Working on 2.0

Hi all,

I’m still chugging away at Lane-Pushing Games 2.0. Here’s some pics to whet your appetite.



Yes, I am very excited about this. It will be cool, and you’ll get the DotA1 article as well when it launches.

Since Master X Master will be closing recently, I decided to post that one here in advance; you can read it below!


Review: Master X Master

MXM_PreviewJack of all trades and master of few, Master X Master (NCSoft, 2016-2018) is a playfully presented game with lots of modes, features, and character customisation options to tinker with.

Among its many modes is a 5v5 lane-pushing battle called Titan Ruins, which will be the main focus of this review. The game has been announced to close for the last time on 31st January 2018, so if you haven’t taken it for a spin yet, head on over to the official site and give it a spin. It looks like there’ll be a closing tournament to see the game off as well.



The most distinctive trait of Master X Master is that players select two masters at the start of the game. These are played as a tag-team pair: only one will be active on the battlefield at any time. One of your masters is visible to the enemy team when you pick, and the other is concealed until picking heroes has concluded.

Masters have the following abilities:

  • A basic attack, which is always aimed.
  • A passive, offering some twist on how the hero should optimally attack or cast. Typical examples are “every third cast you get a shield”, or “incoming crowd control is reduced by 25%”.
  • An escape, which can be a dash, stasis, speed boost, and sometimes will have offensive potential.
  • A spell, which is picked pre-game from a small selection. Both your masters will share one spell.
  • Four traits, stats which can be upgraded gradually during a match.
  • Four basic skills, of which a player can bring any two into a match.
  • An ultimate, which requires ultimate meter to cast.

The traits for a player’s masters.

This game uses WASD movement, so basic attacks are always aimed towards your cursor. They come in lots of varieties, including cleaving, double-attack, optional charge up, attacks which heal allies, split on impact, long-ranged missiles, goops of acid that stay on the ground, and ranged attacks which deal less damage at short range. Most weapons will overheat after a few seconds of continuous use.

Cagnazzo swings his ball in an arc, damaging everything it hits.

Masters can move and strafe constantly without penalty, so that is exactly what they do at all times. It can be difficult to keep bullets trained on a target if you’re not used to it!

It’s possible to walk up ramps, jump, and shoot down from cliffs, which is fun but the game stays firmly 2.5D.

Basic skills are unusual in that players must choose pre-game for both of their heroes, which skills they will bring into a match. There is also total freedom to decide which will be your Q or your E for each master. That’s important, because masters get skill points as they level up, and the Q on both heroes will take points simultaneously.

Skill points offer a different sort of value to most lane-pushing games. They apply to basic skills and ultimates only. They don’t linearly increase certain values of the skill; instead they offer a different flat bonus at each level. Usually levels 3 and 6 offer something juicy, and it’s allowed to sink six consecutive points into the same skill so that seems like the strictly optimal choice.

The available Spells are:

  • Heal will restore health to nearby allies and buildings over 5 seconds.
  • Shield a friendly tower or Titan for a short duration.
  • Sprint to move faster for a few seconds.
  • Ward grants vision in an area for several minutes.
  • Teleport to a friendly tower or Watch Tower.

I appreciate that all of these are useful and a team might encounter difficulty deciding which to prioritise.

Affinities and Tagging

Every master belongs to one of the three Damage Affinities. These are Argent, Helix, and Kinetic.

It is a simple rock-paper-scissors system, where each affinity deals +15% and takes -15% damage from one of the others. That is a 30% relative dps advantage for an affinity advantage in an engagement! Damage from the same affinity has no modifier.


It would be pretty awful to arrive at a lane matchup and be at a natural disadvantage like this. Thankfully, that’s what the tag-team system is for! As long as a player pick two masters with different affinities, any 1v1 is guaranteed at least an even match.

Each of your masters has a separate health and mana pool, as well as its own ability cooldowns. Ultimate meter is shared, as is the cooldown of the spell. The Tag ability is used to instantly switch which master is active. This has a 14-second cooldown.

There are two things in the above paragraph that I strongly disagree with. In perSonas (the game which first introduced tag-team to mobas in 2006), health and mana pools were shared, and the switch cooldown was 6 seconds. That worked. You switched often because you could, it was offensively good to unload more spells, and there was nothing to lose defensively.

MXM’s numbers mean that switching to unload more spells gives up access to your precious extra health pool for 14 seconds. That’s a really bad idea, and players know it. You really shouldn’t switch unless your current health pool is not going to be useful to you: either because it took too much damage already, or you won’t see combat for 10+ seconds.

I don’t see why MXM went its own way here. perSonas already got this right.

Lanes and Titans

Following some familiar patterns, there are three lanes and a couple of towers protecting them. Only the mid lane has outer towers, and there are no barracks or similar objectives inside the bases.

During gameplay, teams collect Points through almost everything they do: with last hits awarding 1 point each, and clearing the jungle, hero kills, and so on more. Points are an important factor towards winning, and will directly grant victory if you reach 1000, or the game lasts 25 minutes without a core being destroyed. (The team with higher points in that scenario wins.)

Every 100 points collected by a team will cause a friendly Titan to spawn and march down the mid lane. These powerful troops are tanky and good at hurting everything, particularly enemy towers and their core.

