higher overall, but it would require the individuals to take all 3 hits.

I went with the single blast over the triple blast because Koakoa is already heavily focused around zoning abilities, and having another hard to land ability with high zoning presence might make Koakoa little more than a nuisance.

Level Traits:

  1. Reduces cooldown by 2 seconds
  2. Increases cast range by 1.5 meters
  3. Reduces detonation delay from 0.75 to 0.5 seconds
  4. Increases skill power by 5%
  5. Increases skill power by 10%
  6. Increases the length of the explosion by 1.5 meters and the width by 0.5 meters

Ability 3:

Bomb Party

Range: 20 meters
MP Cost: 250
Cooldown: 14 seconds

Creates a bunch of bombs in a 6 meter radius.  The bomb explosions do not overlap.  Bombs explode after 4 seconds.  If a bomb is destroyed, it does not detonate.

There are three bomb sizes – Small, medium, and large.

Small bombs die after 3 instances of damage.  Medium bombs die after 5 instances of damage.  Large bombs die after 15 instances of damage.

All bombs do the same damage (very high).

Design Insight:

While this ability is easily countered (by simply destroying a small bomb and standing in the gap), it is an ability that will heavily restrict enemy movement on the battlefield, making it a powerful combo ability.

Level Trait Ideas:

  1. Reduces cooldown by 1 second
  2. Increases skill power by 10%
  3. Increase the density of bombs by 10%
  4. Increases skill power by 5%
  5. Increases radius by 1.5 meters
  6. Increase durability of bombs by 100%

Ability 4:

Blank

Cast Range: 5 meters
MP Cost: 125
Cooldown: 15 seconds

A short range cone attack that deals low damage and applies a 1.5 second fear to everyone within the cone.

Design Insight:

This gives Koakoa a way to remove control of enemy masters at the expense of damage.  I had considered a 1.5 second blind as well.

Level Trait Ideas:

  1. Increases skill power by 5%
  2. Reduces MP cost by 25
  3. Increase duration of fear by 0.5 seconds
  4. Increases cast range by 1 meter
  5. Reduce cooldown by 1 second
  6. Blank reduces the defense of victims by 10% for Blank’s fear’s duration

Ultimate:

Bombs Away!

Cast Range: Global
Ult Cost: 2 bubbles
Cooldown: 45 seconds

Drops a bomb on top of every enemy master on the map.  The bombs are tethered to the master that they were summoned upon.  After a 5 second delay, the bombs all explode, dealing massive damage to any nearby enemies with a small falloff radius.

Design Insight:

The idea for this attack is that it drops the bombs instantly, but the bombs explode after a while to make Koakoa’s ult more of a fight disruption ult than a straight up team fighting ult.

The tethers would have a short delay before moving in the direction that the master moved as well.  This would allow the enemy masters to run from the bombs and outrange them.  The falloff radius is intended to match up with the base move speed of masters while attacking.  In this instance, it does slightly reduced damage to the player if they’re moving away while running.  If you run at full speed, you will outrange the bomb’s explosion radius.

Swapping masters will break the tether, but the bomb will persist.

Level Trait Ideas:

  1. Increases skill power by 10%
  2. Reduces cooldown by 5 seconds
  3. Increases explosion radius by 25%
  4. Increases skill power by 5%
  5. ??? [I suppose another increase skill power would fit here, but it feels generic having that fill out the ult]
  6. While a target is tethered with Bombs Away anywhere on the map, Koakoa’s non-auxiliary abilities have a 75% reduced cooldown.

Level 3 trait – Increases the explosion radius by 25%.
Level 6 trait – While a target is tethered with Bombs Away, Koakoa’s non-auxiliary abilities have a 50% reduced cooldown.

Passive:

Resupply

Every fourth weapon attack that deals damage to enemies restores 45 mana to Koakoa.

Survival:

Rocket Jump 

Koakoa detonates a small bomb to launch himself towards the target location.  Can be used to jump over walls.  While in the air, Koakoa takes 75% reduced damage from all sources.

Project Infinity April 12th, 2017 Patch Notes

Now With Pictures!

HUD:

  • Inventory name searching is now operational
  • Inventory sorting is now operational
  • Font size for chat bar increased from 10 to 12
  • Looting Widget added
    • Has no functionality in this build

Bug Fixes:

  • Rapidly moving items over and over again followed by using the sort button in the bottom left no longer results in the item duplicating.
  • AI release on death.
  • Dropping a spell onto itself no longer results in the spell being deleted from your spellbook.
  • Players should no longer immediately upon joining a server with multiple living players.
  • Spells no longer persist forever in the game’s memory.
  • Stats are recalculated upon leveling.
  • Right click turning now properly replicates.
  • Characters now turn to face the target they are casting on.
  • Spells can no longer be targeted on dead targets, but there is a persisting issue with AI (see below).

