Wednesday, January 21, 2015

Chemistry: Tendencies, Playstyles, and Comfort Zones Matter

I'm gonna start this blog off with an anecdote from my job. For those who don't know, I work in a local fast food kitchen. It's not part of a chain, and because business isn't great, we don't have the benefit of a high turnover rate with employment...which usually means shit employees and problems just have to be accepted, rather than replaced.

We have 8 different employees that regularly work in the kitchen, and everyone is supposed to be able to handle all the responsibilities in the kitchen. One person is dedicated to keeping us stocked on cooked chicken tenders. Apart from that, we also have to have wings, fries, and toast cooked (with some additional cooked-on-order foods like catfish, mozzarella sticks, fried mushrooms, or burgers). And then there's a need to plate/box all the food. I've noticed that depending on which 4-5 employees we have present on any given day, the nights can go a lot differently. Prior to clean up time, there are essentially three responsibilities in the kitchen to keep things running smoothly.

We'll call these 8 employees Jed, Mary, Roger, Bessie, Dana, Andy, Tim, and Mike.

Mike is the owner of the restaurant. He only works lunch shifts, and prefers to handle cooking when he's not doing business-related things. He's good about keeping the chicken cooked -and- being able to keep fries, wings, etc. stocked so we don't run out.

Tim is new, he's essentially useless and shows no desire to actually learn how to do anything.

Jed and Roger both like to plate. Although they sometimes cook the chicken, regardless of whether they're plating or cooking chicken, they almost never pay attention to the stocks of fries, wings, etc. So when they are around, someone else has to pay attention to those tasks. They tend to only cook those things when we have completely run out, meaning customers have to wait on their food.

Mary and Bessie generally prefer to cook. When they're not cooking, they prefer to leave the kitchen and work on stocking other things like our baked beans, sauces, and whatnot. Bessie is hated by everyone for her inability to do any of the tasks the correct way, and tends to have the worst-tasting chicken (because the fryer burns and that taste gets into the chicken) so other people in the kitchen with her don't want her cooking. Or stocking.

Andy does all of the tasks well, and focuses heavily on making sure we're stocked on fries/wings/etc. at all times and never run out, while also making sure we're never overstocked so food will get cold (which Mary and Dana will tend to do.)

Dana is the owner's wife. She almost exclusively plates, though she is good about handling the other duties as well when necessary. Regardless of the fact that she is basically the immovable rock of the plating table, Jed and Roger will still try to exclusively plate while she is there (and it really only takes 1 person to plate unless we are super busy). This typically means that if she is there with Jed/Roger, we'll run out of fries if the chicken cook is in the middle of cooking chicken, because no one will be looking at the fry pan.

So, let's take a four names out of that and create an imaginary scenario of how a night would go.

Dana, Jed, Tim, and Roger. You have 3 platers and a useless body. Whether Jed or Roger are cooking doesn't really matter, because neither of them will see that we are running out of fries until we are actually out of fries. Tim will keep us stocked on toast, but other than that will only do what he is directly instructed to do (unless a salad gets ordered, he can do that! What a swell guy.) But because of the lack of attention to sides/wings, this kitchen staffing will likely be behind for the entire night, leading to a lot of frustration from Dana who will be repeatedly waiting on food so she can plate, while Jed or Roger stand over her shoulder trying to help plate only to find there is no food.

Now let's take...
Mary, Bessie, Andy, and Tim. Now you're left with no one that actually plates, unless Andy plates. Tim doesn't really know how. Mary and Bessie are both very slow at plating, and Bessie tends to get confused by the tickets. But, because Mary and Bessie tend to leave the kitchen if they aren't cooking chicken, this leaves you with one cook, Andy plating, and then Tim. Unless Andy continuously tells Tim to drop food, then either the chicken cook Bessie/Mary has to keep foods cooked, or Andy has to leave the plating table and put orders on hold just to keep the food supplies up.

So far, we're 0 for 2 in making a good crew for a night.

Let's take a lunch shift. Which is usually Mike, Mary, Bessie, and either Jed or Roger. Mike's good at keeping everything stocked up while he's in the kitchen. Jed or Roger can handle plating. Great! Doesn't matter if Mary or Bessie peace out of the kitchen, you've got plenty of food, and someone to make the orders. The only problem arises when Mike wanders off to sit at the desk / place truck orders / goes to talk to someone he knows in the dining room (he's the owner, he can do what he wants.). Then you're left with Mary or Bessie to cook, which typically won't happen until you've run completely out of chicken and customers are left waiting.

A good crew typically involves pairing Dana, Mary, Andy, and either Jed or Roger to be a gopher for things away from the table that Dana may need (whether it be slaw/beans/potato salad for side items, or to grab the non-hot/mild sauces for wings, which are stored on a separate table from our steam table). Andy keeps the fries/wings stocked, Mary keeps the chicken stocked, and Dana+Jed/Roger get the plates out in a timely manner.

Without Dana there, adding Jed or Roger to keep plating controlled, along with Andy to keep fries/wings stocked, and anyone assigned for cooking keeps the kitchen running fairly smoothly.

If Mary is there but Dana isn't, Mary usually is not the chicken cook because she's the Manager for the night, which means Mary is usually absent from the kitchen if Dana isn't present (#TheDuo?). If Mary is not in the kitchen, and Andy is not working, then problems begin to arise of running out of fries/wings/etc.

Now... what does all of this rambling about kitchen staff have to do with gaming?

None of us in the kitchen have a specific role. We are all (except Tim) trained and expected to be able to do everything as required. But, because of the tendencies of how people work, shifts can go a lot more easily or a lot harder depending on who is scheduled for the night.

Apply this to Halo. You can say all you want that roles don't exist in Halo (and, to some degree, it is true). But everyone has their preferred playstyle, and tendencies they've developed over thousands of games.

Whether you're a player who likes to sit in power positions wailing on people with your BR, or a player that likes to put 1 or 2 shots and play a rousing game of hide and seek, or a player who likes to focus entirely on the objective, or a player who likes to grab Snipe and go blain kids all have a playstyle. Finding a team of 4 players with playstyles that don't clash greatly improves the way that team can work together.