There are three types of Titan that can spawn: one for each damage affinity. They have separate strengths and abilities, and of course are resistant or vulnerable to masters based on how the affinities match up.

Here’s the breakdown:

  • mxm_argentArgent:
    • Globally, allied masters gain 20% lifesteal and increased stamina regeneration.
    • Each nearby friendly master increases its damage by 20%.
    • It will sometimes fire a damaging shockwave at enemies.
    • On death, it explodes to hurt nearby enemies.
  • mxm_helixHelix:
    • Globally, allied masters gain armour, magic armour, and increased health regeneration.
    • Blocks incoming projectiles from its facing direction with its massive shield.
    • It can heal all allies in a large area.
    • On death, any nearby allies are shielded.
  • mxm_kineticKinetic:
    • Globally, allied masters gain cooldown reduction and increased mana regeneration.
    • Leaves behind a trail that boosts the speed of friendly units and slows enemies.
    • It will sometimes pull all nearby enemies to itself.
    • The same will happen when it dies, with the addition of a heavy slow.

Warning: this is not a drill.

Titans are worth less experience to the enemy team the more they have travelled. Also, when one Titan kills another, it recovers about a third of its health. This avoids stalemate situations where Titans keep cancelling each other out.

The first Titan to spawn each game has two properties:

  • Its type will be random. All Titans thereafter will follow the sequence Argent, Helix, Kinetic, Argent…
  • It will appear with a substantial amount of shields, as a perk for the team being the first to 100 points.

I’m not a fan of the random/uncontrollable Titan types. If this system allowed player interaction, then maybe players could think about which Titan they wanted and the differences between Titans would be something to celebrate. In their current state, there is not much to be done other than get points and Titans as quick as you can.

Monolith and Shards

Among the various ‘unfriendlies’ to be found off-lane is the Titan Monolith, a strange turret device that can only be attacked from short range. Like all the unfriendlies, it uses pattern attacks which are usually marked with red areas to allow a moment to dodge.

When the Monolith  or a Titan dies, they drop a Titan Shard. These behave a little like a “capture the flag” objective. Walking over a Shard will pick it up, revealing that master to both teams for a brief moment. It slows the master carrying it, and bringing it back to the safety of the main base will secure it for your team.

Once five Shards have been assembled, anyone on the team can temporarily transform their master into a Titan Incarnate. This is a better version of the Titans above: packing tons of health and damage, and having the advantages of being deployable anywhere, and being able to focus enemy masters or buildings as the situation requires.


In fact, Titan Incarnate has two forms: a “brawling” one (active by default) which has huge line damage and can leap, and a “hammer” form which excels at hurting buildings and can shield all nearby allies and itself.

The Incarnate’s basic attack has a pattern: its third attack sends out a shockwave, while in hammer form it uses a fast weaker swing followed by a strong heavy one. It takes a bit of practice to learn to use these special forms.

The transformation takes a couple of seconds, during which any two players can veto it, or any two players can okay it (in which case the transformation completes instantly). This feels like a sensible way to let the team contribute on when precious shards get used, and allowing ‘yes’ votes to accelerate the transformation means voting yes has tangible value: encouraging players to do it whenever appropriate.

An important note is that the Incarnate will adopt the affinity of whichever master did the transformation. That is really important when you’re outputting the kind of damage that an Incarnate does (or taking the kind of focus-fire as well!)

Buffs and Bosses

Four Watch Towers are placed on hills around the map. They take a short channel to claim, and grant vision for 5 straight minutes thereafter (during which the enemy team can’t contest them). Since matches last at most 25 minutes, taking a watch tower secures its vision and value as a Teleport target (for the Spell that all players can take) for at least 20% of the game.

The jungle contains some buff camps: two red, two blue. Both colours provide identical bonuses (reduced tag cooldown, and health/mana/stamina regeneration), which would be redundant except that the blue and red buffs stack with each other. Together they’re a considerable powerup. (Getting two blues would simply refresh the duration.)

The weak boss on the map is King Tanian, who grants 30 points and some experience when defeated.

The King’s Cross. One of many delights he has to throw at us.

Opposite him on the map is Rozark, the biggest objective. He lives inside an underground chamber. To enter, masters must briefly channel without taking damage to enter. Leaving a hero on lookout seems like a handy way to take him with minimal interruption, especially if the nearby Watch Tower has been claimed.

Rozark awards an immediate 100 points, and places a curse on enemy towers when defeated.

Busy being a pattern boss and whatnot.

In the corners of the map are two Fallen Altars, which are protected by a pair of guardians. It’s possible to defeat the guardians without taking any damage since they signal their dashes and swipes with enough time to dodge. As previously mentioned: this is true for all the unfriendlies, but it matters more for an on-lane objective where having to tank damage would be unfavourable for laning. Instead, we have a situation where the objective is tricky only when enemy players show up.

Once slain, a modest channel will capture the Altar, freeing a Fallen Hero to push that lane for your team. Usually the enemy will try to interrupt with an attack or two, so finishing the channel isn’t always easy. Especially not late-game, when its duration reaches 15+ seconds…

The Fallen Hero is a great pushing companion, who will pick a nearby ally to link with, steadily healing them and granting damage reduction. Very handy! Occasionally it will also try to hook enemy masters to itself, use cone attacks and so on.

