Isle of Arrows

Isle of Arrows offers an interesting take on the tower defense genre, but a dependence on good RNG to survive pulls it down.

I don’t hate Isle of Arrows, but I can’t recommend it. It has a nice art style, and interesting core mechanics. But the reason the game took me 12 hours to beat is that randomness plays too heavy a factor. Having good runs is difficult, and having satisfying runs almost impossible.

Very minor quibble: the situation present here in the game’s Steam banner can never happen in game. And now you will share in that suffering with me.

Isle of Arrows is a tower defense tile placement game. You start with a single pre-placed arrow tower, a few pre-placed road tiles, a bomb and a bridge.

Your job is to turn this.

Into something like this.

Game Mechanics and Systems

You have 4 resources: Life, Gold, Bridges, and Bombs. If enemies reach the tiny gold octahedron at the end of the route, you lose a life. Enemies themselves will always spawn on the last road tile on a path to your octahedron.

Gold is the only real spending resource. Each round, you’re given a tile for free that you can place. You can either place that tile, or skip it. Then you can spend 2 more gold to draw the next tile. Repeat until you’re out of gold. For example, if I had 10 gold, I could place my current tile, then advance choosing to place or skip up to 5 more tiles.

It’s important to note that you can only ever see the next single tile that you’ll draw. While enemies can drop gold, your primary source of gold is income. It’s almost interest: each round, you’ll get 1 gold, plus 1 more for each 10 gold you have stored. (Up to a max of +4 at 30 gold banked.)

The tiles you draw have placement requirements, though. Tiles must be connected to another tile you’ve already placed, they can’t overlap with existing structures, and they have to be placed on ground.

Which brings us to the last two resources: Bombs and Bridges. They both serve a similar purpose in allowing you to break some of the rules. Bombs can be spent to place a tile on an already existing tile, and bridges can be spent to allow you to build out into the air. While there are ways to get more bombs and bridges, there’s no way to guarantee it.

There are a few more tile types. There are roads that can be placed to expand the path to your shiny defense shape, and there are towers that can be placed to shoot enemies as they travel along the roads. There are also traps, which have some effects on enemies placed along the road, and non-combat tiles such as water that often come attached to other tiles. Finally, you can place economy tiles to generate resources, usually as a one time effect based on where you place them.

Isle of Arrows has a beautiful isometric look. It’s fun to watch your island expand! But Isle of Arrows has a big fundamental problem: the randomness.

The Problem

At its simplest, the problem boils down to a question of DPS. There’s no guarantee that you’ll draw enough towers with tiles to have enough damage to clear late game waves. But more importantly, there’s no guarantee you’ll see any specific tower. So it’s hard to strategize. You can’t build or plan with the expectation of “Eventually I’ll get X.” You can play 40 rounds, and never get a ballista. Or you can get 3 sniper towers in the first 10 tiles, and zero in the next seventy.

As a result, I couldn’t set up and build to do anything clever. Sure, I could try to to build a long straight road that would be cleared with a boulder tower, but I’d often never see one. I could set up a winding route that would be perfect with ballista, but if it didn’t show up, that route was just a waste of space.

As a result, my strategy was pretty much always the same: pool gold until I got to 30 banked. After that, I’d just place tiles wherever they provided maximum utility in the moment. There was no grand strategy, or clever synergies. Instead, it was just a fight stay ahead of the curve.

I think this is why my brain feels a bit melted after playing Isle of Arrows. I never really changed up my core strategy. Instead, I just tried to have any tower I placed cover as many tiles as possible, and hoped for the best.

I think a peak example of this randomness was visible in my clear of the game’s final level, Burning Embers. Burning Embers has a fire theme, and a tower called the cluster tower. Cluster towers get a damage bonus for each other cluster tower they’re linked to. But they’re fairly rare, and in my first several attempts, I saw either one or zero of them.

But on what would be my final run, I lucked into an early cluster tower, and then several more later on. Those cluster towers made the difference, bringing down the final tanky enemies, because of that cluster damage bonus.