Right now, among teams competing in the HCS, I'd say there are 4 teams with an actual “support” type player – a player with a selfless playstyle or focused on setting up the rest of the team to succeed while their performance doesn't NECESSARILY look that fantastic (disclaimer: being labeled a support player does not mean you go negative, it just means if you are going negative, your performance is still benefitting your team). Those 4 teams I would label as having a player of that style...also happen to be the top 4 seeds currently.

Str8 Rippin and Cloud 9 both have tremendous talent on the team, but struggle against these others. Part of the blame belongs on the natural chemistry of those teams' rosters. The playstyles and tendencies of those players are too similar, so it's hard to just naturally end up in a proper set-up in an objective gametype. Does that mean it -can't- happen for them, or can't just click and be successful? Not at all! After all, these players know how to do the other tasks...they're just not used to them, and when doing them they're not in a comfort zone with tons of experience knowing exactly which way to juke with the flag, or where to set up to zone people off going for the hill even if they're not going for the hill themselves. They just simply lack the experience from being forced to do those tasks.

Hopefully this blog gives you some insight on why I'm a preacher of the “playstyles matter” school of thought.

Monday, December 22, 2014

Designing Tensai, Part 5: Making the Stars Align for Perfect Stats

So, since I've pretty much beat the core combat system to death for these Tensai blogs, I thought I would cover something a little lower-impact but, in my eyes, equally important to setting up a Pokémon clone for competitive success. And that's individual monster customization.

I already touched on the equipment that can be equipped and swapped to modify and adapt strategies on the fly, but there's an additional bit to the Pokémon formula that competitive players love to tweak, but it's covered in such a horrendous mess of never-really-explained mechanics and bits and pieces of information that were hidden from the player until the most recent generation.

I'm talking Effort Values, Individual Values, Natures, et cetera. In Pokémon, your monster gained “effort points” or “stat exp” each time it defeated another Pokémon, which could build up to boost your effectiveness in that stat. This was a sort of under-the-hood customization option that wasn't really clear to players until Generation 3 where the formula was adjusted to be a stat point for every 4 Effort Points you accumulated with a max of 255 effort points (the formula in generation 1 was the square root of the stat EXP you'd gained, with a max amount of 65,535, so 63 stat points just like the latter system). An individual Pokémon could not accumulate more than 510 effort points (which meant if you got 63 points in 2 stats, you had only 6 points, or 1 stat point left to be acquired).

If you didn't already know any of this stuff, there's a chance you're confused right now. And that's a big problem. The EV system is a convoluted mess that's hard for a player to learn and track. Furthermore, since the Evs in a stat could max at 255, but the point yield stopped at 252, there was an ability for players to “waste” a stat point they could've acquired had they not stopped training a specific stat. This is not true in Generation 6, but the fact that three generations existed with this limitation is a rather depressing notion.

When it came to the competitive scene, EVs were generally used either to boost offense or defense depending on whether a Pokémon was a sweeper or a wall, with EVs being assigned to Speed to reach certain break points to out-speed certain common match-ups the monster may face. Overall, this is a great use of the customization stats – players could choose to risk going second against a bad match-up for an extra punch against match-ups where speed was irrelevant. A meaningful choice had to be made in team building for what the player wanted out of their Pokémon's capabilities when building their EVs.

Then we have Individual Values and Personality Values. These have effects on your Pokémon's stats, appearance, and which ability (passive) it gains. IVs affect which version of Hidden Power your Pokémon gains, and whether or not your Pokémon can truly max out their stats. And yet, if you read the Bulbapedia pages I linked for them... it's an even more convoluted mess than EVs I described above. I'm not even going to attempt to describe how IVs work, because the system is so needlessly complex just for the sake of adding grinding to the single-player game for completionists who want the perfect creatures.

Well, Tensai's original designs were to be a standalone battler; no single-player. The grind is unnecessary, so variations in stats between different creatures of the same type were not a requirement for the game. Leveling up was also not present in my game, so having additional stat bonuses (like EVs) gained from battling other creatures was another unnecessary inclusion. But I still wanted the level of competitive customizability offered by natures, EVs, and the like, as well as the inclusion of a Hidden Power-like move with a variable element based off something other than the element of the creature using the ability.

Fortunately, Tensai was set in a fantasy world I've been world-building for years, and one of the core concepts of that world happened to fit perfectly into what I was looking to do in order to emulate Pokémon. So now it's time for a bit of a fantasy storytelling about the world of Astral Gate.

I mentioned back in the first blog the world had seven elements: Fire, Metal, Ice, Wood, Air, Water, and Earth. In addition to this, there is a duality of the spiritual and the physical, which I borrowed from Plato and labeled Aether and Eidos. Each of these seven elements pair with the duality for fourteen signs of their astrological Zodiac. Some examples are the Eidos Fire sign, a flaming bear known as Guiredaro, the Aether Wood sign, a giant rooster with leaves in place of feathers known as Cockatrees, or the embodiment of terror from the Eidos Metal sign, the Razor, a creature made of sharp bladed edges with the body of a scorpion and the head and aggression of a wolf.

With the existence of this concept, I not only had 14 creatures ready to add to my game, but also the ability to compress EVs and Natures into a single menu option that players could change when setting their team in order to determine which stats were boosted, by setting a critter's star sign. Each Zodiac would boost one stat by a reasonable amount, one by a small amount, reduce one by a small amount, and another by larger amount in my initial designs (there was a distinct possibility these would've been changed, especially stat penalties, which are generally not well-received by players, even with the positive trade-off of gaining the stats they want.)

I mentioned Hidden Power as well, a move that in Pokémon, could be any type depending upon the user's IVs (and prior to Gen 6, had variable power as well). For the sake of porting this move into Tensai, I simply created an attack move called Zodiac that took the element and essence depending upon the star sign assigned to the creature using the move, enabling some creatures to use moves not of their own element (but with a lowish base power). This was great for those who wanted to use a Health Pack or non-elemental weapon but still wanted the additional option of elemental coverage. Whether the move was physical or 'special' (in Pokémon terms) was determined by whether the Zodiac sign was Aether or Eidos as well.

In the event that this system was oversimplified and caused player frustrations between stat customization and the assigned Zodiac move element, I had the option to more deeply mimic the Chinese Zodiac and its use of Inner Signs and Secret Signs (since the primary Chinese zodiac is based off the year in which you're born, while Inner is based off the date, and Secret based off the time of day.) where a player could assign a second sign solely for the purpose of determining the element/duality of the Zodiac attack move.