Fallen, but won’t allow others to fall so easily!

Even towers get in on the “action” style of the game: they will occasionally release a burst that knocks enemies away from them if they take enough damage. It keeps pushing interesting I guess, if the various super troops weren’t enough to do most of the job.

In the middle of the map, there are two bodies of water each housing a River King. This little fishey is available to hunt from the start of the match, and grants some healing and ultimate meter when killed. It is almost harmless to fight, unlike everything else.

Client and Atmosphere

It’s worth noting that MXM is not just a lane-pushing game. It has lots of short co-op stages which can be cleared, many of which are nicely decorated and have fun enemies. It has a 3v3 arena mode, and a 4v4 capture-points mode as well. You use the same masters and swapping mechanic for all of these.

The client itself is themed as the hull of a spaceship, and your character actually stands there and can walk around/emote/chat with other users while you wait between games. There’s even a jukebox where you can put on music for everyone else to listen to. Pro tip: always choose this track.

Hanging around: in space! This used to be packed in early alpha.

The music is upbeat and fun, and the heroes are a mixture of everything: stone beasts that look like they’d fit right in with the Titans, blobs of orange goop, baseball players and boxers, idol singers… there’s something (or a skin) for everyone.

Frankly, a lot of effort went into this game (certainly for the western market), and it’s not entirely clear why it’s closing down. My suspicion is that something went awry in NCSoft’s home market of South Korea, and resources were moved elsewhere. If so, that sucks because MXM is an entertaining game.

Stat Padding

Among the many, many things you can spend your time collecting or crafting in the client, are Nodes, which are functionally similar to League’s former masteries. There are six nodes to a page, and you bring one page into any match that you play in, to be shared by both your masters.

The nodes are split between offensive/defensive/utility, and offer effects like +10% crit chance, +3% movespeed, +27% tagged-out regeneration, +5% weapon damage, +8.7% cooldown reduction, and so on.

Combined, a node page could offer +30% weapon damage(!) if you had six of the good ones. However, this is slightly dampened because each master has a “preference” for which node types it wants. Vonak wants three offensive, one defensive, and two utility (3/1/2). Lilu wants (1/1/4). If you use more of a colour than the master’s preference, nodes of that colour will have reduced efficiency.

Here, my extra offensive and defensive nodes are at reduced efficiency. The large number displays overall efficiency.

For further customisation goodness, every master’s basic attack has three upgrade paths which can be pursued as well (requiring assorted junk to actually do the upgrades). The paths include bonus range, damage, splash angle, projectile speed, charge time (for masters with a charged up attack), healing, reducing heat buildup, and more.

Closing Thoughts

I enjoyed messing around in MXM, but three things made it hard to enjoy the game.

Firstly, WASD movement is a nice change of pace, but I still have some reservations about how MXM handles it. Masters suffer no movement penalty when they attack, so kiting around is automatic behaviour for everyone all of the time. It feels very thoughtless.

Secondly, the game has a min-maxer attitude underlying it. The combination of nodes, attack upgrades, the unusual skill point stats, traits, and damage affinities means there is always a percent or two to be improved upon somewhere if you put your mind to it.

There is a point where adding more knobs and whistles to the definition of a hero is too much. MXM sails right past that. If all the passive complications weren’t enough, masters don’t even consistently have the same skills: their basics and their attack could be different depending on what was picked.

This is while there are five pairs of enemy heroes on the enemy team to keep track of in each match. There are too many little things adding up, and not enough big things that mean something.

Finally, if we consider the “hook” of MXM to be playing as two heroes and switching back and forth between them, two critical design choices (separate health pools and Tag’s lengthy cooldown) are poisonous. They encourage playing like a single-hero moba as much as possible, where sometimes you have a free health refill.

I expected that the promise of MXM would be the same as the promise of perSonas: that I would get to be swapping back and forth mid-combat doing clever things. This iteration of MXM seems to be for someone different: someone who doesn’t really know what they want but it definitely involves practising lots of small tweaks and procedures.

Differences in culture, or target audience, perhaps? I don’t know, but more could have been done with these ideas. Maybe someday under a new Master.

Interview: Cyclotrutan

Today: a special feature! Martin ‘Cyclotrutan’ Schwesinger, the developer of Darkening of Tirisfal / Crimson Coast (2009) joins us for an interview! Take a look through the review for a quick refresher, and I’m pleased to share that the map has received a substantial update after 7 years! Go check it out!

Martin, good morning! How are you doing?

Hello, thanks for having me!

I’m going to ask a bunch of questions and talk through some things; hopefully have a bit of fun while we’re doing it.

Yeah, it’s all saved somewhere in the very back of my brain because it was so many years ago, but I’ll try to have something to say about it.

To start from a very early stage: when did you first get into gaming? What kind of games were you playing?

Well, that was mainly shooters at first. I played Counter-Strike, and also used the Worldcraft editor to make maps for the Quake 2 engine. That was the first time I had got into mapping and creating my own games. Before that, when I was 6, 8, 10 years old, I was creating board games: very simple ones and I forced my family to play them with me. So I’ve done this my whole life.