Known Bugs:

  • AI still attempt to target dead characters if they have no other target in range.
  • Overhead widgets don’t appear for targets other than yourself.
  • Equipment menu shows as a cluttered mess for the authority client.
  • If the AI is targeting a flying unit, the AI stutter steps instead of reconsidering the target.
  • After you go from walking to flying and back from flying to walking, you will experience minor rubber banding every time you stop moving.
  • The effect tracker for your target persists even after you change targets.
  • Any character (PC or NPC) dying applies the sepia tone death post process to the authoritative player.
  • Stats aren’t recalculated upon equipping or unequipping items.
  • Class information is shared across all classes.  Classes should level independently.
  • Too many units being spawned simultaneously results in a bottleneck wherein most of the characters fail to possess their pawn and are left in a limbo state.
  • Attempting to use a consumable while a spell is being cast consumes the consumable but does nothing.
  • Cooldowns do not persist after logoff.
  • The trees still have horrible collision models.
  • The camera attaches to the mouse as if right clicking after sending a chat message.

— Anaeradranax

Project Infinity March 28th 2017 Patch

Patching Process:

  • This version will be the final version that is a standalone version that requires all of the players to download the game in its entirety.  In future versions, a small patch file will be available to jump users from version to version.  This will reduce download sizes from 5 GB to 20-25 MB (except in graphically large patches).

UI Changes:

  • Spell effects now display on the HUD for both yourself and your target!
    • See below for a known bug involving the spell effects display.
  • Gear is now equipped or unequipped by right clicking not by left clicking.
  • Left clicking on spells still works.
  • Left clicking on a HUD element should not have the unintentional consequence of clicking through and clicking onto a player or object underneath, causing you to potentially change targets.
  • Right clicking on a player’s status bar will now properly bring up a context menu, allowing you to invite to a private party or to mark the target

Console Commands:

  • SetPlayerClass X where X is the number of the class you want to become (stored in alphabetical order).  This will be disabled when there is a better way for players to switch classes.  This command is case insensitive.

Private Groups and Marking:

  • Private groups have been added to the game.  You are able to opt out of the public group experience and instead control who is allowed into your party.  By entering a private group, you also gain the ability to use the mark component.
  • You mark someone by applying a number or symbol above their head.  Once a target has a number or symbol above their head, it follows them around, and the user is able to target that specific mark with a hotkey, making for easier targeting.  As stated in the private groups section, you cannot mark in public groups.  This may change in the future.

Chat:

  • Chat now wraps instead of truncating at the end of the line.
  • The chat scroll bar will now stick to the bottom of the chat log if you have not adjusted the offset, and it will stick to your offset if you have.

Threat:

  • Added a threat management component to all pawns so that threat is properly propagated across all of the targets.
  • Threat is handled through an algorithm that combines enmity, distance to target, and health of target.
    • Targets that are closer are favored over targets that are farther away.
    • Targets that are low on health are favored over targets that are fuller on health.
    • Targets that have more enmity are favored over targets that have less enmity.

Class Statuses:

Necromancer moved from “class fantasy design phase” to “first ability pass: implementation phase”.

  • Assassin – First ability pass: Test ready
  • Berserker – First ability pass: Test ready
  • Chaos Blade – Undergoing class concept revamp
  • Cleric – First ability pass: Test ready
  • Dreadnaught – First ability pass: Implementation phase
  • Druid – First ability pass: Test ready
  • Magus – First ability pass: Late design phase
  • Mesmer – First ability pass: Early design phase
  • Monk – First ability pass: Early design phase
  • Necromancer – First ability pass: Early design phase
  • Oracle – First ability pass: Implementation phase
  • Paladin – Class fantasy design phase
  • Ranger – First ability pass: Early design phase
  • Shaman – First ability pass: Early design phase
  • Troubadour – First ability pass: Early design phase
  • Warrior – First ability phase: Late design phase [Guardian was merged into Warrior]
  • Warlock – Class fantasy design phase
  • Wizard – First ability pass: Early design phase
  • Zealot – First ability pass: Late design phase

Bug Fixes:

  • Frostbolt no longer hits friends and foes alike.
  • Chat log no longer cleared on death.
  • The local player no longer sees every chat message as originating from them.
  • Overhead mark is now invisible until set, and it returns to being invisible upon unmarking.
  • Auto attacking now turns off if you are dead.