For the sake of a standalone battler, this Zodiac sign based system (which, admittedly, isn't a unique idea; Final Fantasy Tactics has used it, and I'm sure others have as well.) happened to fill all the needs of the stat customization options of Pokémon with none of the grinding or needless complexity from the original iterations.

That's all for part 5, regarding simplifying the tangled web of numbers and bits and bytes and nibbles of hidden or only partially communicated information that is the Pokémon IV EV PV LV DV R2D2 system of values that create the variations between monsters in the game. Not sure what the next part will include; I still have to talk about the passive abilities in more depth, but I've nearly covered the majority of the core design of the game at this point. Hope you enjoyed the read!

Tuesday, December 16, 2014

Designing Tensai, Part 4: Weapons of Mass Prediction

In the early stages of Tensai's design (well, early is relative, I spent less than a week on the actual core systems design on Tensai before going straight into content design of moves, creatures, and the items), I made the decision to give each creature two equippable items: a Weapon and a Trinket.

Trinkets were essentially Pokémon held items. There were analogous ports such as my own version of Leftovers, Life Orb, and the Choice items from Pokémon, but of course... because of the good ol' trusty ACTION TYPE SYSTEM I've been beating an entire graveyard of dead horses about... there were some unique additions to the system.

The Primordial Switch and Reversal Charm trinket would be announced when a critter holding them swapped into battle, because their effects were QUITE important. The Primordial Switch trinket reversed the elemental hierarchy (so Fire would no longer be strong against metal, ice, and wood, but instead be strong against rock, water, and air.) The Reversal Charm would do the same for Action Types, reversing the flow into the opposite direction.

Pokémon's Arceus “Plates” made their own analog home in Tensai's treatment in the form of Amulet trinkets, which boosted the power of elemental damages by 20%. Of course, since elements aren't the only damage types it meant I also got to make Trinkets for...ACTION TYPES! These trinkets carried a bonus effect in addition to boosting damage of that Action Type: the creature holding that item would enter the battlefield in the Stance of their Trinket. (This Stance ONLY applied on entering the battlefield, not when idling on a turn, so if a Creature had a passive stance already, the Trinket wouldn't overpower the Creature's innate ability.

I tried not to go too heavy-handed with Trinkets, and stick close to the proven designs from Pokémon; with plans to balance or make additional creations/subtractions after playtesting to see what worked and what didn't. With trinkets out of the way, it's time to talk about the other items: Weapons.

I created the Weapon slot as a second answer in addition to Action Types to help alleviate problems of being in an elementally disadvantageous situation. Basically, Weapons were a generic move that is available to all creatures, regardless of their element. There were Rods, Halberds, Shields, Clubs, and Slingshots (to represent Magical, Aerial, Defensive, Melee, and Ranged action types) of each of the seven elements as generic 25 Essence-cost moves.

The general expectation behind the standard weapon choices was that players would look to equip a weapon that fully covered their type weaknesses (A Fire creature would want a Wood weapon to have strength against Air/Water/Rock creatures that counter it.) Of course, with Pokémon moves like Toxic, Substitute, and Rest being prevalent in TMs and able to be learned by nearly all Pokémon, there were weapons like the Kitchen Knife, Decoy, and Panacea respectively to translate those moves into a Weapon option for critters of Tensai, at the opportunity cost of type coverage.

But the buck doesn't stop there for items. I'd added one additional battle command option for players that didn't exist in Pokémon: The ability to swap items between your Critters. For example if you had an active creature that was getting low health (but likely able to survive an idle turn), you could swap your damaging weapon with a Panacea to fully heal and fall asleep for a few turns. If you had a creature with the Flying passive (Aerial Stance) that also relied on Aerial moves for its best damage, and had reason to fear your opponent's ranged moves, you could use your turn to swap your Trinket for a Reversal Charm held by another member of your team, adjusting your strategy on the fly.

If you wanted to bring in a creature with a Choice item in Pokémon, after using the move you are locked into using that same move until you swap out the Pokémon, sometimes losing a type advantage you'd backed the opponent into just because you couldn't undo your move selection. With the ability to swap items in Tensai, bringing out a creature with a Curse of Speed (Choice Scarf) could enable you to use your superior speed to knock out an enemy, then instead of swapping out your creatures, simply swap your Trinket to another member of your team, freeing up the ability to use any of your moves. While it would cost you a turn, you would not forcibly lose any positional advantage you had earned through the use of the Choice item.

The ability to change weapons also gave one other option: to sacrifice the elemental advantage your weapon was intended for to ensure you had Action Type coverage over your opponent's creatures after scouting his moves. Say, for instance, your opponent favored a Magical move for damage, and you did not have a Melee move in your creature's 3 move set, and currently had the Flowing Halberd (Water Aerial) weapon equipped. You could trade the Halberd to a creature with the Stone Club (Rock Melee) in order to get access to a move to negate the incoming damage of their largest threat, either forcing them to use a less efficient move by threat of you having a counter or forcing them into a game of chicken.

This ability to trade items between creatures could allow a player to cover situations their team was not truly prepared for by giving creatures action types that weren't prepared in team creation, and created dynamic customization as the battle unfolded.

I was a bit worried that even with the weapon slot, players might feel only 3 moves per creature was a little underwhelming, so I also made sure to design each creature with its own Signature Move. The giant flaming bear zodiac (I'll go into this when I talk about the world design of Tensai) creature Guiredaro had the ability to use Bear Hug, grappling its opponent and transferring any other negative status effects from the user to the target. Guiredaro was designed around setting himself Aflame (a damage-per-turn status effect) and then transferring that status to the opponent. The other fire zodiac creature, the Firefly, was designed around setting itself Aflame to heal itself, as fire-elemental moves healed it. Its signature move, Burn Up cured negative status effects and THEN set it Aflame.

One of the metal zodiac creatures was a squirrel themed around magnets; its signature move was called MagLev, which inflicted the status effect “Juggled” for one turn – a status effect that tied back into my Action Type system by ensuring if an Aerial move hit the target next turn, it took double damage (similar to a critical hit, only specific to one Action Type). If your creature was Juggled, you had to be extremely wary of an incoming Aerial attack. So you could prepare a Ranged attack...but again, the layers upon layers prediction come forth.