But there was no way to build or plan for this. There was no grand strategy. It was just “Hey, you got lucky on the rolls, here’s some good stuff.”

And more irritatingly, there are no systems to compensate for the inverse happening, and just getting screwed by RNG. Random draw is the only way to get towers and tiles. You can’t choose to spend an exorbitant amount of money to buy a single specific tower. You can’t choose to modify the pool you’re drawing from. You can make all the right choices, and still lose.

Getting Kobyashi-Maru’d isn’t fun, but it’s made even less fun by the fact that there is really only one strategy. As I mentioned before, it boils down to just getting as much value out of tiles on the turn you draw them.

Isle of Arrows is nicely polished. I want to like the random tile system for an interesting take on tower defense. But the randomness and capriciousness makes it feel like there’s no point to trying out multiple strategies, or trying to be clever.

Give me ways to mitigate being screwed by your system! Let me pay a premium to bring in certain towers if I can pool the money. Or do something akin to Loop Hero, where certain arrangements of tiles and towers morph into a larger super tower structure. Give me some form of decision making that isn’t just pushing my luck, or maximizing current utility.

Isle of Arrows is $13 on Steam, at time of writing. Any desire I have to continue playing the game is mostly out of a sense of spite, and to vindicate my own opinions. I’m confident I can clear all maps with all factions. After all, I’d just need to get lucky.

I’ll leave it to you, dear reader, to decide if that constitutes a endorsement.

Author’s Note: I didn’t really mention the faction mechanic. There are five factions outside the starter faction, they all change up the gameplay slightly. They usually make it harder. Architects don’t get flags to expand the island, but gets bridges after each wave. Cannoneers don’t get flags, but convert any extra life or bridges into bombs. Pathfinders can build roads both ways, but get swamped with road tiles. I’ve got two problems with factions. First, the general high level strategy doesn’t change per faction. Second, each faction is usually more difficult than the preceding one. As a result, clearing each map with an additional faction just feels like grindy busywork instead of an interesting choice of challenge, or a fun variant to try out on a map.

TLDR: There’s a faction system, but it feels like it exists to pad playtime, not make it more fun.

PAX Unplugged – Mythic Mischief and Klask

Mythic Mischief and Klask don’t really have anything in common with each other. It’s not even like they had booths next to each other or something. Mythic Mischief is an action economy and movement-based game with victory points that almost reminds me of Chess. Klask is a skill-based dexterity game that feels like miniature air hockey.

So why am I covering them together? Because I don’t have enough to say about them separately to fill writeup! Anyway, let’s get to it.

Mythic Mischief

Mythic Mischief is an asymmetric grid-based movement game, designed by Max Anderson, Zac Dixon, Austin Harrison, and published by IV Games.

The best summary I can offer is that you and your opponents both control 3 miniatures on a 5v5 grid. Alternating turns, you attempt to spend your actions and use your abilities to place your opponent’s units in the path or directly on an NPC unit called the Tome Keeper.

Editor’s Note: Tome Keeper not to be confused with Dome Keeper

At the end of a player’s turn, the Tome Keeper moves towards specific locations. If there are units in its way, the Tome Keeper knocks them out, and the player who didn’t control those units scores points. Units that get knocked out can be replaced at the start of the next player’s turn.

There’s a fair amount to the movement and action system, and how it plays with the game’s upgrade choices that I don’t think I can summarize effectively, so I won’t try. It’s a perfectly fine system, but I would not describe it as “Sparking Joy,” at least for me.

It is worth noting that each player will be playing a different faction, with unique abilities and so keeping track of what your opponent can do is necessary to succeed.

I only played one game of Mythic Mischief, and it was a combination of a demo and an ass beating. I wouldn’t say that I hugely enjoyed it. That might have been because I lost, and because I get salty easily. But I also struggled with two other factors.

First up, just because of how the game works with scoring, it felt very difficult to make any sort of comeback once I fell behind. Secondly, the game reminded me of Chess in that it felt like a game of trying to find the “Correct” moves, and like a puzzle of chaining things together. That’s just not something I find very fun.