When you say you were making board games, would you have got together bits of paper, drawn rules on cards, that kind of thing? Or modeled of existing board games; what was that like?

I had some cardboard and painted some fields on it: water fields and grass fields in Settlers of Catan style, and iterated on it all the time like a 10-year-old would. You can imagine what a game developed by a 10-year-old would look like.

Tell me about your first experience with Warcraft 3.
When did you pick it up, what did you think of it as a game?

I picked it up when it came out, and I enjoyed it quite a bit. I played competitively in 1v1 and was in a clan; we did clan wars back then. I wasn’t the best, but was kind-of decent with Human. I went into mapping very early on, and made my first very crude maps. Later, I enjoyed Free For All a lot in Warcraft 3 because it’s actually very… it can be competitive if you play with good opponents that don’t just kill you out of spite if you attack them; stopping trying to win to just screw you over. So Free for All (with 4 players) was a lot of fun competitively. After that I played mostly custom games because yeah, that was what the game evolved into, mostly. Ladder died out and the custom game section grew.

When you started playing custom maps, wh
ich ones were you playing? There’s a huge variety out there.

Let’s see… so I enjoyed Uther Party – the minigame compilation map. The original DotA, of course. I really enjoyed the Gaia’s Retaliation RPG though that came out a little later – this was one that fascinated me, and inspired me to do the latest iteration of my own map. Element Tower Defence, and some other tower defences… but you know, you play a tower defence 2-3 times and it’s like: okay, I’ve solved it, nothing more to see, but Element TD was a cool map also, you can do a lot of fun strategies and variation in there.

Gaia’s is a huge project. But lets go back to that interaction with Eul’s DotA: I would assume that was the first AoS map that you played?

The first DotA I enjoyed a lot. When DotA Allstars came about, I really lost interest in it and that was mainly because it was too much on the custom maps section; 80%, 90% of the games on the list were DotA Allstars and there was too much of it. The other thing was that it moved into a direction that I didn’t really enjoy: it became much faster and more like a tactical shooter than what I would like in an AoS map. So when DotA Allstars came around, I didn’t play DotA anymore.

Were there any other AoS maps that caught your interest around that time?

I played a few. Tides of Blood I remember, that was a cool map. But none that I really played extensively.

At some point, you decided to start your own AoS project. How did that come about?

Well, it was just me experimenting with the editor at first. Until now I’ve had five different versions with different names of my AoS map: and always because I wanted a new terrain. It started with an ice terrain called the Frozen Blood, and then it was a forest terrain called the Forest of Blood so it was not very creative naming. Then it was Sands of Blood, and I kept improving on the mechanics and introduced new heroes, and made my first attempts at complex abilities with triggers and so on. It was when I did the Crimson Coast version that I decided maybe I should publish it on some website, and I published it on Hive Workshop where it got some good positive feedback.

And later, inspired by Gaia’s Retaliation RPG; I saw the terrain there and it was so fabulous that I thought: “well maybe I can do this myself”. I tried to do new terrain and then I renamed the map again because it was no longer a coast, and then it was the current version which is Darkening of Tirisfal.

Those early versions had a consistent naming trend, which was
Forest of Blood, Sands of Blood… was there a little bit of a Tides of Blood inspiration there, or is that just coincidence?

I think my map pre-dates Tides of Blood, so it can’t be.

One thing in my map that I think was inspired by Tides of Blood was from one of their heroes: the Blood Mage. It has this huge blood wave and blood explosion, and I made a hero the Blood Lich which also does everything like that. So that was maybe the inspiration I took from Tides of Blood.

When you say that your map pre-dates
Tides of Blood, that means it made before the Frozen Throne expansion, right? Were you doing this in the Reign of Chaos editor?

I think so. When I started there wasn’t even an ability editor in the World Editor, so what I had to do was edit the “.slk” file, this table that’s in the map archive, and it’s still in the map which is something that isn’t very efficient because it takes up a lot of data and makes the map much larger. I wanted to get rid of it, but it’s really difficult because it’s so ingrained in the map that if you take it out it just crashes.

For a map that started with
Reign of Chaos (2003), it wasn’t until quite a lot later (2009) that the map ended up in the public. That’s a six year gap! Who was playing the game?

I played it with friends and my clan and with testers. Of course I don’t remember the exact timeline, but I know WoW came out in 2005 in Europe, and during Burning Crusade we still played Sands of Blood; this was a version that pre-dated Crimson Coast. I don’t put so much focus on publishing stuff I work on; I just enjoy doing it. If someone likes it, of course that’s great but its not something that really drives me towards my creations. I do it just for fun and for creating something.

I’m also quite a perfectionist, and if something isn’t as good as I think it could be, I tend to be very hesitant to publish it because I think it will be critiqued and I can improve on it, and then maybe later I’ll publish it. So I’ll delay until I think it’s perfect. Of course it’s never perfect, but I’m obsessed with perfection I guess.

You wouldn’t be the first artist to feel that way! Apart from early
DotA and ToB, were there any other maps that influenced your work, or was this a work that evolved in isolation over this long period of time?