By making Signature moves for each of the creatures, it also adds a specific expectation of what the Action type that creature will use for its primary damage source once a player becomes familiar with the game. If you see a Guiredaro Aflame, you know he wants to Bear Hug you, a melee move. This preconceived expectation of a player's moves helps dictate the flow of an average battle, but as players become more intimately familiar with both the game and one another, it adds inherent depth into the possible interactions.

Combined back with the ability to swap out items, the semi-scripted nature of battle created by Signature Moves helps allow a player know what's coming before it comes and prepare for it by getting their items where they needed to be before they needed to be there. Or after. Whatever.

Regardless, the Action Type horse army has been sufficiently beaten to death, so my next blog on Tensai is going to focus on something else. Not sure yet what it will be. Could be the Astral Gate world in which Tensai is set (a fantasy world I've been worldbuilding for nearly 6 years now.) Might be just the Zodiac alone. We'll see when I get inspiration to write again. Thanks for reading!

Part 2
Part 3

Monday, December 15, 2014

Designing Tensai, Part 3: Hate PP? Urine Luck!

Okay, I'll admit that's an awful title. The real subject of today's blog is the resource systems of Pokémon and Tensai. I haven't mentioned it in the last two blogs, but Tensai wasn't designed to have a single player experience; I was creating JUST the Battler, and because of that, Pokémon's PP system, which was designed to be a sort of dungeon attrition system like a D&D “casts per day” or Final Fantasy 1's system simply wouldn't make sense.

Apart from the moves in Pokémon that only have 5 base PP, it's very rare to actually run out of uses of a move in a battle between trainers unless you're up against a Stall team, or you're playing 6v6 Rocky Helmet Magikarps. The only other exception came if you had a Pokémon with the Pressure ability, but even then, it wasn't that impactful in actually making the casting resource feel like a worthwhile part of the battle to worry about.

Because of this, I wanted a resource system in Tensai that would actually have an effect on the battle. And so, “Essence” was born. Essence basically works like mana systems in today's card games; you gain a little bit of essence every turn (however, essence spent is gone once it's spent, rather than you gaining even more the next turn.)

You would start and cap out at 50 Essence and most basic moves would cost around 10-25 Essence, depending on their base power and their effects, while the more powerful moves would cost 35-50. You gained 20 Essence per turn, and could elect not to attack and instead use the “Rest” command to recharge an additional 10 Essence for a total of 30, which would put you at the cap on the ensuing turn.

Draw a Card, Play Your Opponent
This system fed directly into the Action-Type system (see parts 1 and 2) as well, once you learned your opponent's team. For example, if your opponent only had 30 Essence, and you knew his strongest move was a 35 or 50, he couldn't use it this turn. That's one Action-Type you can eliminate from the realm of possible uses this turn. He can play Rock or Paper, but he can't play Scissors. Better play Paper yourself for the best odds!

If you saw your opponent rest on the first turn after hitting 0, it means he's rushing back to 50 Essence and probably wants to immediately use his strongest attack. If his strongest attack is a Magical Action-Type, ready up that Melee you have equipped and go to town while you take no damage!

From the attacker's viewpoint, it also made a player be wary on when to choose to use his strongest attacks; if you made it too obvious, you could be countered rather easily and use all that essence for naught. You would have to set up your opponent to unexpect the expected in order to deliver your wrath. Keeping smaller cost moves around so you can stay at max Essence and leave that threat of a high damaging move available was an important strategy.

By making the resource system actually impactful on the battle and give soft limitations on what moves could be used at a given time, it gave me as the designer a lot of control over the flow of a battle and more room for the direct Player-versus-Player Prediction interaction granted by the inclusion of Action-Types.

Mix, Mix, Swirl, Mix!
Furthermore, the Essence bar was split between whether the attacks were Aether (Magical) or Eidos (Physical) – if your move was Aether, you used Aether Essence. If your move was Eidos, you used Eidos Essence. Only one minor move in the game took Essence from both bars.

This meant that if you had Mixed-Damage type Sweeper with a powerful Eidos move AND a powerful Aether move, you could use the moves in succession, whereas if you had a Sweeper of just Eidos moves, using a 50-base power move would leave your bar depleted and force you to rest or use only weak moves until you recharged. While Mixed Sweepers weren't necessarily STRONGER than the more focused ones, but because of having their casting resource split into two bars, they could use more powerful moves more often, which could've give them a larger place in the eventual meta-game had Tensai ever been completed.

The separated Eidos/Aether resource system also encouraged players to potentially invest in one of their three moves as a Defensive or Magical Action-Type move of a different cost type (Clarification, Not all Magical Action-type moves were Aether. The two concepts are separate.) than their primary damage stat, even if they weren't mixed. Defensive-type moves almost all dealt a status effect in addition to a mild amount of damage, so it could come in handy having the ability to inflict Unstable or Stunned to an opponent with a Defensive move at any given time. By making this option on your Creature not cost the same resource as your attack moves, you could use it solely for its utility rather than caring about its inflicted damage.

A Balancing Metric, Should I Ever Need One
Last but not least, the revamped resource system of Tensai gave me an additional slider to use to balance moves, if the game had ever been completed. Say, for instance, a move felt like it did the right amount of damage, but its additional effects just seemed to much. But, if you removed any of those additional effects, the move wouldn't feel worth taking.

If they put that in Pokémon, and attempted to reduce its available PP, it wouldn't really affect the player-versus-player metagame too much. But in Tensai, I could increase the Essence cost and greatly affect its availability over the course of a battle.

I used this pre-emptively on Stun moves (read: Flinch from Pokémon.). Any move with a Stun except for a few exceptions costed AT LEAST 35 Essence. In other words, if you used it at 50 Essence, you could only use it one more turn successively before being out of Essence. This meant that you couldn't go Jirachi Serene Grace Iron Head RNG cheesiness to open up a battle. (Of course, since I gutted RNG, the Action-Type system alone nerfed this strategy, but it's still worth noting.)

I also used it on Increased Priority moves to make the majority of them cost over the 20 essence mark to ensure a player could not follow-up a Coup de Grace 50 Essence move with a high priority move; this may have been a heavy-handed over nerf before such a strategy was even tried, but my gut told me it would be necessary. The world may never know!