Known Bugs:

  • Clicking off of a target does not remove the spell effects display.
  • Right click turning does not replicate until you move without right click held down.
  • Minor rubber banding when you stop moving unless you are an authoritative client.
  • Ammunition is not consumed even if it is required for a weapon.
  • Only one character can be saved per server.  This is a shared quantity across all users.
  • Frames outside of the hotbars are not able to be dragged.
  • Blueprint Nativization crashes the game packaging process due to errors in Boost Bimap.
  • Sending a chat message attaches the camera to your cursor.
  • NPCs will continue auto attacking you after your death.
  • Spawning too many NPCs simultaneously will cause any player that also attempted to join the game to crash.

— Anaeradranax

Project Infinity March 14th, 2017 Patch

UI Changes:

  • Hotbar buttons now properly display GCD.
  • Status bar’s MP and HP values transition between the old value and the new value in a much smoother fashion.

New Weapon Features:

  • Extended weapons now exist!  Extended weapons have an additional 25% range compared to that type of weapon.  This is an exceptionally rare affix for a weapon.
  • Weapons are capable of doing custom damage types!  A sword that deals fire damage or elemental damage is completely possible in this new system.

New Spell Features:

  • Damage types can change to the weapon’s damage type.
  • Weapon dependencies (both optional and required) now exist.  An optional weapon dependency is a situation where the ability derives its stats from the specified weapon, although it can use hands.  A required weapon dependency means that specific weapon types are required in the specified hands.
  • Some spells now have a range that is dependent upon the range of the weapon they have equipped.
  • Instant cast spells no longer wait a tick to activate and instead activate instantly.

New General Features:

  • Auto attacking has been added
    • Auto attacking does not require a main hand weapon to be equipped
    • Dual wielding supported
    • When an on-GCD ability is used, auto attack timers are reset
  • Influence Map
    • The new Influence Map supports multiple teams and uses 1/1000th the memory (literally).

Class Statuses:

  • Assassin – First ability pass: Test ready
  • Berserker – First ability pass: Test ready
  • Chaos Blade – Undergoing class concept revamp
  • Cleric – First ability pass: Test ready
  • Dreadnaught – First ability pass: Implementation phase
  • Druid – First ability pass: Test ready
  • Magus – First ability pass: Late design phase
  • Mesmer – First ability pass: Early design phase
  • Monk – First ability pass: Early design phase
  • Necromancer – Class fantasy design phase
  • Oracle – First ability pass: Implementation phase
  • Paladin – Class fantasy design phase
  • Ranger – First ability pass: Early design phase
  • Shaman – First ability pass: Early design phase
  • Troubadour – First ability pass: Early design phase
  • Warrior – First ability phase: Late design phase [Guardian was merged into Warrior]
  • Warlock – Class fantasy design phase
  • Wizard – First ability pass: Early design phase
  • Zealot – First ability pass: Late design phase

Bug Fixes:

  • Characters no longer have a frame between when their HP changes and when their HP is clamped to its acceptable range.
  • Dedicated server now works again.
  • Server no longer crashes if the owner of the buff/debuff is no longer available (offline or dead).

Known Bugs:

  • Reidrectors do nothing right now.
  • Too many NPCs/PCs being spawned simultaneously causes problems with players possessing their pawns.
  • AI does not know how to treat flying.
  • Equipped items do not display on the character, but they display inside of the HUD.
  • Attempting to use a consumable while a spell is being cast consumes the consumable but does not give the benefits of the consumable.

–Anaeradranax

Project Infinity Influence Map

First demo with decals enabled:

Second demo with decals toggled on and off to show the performance of the system when decals are enabled vs. disabled:

This map is hideous.

Yeah it is.  This is a sandbox-test map that I use to demo anything before putting it into a “real” map.

How does it work?

It is a double buffered event-driven influence map that is based off of this article by Alex Champandard.

It appears to update in rows.  Why is that?

The amount of grids that are updated per tick (from the server’s perspective) are editable in the actor details panel on the right side of the picture shown below.

It updates a limited number of grids per run to keep the overhead low.  Setting the “Updates Per Tick” value to 0 results in it updating all of the grids at once.

What is the purpose of the influence map in this game?

The purpose of the influence map in this game is to simulate areas of control by the different factions within the game to indicate to the AI where it should focus their efforts.

What is the “BP_Outpost” that was added to the map?

The BP_Outpost added to the map is a constant influence point that the AI will construct to add spawn points closer to the front lines of battles.  The BP_Outpost, in its current state, adds no true gameplay ramification outside of the influence applier.  In the future, it will add spawn points to the world.

What does all of this lead to?

The goal for this is to create a system wherein the actions of the players or the NPCs can sightly shift the behaviors of the NPCs in the world.