There were certainly some minor influences from different maps; Gaia’s Retaliation and maybe I had some influences on heroes but I can’t remember the specifics now.

One of the things I first noticed when playing
Crimson Coast was that it’s 4v4, rather than 5v5 like most AoS maps. Was there anything behind that choice?

It started as 4v4 because I think there was a cap on the number of players back then and you couldn’t do 5v5… I don’t know if I’m remembering that correctly. We never played 4v4; we only played 3v3 and the map was balanced around that, though there’s the option of 4v4. Later, someone told me that I should probably increase the cap to 5v5, but there were so many triggers and abilities that were written for 4v4 that making the shift would have been an immense effort.

In some senses,
Crimson Coast would feel familiar to any AoS player, because you’ve got three lanes, the regular troops, creep camps, and some recipe items in there as well. One of the things that makes the game a bit different is that every hero starts the game with a Staff of Teleportation item, and it lets you teleport to any allied unit after a 10 second channel time.

Right. At first it was just an item you could buy at the shop, and it cost gold like anything else. We found that everyone would buy it immediately with their starting gold, so it was really obvious that it should just be a standard item that everyone gets. Then they can buy some other item with the gold they have.

Was this because everyone needed to have a
Staff to play effectively?

I think it was very necessary. One of the tactical depths of this game is quickly switching lanes and going where the creep wave is currently pushing towards the tower. Battles are much slower in Darkening of Tirisfal, so the computer-controlled creeps need a lot longer to kill the other wave. As a result, they spend more time between the towers before they reach the tower, and when they then finally arrive: more creeps have accumulated at the tower. So you wait for this to happen on a lane where nobody is, teleport there, and use the momentum of that creep wave to push against the tower. The towers are quite far apart, so the teleport is a valuable resource.

You have to really switch quickly between the waves, because as you’re pushing, the other wave you just neglected comes up to your own tower. You have a lot of dynamic gameplay this way.

You changed the terrain a number of times over the map’s development, at least in terms of theme and setting. Were you iterating on the layout as well?

I made one change to the mid lane when I switched from Sands of Blood to Crimson Coast. Previously it was a straight line in the middle with L-shaped lines at the edges. What happened was people would always push the middle lane and the outer lanes had little relevance.

I don’t know how it works differently in DotA or League; a straight mid lane didn’t work for me, and it certainly works for them. I made this S-shaped lane so the overall length of the lanes remains the same and all three lanes have the same relevance in the late-game.

Right; the troops tend to bundle up in
Crimson Coast so lane length matters a lot. One of the other features in the game are these special passives called Runes. You can only unlock 1-2 of them during a match. How did they fit into the game?

This was something I introduced very late. Previously the level cap was just 20, and in a long game you always hit the level cap and there was nowhere to go. The way Runes work is: you can get to level 20, but the experience bar goes to level 21. If you hit level 21, you go back to level 20 and get a rune slot instead. So every time you hit 21 you get another rune. It’s another way to progress.

The runes are quite creative: one reduces the channel time on
Staff of Teleportation, another explicitly helps you destroy buildings… did you plan in advance to have them interact with so many mechanics?

I was just designing on the go. I started with a few runes, to see what different kind of effects I could do. Also, I wanted each hero to have some kind of rune that they can use. That said, often players would say, “I don’t want to spend too much time thinking, I’ll just get this one rune I get for every hero”. That was Rune of Enslavement: it gives you a summon that follows you around and shoots at enemies.

Crimson Coast has a number of what I’ve been calling chambered abilities. They’re this type of ability where you have multiple sub-abilities that are tied to the one ability slot. It’s as though you have a gun, and can choose what bullets to load it with: the bullets have different effects but the gun has a single cooldown. The idea gets explored quite a bit: how did you refer to that type of ability, and how did you think about it?

The first hero with this large number of abilities was the Shapeshifter hero, a wisp that can learn five different transformations. You can go into treant form, wolf form, etc. and they have their own abilities. You can stay in each form for only 30 seconds, and then you must choose a different form with your next transformation.

I had imagined making a hero like Shapeshifter for a long time. It was my first try at a complex hero, and working on it taught me a lot about the trigger editor. Everyone liked the result, and because it was so successful I decided to try another, similar concept.

The second one I did was Avatar of the Elements, which you covered in your review. She’s the hero whose skills have one sub-ability for every element. Some people liked the complexity and would play her from time to time, but for most people it was too much. They couldn’t remember that many abilities!

I didn’t refer to these as ‘chambered abilities’ or anything; I just saw it as ‘complexity’.

For the next hero I made, I decided to have one chambered ability which would be very interesting, and keep the other skills simple. This way, the hero would have less of a learning curve: you can select something at random on your first time playing and it will be good enough, but on the second try you can think about which sub-ability you want.

The game demonstrates a lot of variety in how often and where you can switch out your bullets: it really explores the
idea of chambered abilities in depth. It sounds like you were prompted by player feedback: by how willing people were to invest in learning heavily chambered abilities versus lightly chambered abilities.

The first heroes I made were very heavy on chambered abilities; these were too much for most players. But I had a variety of players in my playgroup, and some players just wanted to play one hero every time, and it was one of the simplest heroes.