All in all, the resource system of Tensai was a mechanism I was very proud of how well it fit into the rest of the design of the game, and excited to see how it would be handled once in players' hands. That's all for today's blog.

Next time (maybe tonight or tomorrow), I'm going to talk about why there were only 3 moves per creature (hint: there is a fourth move), about signature moves (not the fourth move), and about trinkets (read: Pokémon held items.) Hope you've enjoyed the read. Stay tuned for more!
Part 1
Part 2
Part 4

Saturday, December 13, 2014

Designing Tensai, Part 2: Playing Your Opponent, Not Your Cards

Last night, I wrote about how design by subtraction helped shape some of the primary mechanics of Tensai, and mentioned the use of a two-layer Rock-Paper-Scissors system in order to allow a very readable, clearly understandable elemental weakness chart while also giving room for those at an elemental weakness to be able to outplay their opponents through several layers of predictions-on-predictions-on-predictions.

I also mentioned this Action Type system gave me more control over one of the most bitched-about aspects of Pokémon from its competitive community: RNG / randomness / “Hax.” In Pokémon, some matches can be turned on their head in an instant by the random chance to inflict a status effect or have a stat lowered, or the bane of defensively-boosted stall teams: a critical hit.

The first two (status effects or debuffs) were easily remedied through Action Types. An attack move was considered “effective” if it was neutral or better against the opponent's move (For example, a Melee move was effective against Melee, Ranged, Magical, and Supplementary moves such as heals or self-buffs.) If an attack with an additional effect, such as a debuff or status effect, was effective, its additional effect was applied. While it's still somewhat random how often status effects would actually be effective, it wasn't decided by the game or some dice or some pseudo-randomized bullshit. It was decided by the players.

Because the effective chance would generally be a bit higher than most Pokémon status effects (apart from Spore / Thunder Wave's 100% hit rate), I tended to make the status effects weaker by comparison, including Stun (Flinch) and Unstable (A stun that takes place AFTER the player uses another attack or is hit by another attack...this would work great in combination with low Priority moves on one turn into a higher priority move on the next turn.) as common effects, rather than paralysis or Pokémon-level burns that cut attack by half.

When Triton (the programmer) asked how I could make critical hits not as game-changing, I stopped for a second and thought, “What is it that makes critical hits so bullshit?” A crit in Pokémon ignores any defensive boosts (treats you as +/- 0 stages, so anyone who's spammed Defense Curl or Harden doesn't have an advantage anymore) and deals double the damage. But the defense-piercing effect isn't what made it so frustrating.

It was the unpredictability and the abruptness of the effect. You never know if or when your opponent's next hit will land that lucky 6.25% chance to crit. If your strategy is built around stalling out, odds aren't entirely unrealistic that a critical hit could happen. But it's random, and you can't tell when it is coming. Oftentimes, a crit will completely wipe out the wall you were using to stall, and leave your entire strategy in shambles and you with a very slim chance at salvaging victory. Basically, you were playing against the game, not against your opponent. And that's bad.

So, how do you address that problem? Well, it's a turn-based game. My answer was to simply make the bonus damage effect take place a turn after the critical hit. Rather than a critical immediately flipping the board over in chaos, it creates a tipping point. The board can be flipped, but it is not a guarantee. The game is thrown out of balance (coincidentally, I'd renamed the “Critical Hit” into “Knocked Off Balance” to reflect that the creatures were more vulnerable for a turn.), but the player who was standing tall on the jousting platforms (man I miss American Gladiators) wasn't necessarily knocked into the pit.

By making the critical hit damage take place on the move used the turn AFTER the “critical hit” effect happens, what changes?

  1. The player on the attacking end knows that, their next attack is going to deal a LOT of extra damage.
  2. The player on the receiving end knows that they have to do everything in their power not to let that extra damage take place.

Since the damage bonus is tied into a status effect, the player on the defending end has the option to simply swap out their creature and take regular damage. If this system were applied to Pokémon, the interactions from this situation would be limited, but still add a layer of Yomi – if you predict the player wants to swap out of the damage, you have the advantage in guessing which Pokémon/type they are going to bring out next and can either use a move with a type advantage, or swap to a Pokémon that will have an advantageous match-up. You also have the potential of using higher priority moves (or regular moves if you're faster).

But this isn't Pokémon. Even if you're knocked off balance, the inclusion of Action Types still offers you a manner to have your active creature take zero damage in the next turn. For example, if your opponent's strongest move is a Melee type, and you have an Aerial-type move, and you predict your opponent is going to attempt a coup-de-grace, you can choose to use an Aerial-type move to keep your previous advantage. But he can predict your prediction and use an otherwise-less-optimal decision to go for more likely damage; for example, choosing an Aerial move as well to go for better coverage of your defensive options.

Some creatures also had passive abilities called Stances that made it so upon them entering the battlefield (or using a supplementary move) they were treated as using a specific Action Type, such as birds in the game having an “Airborne” Stance that allowed them to be treated as using Aerial moves when no Action Type was set by their selected move. Because of this, a player could use being knocked Off-Balance as an opportunity for a free swap-in, if their Stance matched up well against the predicted move. You're gonna use a Ranged move to try to capitalize on my off-balance creature? I'm going to swap in my Tortoice with the passive, “Defensive” so he takes no damage as he comes in.

So thanks to the inclusion of Action Types, my “delayed effect” solution to critical hits is given even more depth than it would've had if plugged straight into Pokémon (where it'd still be better than the current iteration). The player “screwed over” by the random chaos of a critical hit is still given some degree of control of what happens next. They're not playing against the game that decided the critical hit happened...they're playing against the opponent who decides what happens next.

And that's all for Part 2 of Designing Tensai. I'm not sure what or when Part 3 will be. Just stay tuned!
Part 1
Part 3
Part 4

Friday, December 12, 2014

Designing Tensai, Part 1: Take Up Your Chisel

A while back, I was pretty gung-ho about making a Battler (basically, a Pokémon clone). A friend had expressed interest in cloning an existing online Pokémon battler in order to practice his programming skills, so I took the opportunity to pitch an idea for a new game, in the vein of Pokémon, designed to improve on what I viewed as some flaws in the formula. Although the game got abandoned as the programmer got busy on other projects, I still dwell on the design and thing, “If I ever made this game, it would be fantastic.”