So yeah, if you do like deterministic movement games, or things like Chess, maybe you’ll get more out of Mythic Mischief than I did.

Klask

Klask is a manual dexterity game by Mikkel Bertelsen.

Honestly, it feels weird to be reviewing Klask here. It’s as if for some reason I felt compelled to write a review of Skeeball, or Soccer. The closest games I can think of as a comparison to Klask would be Air Hockey or maybe Foosball.

Those chips look really good.

All of this to say that the “Manual Dexterity” part of the game is absolutely not optional. Klask is played in an elevated square wooden box with sides. Each player has a magnet with a stick in the end that they hold under the box, and a pawn they place on top. The top pawn is moved by dragging it with the magnet from under the box.

The pawn and stick aren’t the only magnetic pieces, though. Klask also has 3 small plastic beads with magnets in the center that are placed equidistant in the middle of the playfield at the start of a point. These beads will jump and stick to your pawn if you get too close, and if 2 of the 3 stick to a player’s pawn, their opponent gets a point.

Points can also be scored by a player hitting the ball into the goal indent on the board, or if a player messes up and gets their pawn stuck in the indent.

The interesting part of Klask for me is how the tiny white beads open up strategy. Without them, the game is pretty much just air hockey with a marble. But with the beads, you can do interesting stuff like hitting them towards your opponent in order to close off parts of the board.

Overall, I like Klask. I just don’t like it enough to really want to buy it. That said, if someone asked me if I’d play, my response would be a semi-enthusiastic “Sure!”

Conclusion

I don’t think there’s any meaningful conclusion you can take out of things like both Klask and Mythic Mischief being present at PAX Unplugged. Maybe there’s some sort of testament to the diversity of mechanics and games present. Maybe there’s something to be said for the sorts of games you’d play if someone else is footing the bill.

And maybe there’s nothing. Maybe there is no purpose. Maybe the real journey was the friends we made along the way.

If you want more nonsense and to be notified whenever I write new stuff, maybe consider following me on Twitter? The site still seems to be up and functioning, at least for now.

Dome Keeper

Dome Keeper is a small and solid game, but didn’t offer enough variety in runs to keep me hooked.

Author’s Note: Not big on reading writeups? Why not just watch me play the game here?

Dome Keeper is a score-attack mashup of Motherload and Space Invaders. I think it’s a good game. I don’t really recommend it, for reasons I’ll get into in a moment. I know it sounds weird to call a game good, and then not recommend it, but I promise it will make sense in a bit.

Anyway, back to Dome Keeper. You have a Dome, which functions as your little base. You venture out from it into the earth to mine minerals. And this dome needs to be Kept. Specifically it needs to be kept from being shattered into a million pieces by various spooky shadow monsters that show up in waves on timed intervals. This is the game’s core tension: mine resources, drag them back to your dome, and try not to get caught at the bottom of the mineshaft right as the next wave comes in to smash it.

You defend the dome with its weapon systems. There are two dome weapon base sets, the Laser Dome, and the Sword Dome. The Laser dome plays somewhat like a turret defense game. You have a big laser, you can rotate it alongside the outside of your dome, and you press another button to fire. The laser moves slower when fired, so it’s faster to move it into position, and then fire the beam. It has various upgrades, including moving the laser head faster, having the laser deal more damage, etc. Honestly, outside of a double laser upgrade, there’s nothing very exciting.

The Sword Dome is unlocked later. Instead of having a projectile weapon, it has a large sword that can be swung back and forth across the dome. It can also be launched like a harpoon to skewer long ranged projectile using enemies, or even to just tap melee enemies a bit before they reach the dome.

Personally, I very much think the game was designed with the sword dome in mind instead of the laser dome. Certain enemy behaviors and patterns interact in a much more interesting way with the sword than with the laser. As an example, one of the earlier enemies is a small bat-like creature. It flies onto the screen cloaked and unable to be hit, flies to an area on either the left or right of the screen, uncloaks, shoots a few projectiles, then recloaks and flies to the other side. Rinse repeat.