I wanted to give them all a chambered ability and make them a bit more complex (and in my estimation more interesting) but there were players who just liked these very simple heroes. I added a bit of variation, but kept them simple so that those players would have a hero they could enjoy. And then there was the Avatar of the Elements and the Shapeshifter that were very complex, and all subsequent ones were at a middle level of complexity.

You were a player of
WoW for a number of years, and that’s a game where you have access to quite a lot of abilities. Would you consider that to be an influence on the relatively large number of abilities your heroes ended up with in Darkening of Tirisfal?

Sure; there were certainly some heroes that were heavily inspired by WoW. The Death Knight, the Shaman, and the Dark Naaru. The Dark Naaru was a very controversial hero: every time I played a game this became the main focus in the chat: it overshadowed everything else.

I wrote in the hero description that “this hero is the most difficult to play that you will probably find in any AoS map”. Still, some people who played the map for the first time would just slam the Dark Naaru and complain that it’s impossible to play and they’re dying all the time. On other hand, when I was playing the Dark Naaru, I made mistakes, but I got it under control fairly well and people were complaining that it was too powerful. It was very funny.

What the hero does is: it cannot move, and cannot attack. It can summon two voidwalkers; they can move and attack, and scale with the stats of the Dark Naaru. They have an ability called ‘Recall‘, and can teleport the Dark Naaru to their current location. You move around using them, but the Naaru itself just sits somewhere on the battlefield shooting its spells. It wants to be in the center of the battlefield, and just unleash its storm ability. If you made one mistake, you couldn’t move and would just die. So you had to be very careful.

One of the things I just cannot wrap my head around is how people played against it. Again, when I was playing the hero the complaint was always that it was too powerful. For example, if I was pushing a tower with Naaru, the opposing team might teleport in to deflect the wave. At that point I say: “okay, I have to retreat” and order my voidwalkers to back up. Every other hero could have just walked away, but because I was playing the Dark Naaru I was forced to sit there while the voidwalkers did the walking. Then, the enemy finishes with the wave and starts attacking me, but now I can Recall because my voidwalkers had a 10 second head-start.

The best way for the opposing team to look at that is “he got away”, but they don’t: they chase right across the map even though I already have a lead. Then they complain that this was imbalanced.

It’s something you have to keep in mind while testing; people are not always thinking rationally. If you want to test how people react to it, their emotions are very important. You don’t only want the game to be balanced, you also want it to be fun. If people feel that way, you have to keep that in mind: “what can you do to make them less frustrated?”. But when it comes to balance, people are very irrational.

Did you feel that the better players would pick the heavily chambered heroes and Naaru more often?

Sure. The Dark Naaru was either picked by me, or someone who had never played it and complained all game. There was nobody else. The other heroes were played quite often. There were ones that were more popular, but that just has to do with how the chambered abilities work. On some it’s a bit clunky, and for others it’s a bit better.

It’s still
fascinating to see that type of ability explored in so much depth. Were there any heroes you wanted to try but never got around to?

I don’t think so. I have one hero still left in my file; maybe I’ll finish it. I explored most of the very “out there” ideas.

I’m not an expert on the trigger editor or with the Jass language used in Warcraft 3. Every time I wanted to make a hero, I had to build workarounds and learn the things I needed to make it work. I think many designers fall into the trap that they’re doing stuff just because they can. “I have the ability to make this very complicated ability trigger, and change the terrain and have an explosion here and stuff”; all of these things happen because they’re so adept at the trigger editor and I think it has no gameplay value. It’s something that I found with many of the later AoS maps.

That’s not the case with the original DotA; it’s very focused on combat-oriented straightforward abilities. Some heroes have a special ability that makes them very unique, but it’s not something that is ever “just for show”. It really has some purpose.

I think it’s something that shows up in a number of AoS maps: you build this experience working with the trigger editor, you have this sandbox in which you can play, and
trying out “new tech” is exciting. Darkening of Tirisfal is a very refined map; the level of ‘superfluous’ is quite low compared to other games.

One thing I’m curious about is: you did indulge a bit by having Naaru in the game, because as you say: it was a hero you mostly played yourself. How do you feel about having such extreme designs in the game? Is it a good thing?

Sure, I mean you’re not forced to play it. If it’s too complex for you, pick another hero.

Of course, if you’re a professional developer you have to ask: “do you want to spend time on something that only a few players will play?” But if you’ve already spent the development and creative time on that hero and someone asks you to change it, you could just do another hero and leave the extreme one as it is.

People who like to play it can play it, and other people get a new hero that they would like to play. It’s not the case that it’s too powerful or too toxic for the game. You have to try it for like an hour offline and get used to it, and then you can maybe play it. I don’t know how it is in other play groups, because the map got played a bit by other players and I don’t know if the Dark Naaru was ever used by them.

Could you describe what was going on towards the end of
Darkening of Tirisfal‘s development? Did any other projects spring up to fill in the void?

Well, our play group slowly fell apart. That was at the end of Wrath of the Lich King, so around 2009-2010. At that time, I didn’t have any players to test what I did, so I lost interest in continuing the map.