Don't get me wrong; I have no grandiose ideas of knocking Pokémon off the throne Nintendo's used Charizard's flames to forge together among the many cartridges of the other mobile franchises that have fallen before it, but I definitely see flaws in the franchise's design that irk me as a designer.

I've been pondering writing a blog series for a while about my mindset in designing and addressing Pokémon's issues / designing Tensai, but it wasn't until reading Hardy LeBel's recent blog about what he views as the Universal Truth of Game Design #1 that I really felt my approach was truly correct.

So Design Part 1: The Elements

Now, if you took the time to pause and read LeBel's blog, you saw that he said the second most important tool in a designer's kit is subtraction.

If you didn't take the time to pause and read LeBel's blog first, go back and do so. He says some seriously cool stuff. In fact, I encourage you to go read his other posts as well. Especially if you play Halo.

Back on topic: Subtraction.

One of the biggest flaws to the Pokémon formula in my opinion is the convoluted type strength/weaknesses the different elements all have. Sure, it's cool to have x be strong against y, and d/f dual type being 4 times weak to y while d/x dual type is neutral. But it's convoluted and needlessly complex. The player can't be expected to learn the type interactions through a normal play-through of the game, and there's too much information for it to be displayed in the game's UI in a clearly readable manner.

So when I started working on Tensai, I borrowed from a fantasy world I've been world-building for some time (which borrowed heavily from both Eastern and Western influences)...where there were simply seven elements: Fire, Metal, Ice, Wood, Air, Water, and Earth. (The keen of you will notice those are literally the 4 Classical Elements with the inclusion of Chinese elements Wood and Metal...and then Ice.) The interactions followed a simply RPS7 system; each element was strong against the 3 that followed it and was weak to the 3 that preceeded it.

By trimming the number of available elements from 18/17/15 (depending on generation) to 7, you immediately make it much easier for players new to your system to tell what is strong against what. Fire, for example, is strong against Metal (melts it), Ice (melts it), and Wood (burns it)...while being weak to Air (Blows it Out), Water (quenches it), and Earth (Smothers it / Doesn't Burn). The type interaction chart is so simple it could be displayed with a single chart featuring each of the seven elements with 3 lines of the same color going from each element's symbol to the three it is strong against. Your player base is now given a clear idea of how the elements work, which can be communicated even within a battle.

In addition to limiting the number of elements, I also made one mandate to go along with it that made the design very, very simplified compared to Pokémon: A Creature's base moveset can only include moves of the same element as that Creature. A Fire creature can only use Fire moves, An Air creature can only use Air moves, and so on.

But wait, Audley! I see a problem with that!

Oh, do you now?

Yes! Won't that mean the elemental type advantages are too steep to overcome if you switch into a bad match-up?!

Clever girl, you have found a weakness...and the second major design decision I made on Tensai...which came about as a domino effect of removing a large chunk of Pokémon's clutter.

Design Part 2: Action Types

In Pokémon, moves have their own “attack” element which interacts with the creature's element or elements. Say, for example, you had the Fighting-type Pokémon Hitmonchan. He wasn't limited to Fighting-type moves, thanks to Fire Punch, Ice Punch, or Thunder Punch giving him attacks of those elements to allow him to better cover his weaknesses (for example, Ice Punch or Thunder Punch would enable him to beat up any Flying types that came his way if he was fast enough.)

Since I'd removed this possibility in Tensai, I had to compensate for the aforementioned problem where there was no ability to cover your type weaknesses. And that is where Action Types came in.

I created five “Action Types” that an attack move could be quanlified as. Melee, Ranged, Magical, Aerial, or Defensive. These five also interacted in a sort of Rock-Paper-Scissors-Lizard-Spock format, but not in as straight forward a manner as the RPS7 chain of elements.

Melee attacks were strong against Ranged, but super effective against Magical.
Ranged attacks were strong against Magical, but super effective against Aerial.
Magical attacks were strong against Aerial, but super effective against Defensive.
Aerial attacks were strong against Defensive, but super effective against Melee.
Defensive attacks were strong against Melee, but super effective against Ranged.

Instead of just improving the damage your creature dealt, however, the Action Types did something better: they reduced the damage you took from the opposing creature. Strong = Half damage. Super Effective = No damage.

So if my Fire Creature were out against your Water creature...I would be at a disadvantage. However, if I predicted you were about to use a Ranged move, and my creature had a Defensive move in his moveset, I could use the Defensive move to take no damage, while dealing a slightly improved chunk of damage to your creature.

Since creatures only have 3 moves (technically 4, but I'll address that later) in Tensai, you can use process of elimination to get a reasonable guess of your opponent's moves and attempt to rock-paper-scissors your way to victory. It also leaves room for some Sirlin-loved Yomi Layer 3 in terms of “Well he should do this move for the most damage, so I should use this type of move to negate its damage...but if he predicts I'm going to do that, he may use this move to counter my counter...” et cetera, et cetera. In the few instances of practice battle runs I ran utilizing nothing but a Skype chat room and a PHP-coded calculator, the four of us who tested it all enjoyed the new layer of depth created out of necessity to coincide with the design-by-subtraction sledgehammer taken to Pokémon's element system.

In my next blog on Designing Tensai, I'm going to cover more in-depth how I approached the largest problem in competitive Pokémon, “hax” (RNG) in detail (I touched on it in the past in my blog Fighting Chance.) – but thanks to the Action Type system, I was able to have even greater control than would really be able to be tuneable otherwise.

So stay tuned for Part 2. Don't worry, this one will be updated much, much sooner than my Calling the Shots blog! And thanks again to Hardy LeBel for the inspiration. Seriously, go read his stuff:

Part 2
Part 3
Part 4

Wednesday, September 10, 2014

League of Legends Season 4 World Championships Predictions

League of Legends Season 4 World Championships are coming. Earlier this week, the four groups for the preliminary stages were revealed along with how the seeds will feed into the bracket. So with that said, it's time to make my predictions!

I'm going to be practicing a more comedic, satirical style of writing in this blog, so...if any jokes feel forced, that will explain it.

Without further ado...

Group A: Edward Gaming, Samsung White, Dark Passage, and ahq eSports Club.