With the Sword Dome, there’s an element of skill to this. It takes the same amount of time to uncloak every time, and you can one shot it once you get a damage upgrade. There’s a sort of elegance to predicting where it’s going to be, pre-launching the sword, and steering it into the bat right as it uncloaks.

Another good example can be seen in the later game enemy, the launcher. It’s a large blobby snake that swarms out of the ground, waits for a moment, and then launches a large shadow projectile through the air. With the laser, there’s no real option other than to just blast it down. But hitting it with the sword before it launches the projectile will stagger it and force it back down into the ground.

However, both of these have the same problem. Despite having multiple options for upgrades and changes, there’s no real reason to experiment during a run. Enemies just show up randomly over time, so instead of building for a certain encounter or fight, it felt better to just do the same build each game, and play through. The end result? The fights kind of just feel all the same. Ramping intensity and difficulty, sure, but not changing how things feel mechanically between runs, unless you choose to take a risk and force it.

That’s only half the game, though. The other entire half is mining and digging for resources from under the dome. If you’ve ever played Motherload, this will feel familiar. If you haven’t, it works like this. The keeper is controlled with cardinal directions and will automatically mine walls and blocks if you walk the keeper into them. Different dirt has different strengths, but as you get deeper, the strength just increases overall. This means that it can be easier to mine deeper into weak dirt than to try to dig out stone at your current level. But generally you’ll need to get upgrades to go much further.

The goal of all of this digging is to get resources, of which there are three. Sulphur, Water, and Iron. Sulphur is the rarest, and is used to repair your dome’s health, and buy a single set of special resistance upgrades. Water is used in small amounts for most non-primary upgrades, i.e., anything that isn’t your shield, weapon, or keeper suit. Iron is used for pretty much everything else.

One thing I haven’t talked about yet is the win condition of Dome Keeper. There are two modes: the primary mode is Relic Hunt, and the secondary is Prestige. Relic Hunt is just a standard “Dig deep, find a special relic, and bring it to the surface to win.” Prestige is the primary mode, and is a score attack mode.

I’m personally of the opinion that Relic Hunt is an extended tutorial. Relic Hunt is relaxed, where Prestige is intended to be the primary game. Which is a bit unfortunate, because I’m personally not interested in Prestige very much. High scores are not particularly motivating to me as a factor, unless the entire game is designed around that as a core component, ala Hazelnut Hex.

In Prestige mode, you get points based on spending resources to increase a score multiplier. Then you get points after each survived wave. So it’s beneficial to spend resources early on increasing the score, at the risk of not spending those resources on upgrades. It is a interesting tension, but it’s not one that I’m very compelled by.

There are a few systems I’ve not covered here, like the semi-random relics and the upgrades they offer, but I think I’ve covered enough of Dome Keeper’s system to explain my problem with it.

Dome Keeper is a good game, but any single run can feel indistinguishable from each other run. There’s only a single unique relic that modifies combat, with every other relic modifying resource acquisition. The end result is a game that felt the same each time I played it. It was interesting, but it wasn’t fascinating, or ever really felt like it scratched the itch of something like Inscryption or Spelunky. It never really forced me into a situation where I had to really rely on an understanding of game mechanics or systems to pilot my way out. Instead, it was just more about “Oh, I should have just done X instead of Y.” There was no adapting, just learning, and some small improvements.

This is why I don’t really feel like recommending it. It’s good! It’s well made, it’s polished, and it has some clever mechanics. But I don’t get that vibe of it being a unique or super rich experience that stuck with me.

Anyway, if you think my opinion is stupid, or you really like games like Motherload, you can find Dome Keeper on Steam for $20.

Author Note 2: I played most of my 20 hours of Dome Keeper before the update that added a second playable character, with a different mining style. It’s a neat update, but I didn’t really like playing the Accessor. It’s also entirely possible that Dome Keeper becomes a much richer games with updates, but that’s not what was available when I bought the game on release day.

Marvel Snap and the Inherent Unfairness of Card Games

Marvel Snap can be random and unfair. Personally, I think that’s a good thing. Here’s why.