I’ve since got into card games. I currently play Magic: the Gathering; I played Hearthstone when it came out. I still watch but don’t play. I’m currently working on my own card game called Conquest of Orion. I’ve done this for probably longer than I worked on my moba map… since 2008 almost? It’s the same style of design I described before. I have a new version, brainstorm on it, create, let it rest, and then get some distance to it so I can re-evaluate all of the core mechanics in a few months. Then I say: “Now that I have clarity on this, is this really worth it? Do I have to change this completely because this doesn’t work?” I have it in a state that I’m very happy with, and I’m still testing it but it needs a lot of work.

Nice! It sounds like people have a lot to look forward to in terms of this card game and a new version of Darkening of Tirisfal!

Around 2009, we started to see the first commercial mobas kicking off, such as League of Legends and Heroes of Newerth. What was your opinion of them at the start?

I never got into them very well, because I think they’re doing a bad job at being accessible to new users. When I watch a League of Legends or Dota2 stream, it’s hard for me to see what is actually happening there. I don’t enjoy this very fast gameplay.

Basically in esports right now you have two different games: there is Hearthstone and then there are variations of CS:GO. Of course the variations are quite different games, but they all test for the same skills. The skills that you need to be good at CS:GO for example are the same skills you need to be good at Dota2: team coordination, alertness, and motor skills. Across the entire genre, these are the things that are being tested. It seems that this is the thing that people want to see.

I think this is one of the reasons that Starcraft 2 failed: it didn’t include the teamwork component. It was just everything except teamwork. What I dislike about the new mobas is that they ask players to do something that is not very intuitive. I want to pick a hero and I want to start brawling. I want to attack the creeps on the opponents side and battle with the heroes, but that’s not what you’re supposed to do. You’re supposed to strafe around, wait for the last hit, and then you attack. In some games I heard that they’ve removed this, but you’re also supposed to attack your own creeps.

This is something that seems to me to be a big game design flaw, because for example in Magic: the Gathering, they had a problem for a long time where the cards wanted to play intuitively weren’t very good. People want to smash in the opponent’s face with giant dragons, and for a long time dragons sucked. Newer players would play their dragons and would get obliterated by their opponent playing counterspells and control and “you discard your hand and I gain 10 new cards”. What they had to do was change the game so that dragons are actually good, so a new player wanting to smash in the face of the opponent with a dragon could really do it.

The moment you start a game of League of Legends or Dota2, the thing that you want to do is not something that you’re supposed to do. That seems like a design flaw to me. I’m a bit confused that this isn’t picked up in the community. I think it has to do with that in Warcraft 3, the DotA community became very elitist. Maybe it’s some remnant of it; they don’t care about it… it’s very competitive so for competitive players it’s a very good game, don’t get me wrong. It’s just not very accessible; that’s my take on it. You could have the best of both worlds.

One of the things that many people like about
Dota 2 and League of Legends is that you have a large cast of heroes. That seems to have become the standard in mobas: aim for 100+ heroes, and then you have a drafting phase to produce a lot of dynamics. How do you feel about the idea of such a large cast?

I think it’s quite astounding that… I think 100 heroes were played in one Dota 2 tournament. I think it’s just astounding that they managed to do that: to get the balance so right and make the draft so tactical. I can see that this is something that’s very compelling to watch and play if you’re an expert on the game. I don’t think I could do such a large draft of heroes on my map because it would take too much time!

For Dota 2, I think the heroes are just a vessel to experience this very competitive environment with many different choices, and they’re all designed to fill a specific role. For me, I wanted to do the heroes as just an exploration into flavour and mechanics, so I didn’t have too much focus on the metagame. I think that’s just a different approach.

Have you tried any of the more recent mobas, like

I have to stress: my days of playing mobas were 7 years ago. I played a bit of HotS when it came out but it didn’t spark anything, so I lost interest in it. It’s partly because I think it’s a bit silly. The idea that you smash everything together is something that I don’t like; I would like a game more focused on flavour. You get more immersion that way.

For me, games are just a vessel to explore worlds and flavour and so on. Of course I cared about gameplay, but not in a competitive sense: more the experience of this hero and the flavour of everything coming together. When I, for example, design a card game, that’s a perfect opportunity to design a new world and translate it into a game, and let the players experience the world first-hand. It’s something that excites me about game design.

Has your work over the years with
Darkening of Tirisfal and its predecessors had any impact on your career or what you do now? You certainly put a lot of time into it!

I went to university at the time and studied physics, and I think my affinity for game design and my affinity for physics come from the same character traits that I have: I’m a very creative and analytical person. For example, I’m not as good at writing as some other people, but at game design there comes together this creativity with also this very mathy style.

Another thing is that… I’m a perfectionist type. I think the two components of perfection are striving for perfection which is positive, and there’s obsession with perfection which is negative. I think I’m extremely high in striving, and moderately high in obsession. Striving means you’re trying your best to get it right, but obsession is when you’re writing your thesis but you can’t sleep because you have one comma wrong or something. With my thesis, the day I got it printed, I saw that there was a grammar error in the very first sentence of the introduction. I had to get back there and fix it and call up again and say “NO! I NEED A NEW VERSION!”.