Let's be real here, Dark Passage are Thanksgiving dinner in this group. Samsung White are licking their chops and preparing to dig into the roast Turkey presented on a platter to them by Riot. EDG are leaning back in their seat and kicking their feet up onto this Ottoman Empire wannabe, knowing the poor five wild card players will give them about as much of a chase as Franklin Delano Roosevelt.

AHQ probably celebrated when they saw they were drawn into a group with an IWC team, but that Westdoor got slammed shut the moment White and EDG were roped into the party. Unfortunately, there's no sipping GreenTea for ahq – the #2 seed (and generally regarded as best all-around team) from Korea and #1 seed from China who have dominated the last two seasons of LPL will likely end AHQ's field trip to Taiwan prematurely and send them packing back home to Taiwan after Group A's dust settles.

The big question of Group A comes down to EDG versus White – and while most pundits are giving White the for-sure victory, I'm here to offer a dissenting opinion. And unlike the few voices that are saying “The fix is in.” and claiming White will throw the matches to avoid their abusive big brother Samsung Blue who've knocked them out of the last two OGN seasons...I'm here to say EDG will win the group legitimately.

White somewhat rely on their bot lane to succeed. Imp struggles when he plays against ADCs better than him; just look at his results against Piglet during SKT's hot streak (before Piglet lost his touch) and his results against Deft, who is stronger in lane and team fights. Namei is a better ADC than Imp. And although Mata and Dandy are fantastic at vision control; vision control's value is devalued if you're just being fought constantly. Your gold gets wasted if your opponents don't care you see them and engage anyway. Sometimes I wonder if the Chinese understanding of wards is “That's gold that isn't going toward my ability to tower dive.”

I think EDG's superior team fighting and stronger ADC (I won't say stronger bot lane; Mata is in my eyes the greatest Support player in the game of League of Legends.) will empower them to take first in a shocking upset of the group. I may be the only one with such a faith – but I also said OMG would take their first game off SK Telecom T1 K at Season 3 Worlds after certain analysts insisted China's meta was massively behind the rest of the world.

So, for Group A, my results:
  1. Edward Wong Hau Pipelu Tivruski IV Gaming
  2. Mind over Mata, Samsung White
  3. We're About as Good as a Team from the West-door, AHQ.
  4. Does DP stand for Double Penetration? Dark Passage

Group B: Team SoloMid, StarHorn Royal Club, Taipei Assassins, SK Gaming

This group has the best parity of any of the groups there are. But as far as displaying consistency over the course of the season, this group is a barrel of monkeys. It may end up fun, and if you leave them in Taiwan long enough, maybe they'll produce Shakespeare, but this is going to be the brawl group.

SK Gaming looked good at the European regionals, and despite getting 3-1'd by Alliance, they put up a strong fight in their first two games. But that doesn't change the fact that their ADC is literally one of those comically oversized magnets Wile E. Coyote uses after filling the Roadrunner full of metal pellets mixed into his bird food. CandyPanda's ability to dodge a skillshot is about equivalent to the mother in What's Eating Gilbert Grape's ability to get out of bed. When you're in a group with mechanical greats like Bjergsen and InSec, you simply can't afford to make those mistakes. And with nRated's Crucible allergy (is he a fucking vampire or something?), CandyPanda's positional and dodging woes are compounded further.

TPA aren't getting much respect either. And after their performance at All-Stars this past spring, not even Rodney Dangerfield can complain. Winds looked more like a leaf on the wind – so moving forward he may have a difficult time Washing that performance from his record. Winds may have channeled S3 Worlds Mata at All-Stars, he's still known for being a fantastic jungler in the GPL. Combine that with Bebe's innovation in the ADC spectrum – where he's popularized the Greased Up Deaf Guy Ezreal build (Fuck you Iceborn Gauntlet) and the Why-Can't-I-Hold-All-These-Blades BoRK+Ghostblade build (even before ADC itemization revamp) on Twitch, Lucian, and even Draaaaaaaaaven. If that's not enough, TPA will have throngs of hometown fans bringing the noise to give them that last little edge they may need to advance through the group.

Sometimes, regardless of what the current roster is, people say Dyrus is the glue that holds TSM's chances of winning together. He's consistent, he's solid... But he's totally going to get rattled by that TPA crowd. I mean, when you're used to everyone worshipping you for your globally mediocre play, then having a massive crowd chanting against you is a terrible thing for your mental game. Bjergsen and Amazing are known to tilt as well. Unless LustBoy can quickly implant his Korean Robotics into the rest of TSM before Worlds kicks off, TSM could be in trouble. Lord Loco must teach them how to steel their resolve. Crowd noise, even with noise-cancelling headphones, is a very serious thing.

I saved the best for last. Take one of the most notorious ADCs from Worlds last year. Give him a Korean jungler and a Korean support. Congratulations, you now have StarHorn Royal Club. Zero may have been just that with the KT Rolster organization, and InSec may have been short for InSecure Job...but the two have found success in securing a trip to Worlds in their new region. InSec's balls-out overaggressive playstyle is right at home in the land of “Go” (the verb to advance, not the board game). Royal Club will out-skill and out-play the teams of Group B with ease to earn their place as kings of the group.

I honestly think this group is hard to truly predict, since TPA will be underestimated and TSM are still growing their synergy with LustBoy, but my predictions from Group B are:
  1. KT Rolster Chu-ko-Nu, StarHorn Royal Club.
  3. “Why are they pronouncing S-M wrong in their chants?” Team SoloMid.
  4. Caught-With-Their-Underpants-Down, Schroet Kommando.

Group C: Samsung Blue, OMG, LMQ, Fnatic

Let's get one thing out of the way, Blue will likely run away with this group. And by likely, I mean they came into Worlds holding an Ace and King of Spades, and the flop revealed a same-suited Ten, Jack, and Queen. If Blue drops a game, I will be shocked.

For the rest of the's not as easy to call. Let's start with LMQ. The LPL rejects that retreated to North America to pick on some easier competition and dragged down the former Worlds Semifinalists Godlike with them – dragged him down so much he lost the rights to the name Godlike and simply became Ackerman. Vasilii and XiaoWeiXiao treated their lanes all Summer Split as avenues to collect the US's debt back for Mother China. But all that said, even a mid-tier team from the LPL ended up unable to emerge atop the North American LCS, falling to third in Playoffs. They've got good team fighting and consistently good-not-great solo lanes, but compared to the Eastern regions, LMQ still don't show any sign of being able to compete with their former superiors.