What still needs to happen in the influence map?

  1. Performance upgrades
  2. Adding support for multiple factions
  3. “Smarter” spawning – Some of the grids being spawned simply can’t be pathed to, so spawning them in the current way is suboptimal.

What kind of performance upgrades are you looking at doing?

Now I’m going a bit on the technical end.  Right now, it spawns an actor for every single grid square.  That is horrendous, performance wise, although kudos to the Unreal Engine for being able to handle over 250,000 actors being on the map.  These are not ticking actors.  Only the original “InfluenceMap” actor is a ticking actor, and it chooses small subsets of the population to update.

I have a plan for a major performance overhaul which does involve moving away from actors for the influence grid and moving to a far, far more efficient system.

How large have you expanded it to?

512×512 is the highest I have managed to get into a game.  It is around 6 GB worth of influence map data on its own, although that is partially caused by them all being actors, and it is VERY obvious whenever a chunk updates.  So, that made it painfully obvious that I had to completely overhaul my system.

That seems excessively large.

It is.  Especially since no performance upgrades have been done yet.  This is the goal, though.  512×512 is the “target” for my project.  My project’s world map is intended to be split into an 8×8 grid of 64×64.

When is the next update planned?

I want to have the next implementation of the influence map ready to go by the end of next weekend.  I won’t have the debug visualization completed by then.

 

Hopefully I will have more specifics for the next blog post.

 

MXM Ranked Draft Feedback

To start with, my biggest complaint regarding this test is the ranked draft system.  By making bans only apply to one team, games regularly have identical bans.

In platinum, Sizuka and Death Knight are almost always banned.  In expert, it’s generally Sizuka, Nedien, and MBA.

Why, though?

Because in the current system, you are only banning for the enemy team, which means you are very frequently only banning masters that are either ridiculously strong in general or ridiculously strong against your comp.

The exact numbers I’m going to use for how many picks and bans is under the assumption that the game will reach at least 50 masters.  Currently, there are 35, and I think that 15 more by launch isn’t unreasonable.

In the current system, bans only apply to the opposing team, your team picks a total of 10 masters, and it’s possible to have the same master once on each team.

The smallest number of masters that will be picked and banned combined is 12 (assuming the same 2 masters are banned by both teams and the same 10 masters are picked).  The largest number of masters that will be picked and banned combined is 24 (assuming different masters are picked and banned across the board).

In this current system, though, Sizuka is regularly banned in Platinum and Expert level games.  Your team has no reservations about banning Sizuka because even if you ban her, the enemy still has to ban her or you can get her.  This makes the value of a ban extremely lopsided.

With that said, let’s talk about a different draft system: A semi-blind draft.

It would start as it does now.  The first set of masters would be banned.  Once both captains have made their selections, the bans are shown.  These masters are banned for *both* teams, and it is possible for both teams to pick the same ban.

The second set of masters must now be banned.  The masters that are banned by either side are not selectable, but this is another blind ban phase.

Now the first two players from both teams get to make their picks.  Once they have been picked, there is another 1-ban phase.  For this ban phase, bans are not retroactive, but they apply to both teams moving forward.  A captain can also only ban masters that are not already banned or visibly picked by the enemy team.  That is to say that a master picked by your own team can still be banned, but a master that has been visibly picked by the opposing team cannot.  This leaves situations where the blind picks can trick the enemy captain into a wasted ban.

Now the next two players will pick their masters.

After this pick phase, there will be one more ban phase where one master is banned again.

After this, the captains will pick their masters.

Won’t this take too much time?

Given the current amount of time, yes.  So, I propose a modification of how much time the captains have.

The captains will start with a reserve pool of 60 seconds.  Every banning phase will give the player 15 baseline seconds.  If the captain bans within the first 15 seconds, they do not spend any of their reserve pool.  If they take more than 15 seconds, they spend some of their reserve pool.  None of the phase’s base time carries over to subsequent phases.

Time Difference:

Current system takes a max of 1 minute during the ban phase.

Proposed system would take up to 2 minutes for the ban phase.  Slightly longer, but not significantly so.

How much of a difference could we see?

Raw Master Counts:

We would see 14 masters picked and banned at the minimum (if both teams ban and draft the same masters).

We would see 28 masters picked and banned at the maximum (if both teams ban and draft completely different masters from each other).

Hopefully we will see changes to the drafting system to encourage a larger variety of masters in game, and I think that allowing for more bans is the most reliable way to encourage it.  Balancing is the best way for encouraging a wider swathe of masters, but the metagame in MOBAs tends to shift for weeks or month before it starts to stabilize, so it is also much harder to make good balance changes and not simply make kneejerk changes to early perceived imbalances.