I started this as Marvel Snap Week, but now it’s Marvel Snap Weeks, because I didn’t get these posts finished in time. You can read part 1 and part 2 here.

Let’s talk about something that everyone knows, but nobody really says out loud. Okay, nobody except game designers. They say it, but no one else does. Game designers and… Maxamillion Pegasus from Yu-Gi-Oh. Who, in-universe is a game designer, so I guess it’s still all game designers. Anyway.

This is the ideal game designer. You may not like it, but this is what peak performance looks like.

Card games are not balanced. They are not inherently fair. In a game between two players, the strongest player won’t always win. The strongest deck won’t always win.

This isn’t a design flaw. Allowing weaker or unskilled players to beat stronger or higher skilled players is intentional.

Marvel Snap’s mechanics lean into this in several ways, both with card effects and core gameplay. But they also offer an out, allowing players on the receiving end of the RNG stick a way to minimize losses.

So let’s talk about Marvel Snap’s gameplay for the first time in this short series, and how it handles both having a wide variety of RNG, and controlling RNG’s effects.

Marvel Snap is played with a deck of 12 cards over 6 turns. Each card starting in a deck must be unique. Players draw 4 cards to start, and an additional card each turn. Cards have a cost and power. The cost is energy, and has to be paid to play the card. You start at one energy, and get an additional energy each turn. Leftover energy doesn’t carry over.

The game itself is played across 3 locations. To win, you just have to have the most power at two of the three locations when the game ends. Each player can only play 4 cards at any location. In case of ties, whoever has the most total power wins.

Locations are where the RNG first comes into play. There’s a fairly wide pool of locations, and when a game starts, three are randomly selected and placed face down. A single location is flipped face up when the game starts, and additional locations are flipped on turn 2 and turn 3. Locations can have a wide range of effects. Some buff or debuff units played on them, while others might give additional energy, or create copies of cards at random locations. One location will even play your cards for you. (And he’ll do it very badly, screw you Ego The Living Planet.)

Locations provide a huge amount of variance. Playing to an unrevealed location can be a big gamble. Sure, it might be the location that gives a free 6 drop if you fill it first, but it might be the location swaps the units located there to the opposing player after turn 3.

And this just the start of things that can randomly go wrong. There are plenty of cards with semi-random effects, or that can pull random cards into your hand or from your deck. In short, there’s a lot of space to “lose to” RNG.

The thing about Marvel Snap, though, is that losses and wins are not created equal. Let’s talk about the “Snap” system.

There is no unranked mode in Marvel Snap. Every mode is ranked, and in every game, you’re competing for cosmic cubes. The wager starts at one cosmic cube, and if the game reaches the last turn, the wager is doubled to 2 cubes. That’s if the game reaches the last turn, though, because both players can retreat at anytime. Retreating counts as a loss, but in exchange, you only lose cubes you’d already wagered.

But while you can retreat, you can also choose to snap. Snapping doubles the number of cubes staked, and you can only do it once per match. Your opponent can also choose to snap.

In a normal game, you’ll generally lose win or lose 1-2 cubes. But if both you and your opponent end the game confident you can crush the other player, the amount can go up to 4-8 cubes. And I think this mechanic, where players can state their confidence (and bluff) about whether they’re going to win is part of what balances out RNG. You’re likely to get lucky and unlucky in generally even amounts in the long run. But if you push when you’re winning, and retreat when you think you’re going to lose, you’ll come out ahead.

Players can always retreat, and simply take a one-cube loss. Choosing to push into poor situations with higher losses is an active decision you choose to make. There’s no “I have to play it out” mentality where every game is equal. If you’re doing badly, you can surrender and minimize the pain of RNG.

As a brief side note, I think it’s also interesting that retreating is displayed with the message “You Retreated” and a friendly color scheme, while losses use a harsh aggressive red. There’s a definite goal of making retreating feel like a smart choice and a good option.