Speaking of third place, OMG of LPL! Cool is still ridiculously good. Gogoing showed no fear in Chinese regionals. The two, combined with jungler Loveling, are a really great trio of players. And then you have this enormous anchor by the name of Dada7. If you think Kiwikid is good at getting caught alone and dying... Dada7 has perfected the art. At this point, I'm pretty sure his coach has mandated him to get the Nike Swoosh as a tattoo across his face so he will forever be known as “The Facecheck” – his penchant for dying for no reason leaves his otherwise perfectly fine ADC unable to get his job done.

And if your bot lane's weak, that gives an opening to a team whose solo lanes are nothing special, but whose bot lane is getting vastly underrated on the global scale – Fnatic. Rekkles is the real deal. This team of mostly Caucasian Adonises (disregarding sOAZ) managed to draw themselves into the best pool for their chances of making the playoffs. Rekkles is proof that Swedish genetic engineering is a booming industry – as his mechanical prowess from the ADC position is unmatched in the west. If Fnatic decide to play Benny Hill and give the Chinese teams the runaround with multiple teleports across their team, they can live to fight another round. If they take the Chinese bait of team fights, it will spell their tournament demise.

This group depends entirely on what playstyles the teams decide to exhibit...but if the teams play to their own strengths, rather than to the meta, my picks for the group...
  1. Don't be Sad, Samsung Blue
  2. ratdoto is best doto, Fnatic
  3. OMG it's OMG
  4. “Time for a roster change, oh wait, we can only pick up Americans RIP” LMQ

Group D: NaJin White Shield, Cloud 9, KaBuM! e-Sports, Alliance

I don't think a team could be more aptly named than KaBuM. They're going to get blown up in this group. Obliterated. They'll be lucky to escape the event as anything more than a smoldering mess of eviscerated body parts. Their ADC may be named for the Roman goddess of wisdom and war, but they're more likely to be a display of her birth – a split skull. Seriously, the only wisdom they'll bring to this tournament is the acceptance that they'll be truly blessed by gods to take even a single victory.

Cloud 9...known for Meteos' jungling and Lemon's notebook. The notebook will do them wonders against NaJin Shield, since Ggoong only plays 3 champions ever, but how do you prepare against a Froggen? Cloud 9's chances aren't abysmal. They know how to close out games and get vision control after taking a lead – a talent SK Gaming lacked that cost them in two games against Alliance in the European regionals. On paper, Cloud 9 actually have a decent chance in match-ups against both Alliance and NaJin White Shield. In practice, however, their questionable mechanical prowess in the lanes will probably spell a few sad doodles in the LemonNote.

Alliance have been begging Riot to introduce a new map with only a single lane all season – so the deadweight of Wickd gets mitigated and Tabzz/Nyph can realize they don't really contribute that much to Froggen's victories. A large part of Alliance's success in the latter half of the season was strictly off Shook using AoE CC while Froggen did burst damage with Xerath ultimate from off-screen. Set 'em up, knock 'em down. If Wickd can avoid becoming the star of his very own snuff film laning against Balls and Save, and if Alliance can get enough defensive wards to keep Meteos and Watch from placing deep wards, Alliance can set up Froggen to shine. Tabzz isn't a pushover bot lane, and he+Nyph'll have a good match-up against C9's bot lane (if C9 doesn't lane swap), so having pressure taken off of him by the Threat of Froggen will play into Alliance's favor.

Shield are the favorites to win the group. With Save in the top lane and Watch setting him up for success, it's no surprise. Shield played phenomenally in the Korean Regional gauntlet. But they aren't without their weaknesses. Zefa hasn't looked impressive for an ADC in the strongest region; whether or not that means he'll look weak against Tabzz or Sneaky is still to be determined. Shield's vision control is only outshone by Samsung White, and that will be a large part of their road to victory against both Alliance and Cloud 9. If they can secure a comfort pick for Ggoong, Shield is set.

With that Group D predictions:
  2. Totally Not a Racist Team Name, NaJin White Shield
  3. No Longer On, Cloud 9.
  4. At least Tristana still cheers for us, KaBuM! E-Sports.

So that leaves my bracket as:
Blue vs Shield

White vs SHRC
Fnatic vs Alliance


Welp, we've left the home field for TPA. Now we're on a neutral field, and they're outmatched here. NaMei vs Bebe will be an interesting match-up, but nothing else about this quarterfinal will be exciting. Should be a 3-0 or 3-1 in EDG's favor.

Blue vs Shield

Ouch, no All-Korean Top 3. Blue simply out-match Shield in the late game and Shield's not strong enough early to force Blue into a situation they can't come back from. Blue 3-1.

White vs SHRC

While I mentioned EDG's bot lane outmatching White is going to be a problem for the Samsung hopefuls, I think Mata will be much more prepared for SHRC than for EDG; he has experience against InSec from Champions Spring 2013, and so InSec's prowess on the blind monk will be countered by the All-Seeing Mata. White won't be caught off-guard by SHRC's playcalls. White 3-0.

Fnatic vs Alliance

Alliance will repeat EU Regionals. Fnatic will be sad. Alliance 3-0.

EDG vs Blue

Blue are much better at team fighting than their sister organization, and much better at mitigating early deficits. They aren't a team you can write off if you fall behind, and...the one weakness I think Blue actually has (Dade's champion pool) won't be taken advantage of by the Chinese teams. Blue 3-1. I think Namei will have a powerful performance in one game.

White vs Alliance

This will be an interesting match-up of styles. But White's vision control and tactical superiority will drown out Alliance's chances. Looper vs Wickd will enjoy their island, but Imp will overpower Tabzz, Mata will overpower Nyph, Dandy will overpower Shook and PawN will not be strong-armed by Froggen. White should 3-0.

Grand Finals
Blue vs White

White's mental block continues. I'm not sure what White's problem against Blue is; whether Imp walked in on Deft giving it to his girlfriend or what, White just doesn't seem to show up when they play their sister team. Both teams deserve to be in this final, but I have to give it to the superior team fighting team; Samsung Blue take Worlds.

I lost my creative energy near the end...hence the absence of humor mid-Group D. Oh well. I'm not a humor robot, yet. Please give me feedback regarding the writing; I want to improve my ability to write satire.