A deeper ban system also allows for specific tiers of players to better handle was is strong at their specific bracket.  In DOTA 2, for example, even when Invoker is strong, he isn’t a good hero at 1k or 2k, but he is a strong hero at 6k.  You can’t realistically balance the game for both tiers, so giving more bans is a great solution.


Don’t Automate The Captain Unless Necessary

I know that this change combined with the first one will extend how long the draft phase goes, but I feel these will help the overall quality of matches.  Sometimes the highest MMR player doesn’t want to ban, but sometimes he does.

For this, I propose a 15 second captain selection phase at the start of the match.  People who wish to be captain must offer to be captain during this phase.  If multiple people offer, it will select the person with the highest MMR of those that offered.  If none offer, it will select the person with the highest MMR on the team.

The other 4 teammates will still be sorted in terms of MMR as they are currently.

My Hopes for the MXM Titan Ruins Match API and Why

All examples that I use in the sections are simply arbitrary examples.  They are not meant to represent how I think certain metagame aspects will play out.

Also, I do understand that some of these are crazy and would take up a lot of space, but I am putting these ideas up here anyways in hopes of potentially sparking anidea on how

  • An indication as to who is the “bottom left” team and who is the “top right” team.
    • This may sound obvious, but given the symmetry of the map, it may be very difficult to determine without some indicator.  This is important given that “top lane” is the opposite lane relative to each team.
  • Picks and bans in order; in matches without bans, still indicate the order of picks, and keep them in shown/hidden order to provide more information
    • I would like to be able to see pick rates along with perhaps information regarding counterpicking.  Perhaps when Kromede is a shown pick she has a 42% win rate, but when she is a hidden pick, she has a 55% win rate.  When Sizuka is picked early, maybe it causes a sharp increase in the number of Death Knight and Ignuma picks (or Death Knight and Ignuma bans).
  • Ability breakdown by hero (in the proper order as well)
  • All master deaths, the time of the death, the location of the the death, and the damage breakdown (preferably by abilities and masters as well, but at the very least by masters and owning players)
    • This one would be useful for parsing information about where a player is most frequently caught out of position, or perhaps where a player likes to stalk.  Maybe Kat the Cat is heavily involved in kills that happen in the first 2 minutes on top lane.  If only the master is indicated, it would be difficult to dissect which Kat the Cat (if there are two) was the one involved in specific kills.
    • I changed this from “kill” to “death” in case a player dies to something that isn’t a player (river king is a frequent killer)
  • All neutral kills (the four jungle camps, river kings, ritualists, the Summoned Guardian, and the two altars), the time of the kill, the damage breakdown (only owning player is necessary in this case to me), who initiated the fight, and who secured it.
    • In the case of most of the camps, the team to kill it gets the reward (what I refer to as “securing it”). The dream scenario here would be for the API to provide enough information so that it could be dissected that Blue Team initiated on the Summoned Guardian, dealt 80% of his life in damage, and were discovered by Red Team, who then wiped them and took the Summoned Guardian.  In the case of the altars, this information could be helpful to determine which players are very good at contesting or stealing altars.
  • Skill level up information (i.e. Took Q at level 1, E at level 2-4, Q at level 5).
    • This one would be useful for showing how a player chose to skill up, and perhaps show where certain players feel that a hero’s “breakpoints” are.
  • Trait information, including re-doing traits
    • This one would be useful for showing if a player has a specific trait build against certain masters or perhaps even a certain trait build up until, say, a big push down mid where the player swaps all traits to be offensive
  • A by-wave creep participation breakdown.
    • This is the one that would consume the largest portion of data, and it is likely too much to actually be implemented.  The idea for this, though, would be something like…
    • 1TR (1st wave, top, red) – B2A3 (Blue team, player 2, first [shown] hero, 3 points), B2B2 (Blue team, player 2, second [hidden] hero, 2 points), B4A1 (Blue team, player 4, first hero, 1 point)
    • In that example, the stuff in the parentheses would not be included, but it shows which blue masters were involved in combat with the red Titanites.  It could potentially just be merged between both teams to be…
    • 1T – B2A3, B2B2, B4A1, R1A4, R2B2
    • That would indicated that the second and fourth players from blue and the first and second players from red participated in the first Titanite wave top.
    • This would allow the parsing of lane information.  Perhaps Lua has a 60% win rate if she goes mid, a 42% bottom, and a 56% top.
  • Titan appearances by time (and whether Kinetic, Ardent, or Helix)
  • Titan Incarnate usage by player and time
  • Destruction of towers by time and damage breakdown