Anyway, that’s all I have to say about Marvel Snap and RNG for the moment. Come back later this week, for part 4 of Marvel Snap Week(s), where we’ll talk about convergent game design, and wrap this series up. Or follow me on Twitter to see when that post goes up.

Marvel Snap and Dark Patterns

Marvel Snap has its fair share of dark patterns and skeezy progression design. Let’s talk about them for a bit.

Welcome back! It’s time for part two of Marvel Snap Week! Did anyone who worked on the game see part 1 and think “Wow, that review is positive,” or “He’s an illiterate hack, but at least he appreciates the game?” Well now it’s time to get rid of those nice feelings.

Mobile games are unique in that they’re the only platform where the games are usually “Free,” but have the potential to end up costing you more than a full ticket to Disney Land. As a result, the only sane approach is to enter with caution. They’re the only games I engage with while actively looking for a reason to NOT play them.

When a game is “Free,” you should always keep in mind panel 2 of XKCD #870.

Just replace “typeset” with “Spent hundreds of thousands of dollars to develop and put on the app store.”

To be honest, Marvel Snap isn’t that bad. It wouldn’t rank anywhere in my list of worst video game business models. But not being at the bottom of the barrel shouldn’t be the standard for this stuff. So let’s talk about Marvel Snap’s prices, the concept of dark patterns of design, and how the bar for mobile games is so unbelievably low that something like Marvel Snap seems “Fine.”

Let’s start with pricing. Marvel Snap currently sells two objects. Battlepasses, and Gold. Battlepasses work like most battlepasses in games do, but they can only be bought with real money for about $8. Complete quests that unlock over time to earn battlepass experience points. Level up the battlepass to get additional resources. It’s a fairly standard design, even if it does reinforce a lot of the dark patterns we’ll be talking about later. If you pay the $8 you get some gold, extra resources, and special card styles and avatars.

Now let’s talk about gold, and how Marvel Snap’s progression and collection works, because it’s a bit more insidious.

Marvel Snap has a very unique design for its progression system, one that I’ve actually never seen used before. Instead of opening booster packs to get cards, or pulling from boxes, or having a wildcard system, Marvel Snap has a single value for your collection level.

Your collection level increases as you acquire new cards, and level up your cards. As you travel along the collection level track, there’s a variety of rewards, and some of those rewards are mystery cards. What the game doesn’t really ever tell you, though, is that those mystery cards aren’t random. Instead, they’re random from a pool, and removed once you get them. So as your collection level goes from 1-216, you’ll unlock cards from pool 1. And once you reach 216, you’ll have unlocked them all, and you’ll move onto pool 2. This part is reasonable.

What’s not is how the tracker changes.

As you travel down the collection level, the amount of levels you need to unlock a new card starts to shift. First it’s every 4 levels. Then every 8 levels. I’m currently collection level 358 and I unlock a new card every 12 levels.

This means that even if you play the game the same amount every day, you’re going to start making progress far, far slower. This is because gaining card level comes from upgrading cards, which requires two resources: boosters and credits. And while technically boosters are a limiting reagent, you can get them fairly easily by playing games. Credits are limited by daily quests.

Guess what they sell in the cash shop for gold?

So if two players play, and one spends money, and the other doesn’t, the one spending money will progress their collection faster. This, combined with the fact that your progression is designed to get slower over time, feels scummy. They claim, “You can’t pay to win.” And technically, that’s true. But you absolutely can pay to speed up your collection progress.

Oh, also. The game sells alt-art styles for cards at between $10-$20 in fake in game money. Yes, the expensive art alt styles are 1200 gold. And yes, the closest purchase in gold to buy those styles is $20. So it counts as $20. Skim is a real trick.

Man, I’ve written like a page, and I haven’t even gotten to the game’s dark patterns. There’s nothing super egregious here, but they use a lot of the standard stuff. Daily quests force you to play daily. Limited-time battlepasses make you grind. The aforementioned bullshit where the in-game currency that you purchase is always just a bit more than the expensive item’s cost, so that there’s always some leftovers.

All of this sucks, because Marvel Snap is actually quite fun to play. And that’s what we’ll be talking about tomorrow.