I think this perfectionism is something I see in my maps and I hope that other people see it: being very careful to the details. It’s something that fascinates me with Blizzard games. They are very good at it; they’re very good at having a… the saying in german is “aus einem Guß”. It means “Everything fits together perfectly”. You get the idea that everything works together in harmony; nothing seems out of place.

In general, my time mapping hasn’t affected me. I’ve never been in a professional position as a game designer, though I’m still pursuing it as a hobby and it’s something that I want to do more seriously at some point in my life.

Thanks so much for your time Martin! Where can we find you and Conquest of Orion?

I don’t have social media right now. I have a small web page at where you can find the card game. You can also find some of my fanmade MtG sets on there, and some of my paintings (but I’m a very novice painter). I am currently trying to focus on Conquest of Orion, and want to have a new site up for it in the near future.

All right. Best of luck with your projects going forward!


Moba Survey 2017

Hi folks,

I’m running a survey about mobas! If you have 10 minutes to spare, I’d be very appreciative if you could take a look (now closed!). This is on behalf of an upcoming moba that I’m working on.


Weekend news

A couple of things for your attention:

  • I will be hanging out at the Galway Game Gathering all weekend! Come say hi if you’re attending.
  • For those of you who have it installed, a very committed user arsjac has reproduced the original Defense of the Ancients mod within Dota2! It’s now a full 1:1 port. I’ve been meaning to review this forever; hopefully I get to it before the end of the year… Go check it out!
  • There is a major site upgrade for lpg in the works! Despite having a lot of content, it isn’t indexed very well at the moment. Expect big changes on that front around late October. (If anyone likes drupal/d3.js development and wants a side project; get in touch!)
  • New twitter account! It will have lpg and maybe some other things too.


Review: Keys of Sealing

Preview imageKeys of Sealing (RazorclawX, 2004) is a rare example of an AoS whose gameplay is formally divided into stages: imposing a structured path to victory rather than the blurred lines of “early-game”, “mid-game”, and “late-game”. It also has a special victory condition that is more teamfight-oriented than the familiar “destroy the main base”.

In this review I’m going to cover features from both early and later versions together, as there aren’t too many differences.

We don’t see an air lane very often! It activates later into the game.

The Book of Sealing:

Keys of Sealing‘s most distinctive feature is it’s unusual victory condition. Instead of a main base that needs to be destroyed, the game gives a cinematic introduction to the Book of Sealing which is located near the center of the battlefield. This Book is protected behind a layer of barriers, and three keys must be used to remove the barriers. Once exposed, having a hero channel for 2 minutes in front of the book (without dying) bestows victory.

The Book of Sealing, currently behind barriers.

Thus, the game is divided into three stages, of no fixed length:

  1. Securing the keys needed to control when the book will be exposed
  2. Securing an advantage to expose the book at a favourable time
  3. Fighting to control the area around the book for 2 minutes

The First Stage

Each team starts with one key in their base’s vault. This is a ‘sanctuary’ space: protected on all sides and barred by a gate. It’s not impenetrable, but at least one lane of barracks needs to fall before swiping a key becomes practical.

Outside the vault and occupying a central position in the base is the team’s goalie hero, an NPC tasked with guarding the entrance to the vault. It spawns at the start of the game, and levels up over time to acquire spells and stats. If it dies, it will revive after a short delay at the nearby Altar, ready to fight some more.

The goalie hero at his altar, with a few stationary troops to defend.

That’s two keys accounted for. The third is held by a boss called the Spirit of Destruction; located near the center of the map, and who must be slain before his key can be claimed. The Spirit is quite powerful and has an army of lackeys with him, though neither those nor the Spirit will respawn after death: so whittling his forces down over time is an option. Accompanying the Spirit are two Magic Vaults which contain loot, but take time to break open (and are only vulnerable after his death). This means that a team who gets the key can still be denied of the accompanying loot if the area is actively contested.

In earlier versions, the Spirit enjoyed a private chamber underground; teleporter access only.

The keys are individual items which must be brought to the Book of Sealing to dismantle its barrier. Once presented, a key’s purpose is fulfilled and it vanishes. This can be done one-by-one, but the barrier won’t drop until all three have been consumed. Until then, they are items whose location is known to both teams at all times. They will drop if the hero carrying them dies, and each key also provides a passive bonus to the hero who holds it: splash damage, frost attack, or purging attacks depending on the key.

Thus, a hero can remove the key from their own vault right away and use it for a combat advantage… though generally this isn’t a good idea until later in the game, or at the point where the vault is no longer secure.

An NPC called the Keyseeker roams randomly across the map, and actively snatches up any keys that are left unattended for too long outside a team’s vault. If this happens, he must be hunted down and killed to reclaim the stolen keys, though he respawns immediately to continue his task.

Even when none of the three keys are in his possession, the Keyseeker still has a purpose: he drops a special item called Heart of the Obelisk which guides the bearer to an otherwise inaccessible secret shop. It strikes me as thoughtful to have integrated the Keyseeker mechanic with some other parts of the map: it would have been so easy not to. Even though secret shops are no longer in vogue, this is great flavour for a map of KoS‘s era.

To wrap up: the first stage of the game has a good rhythm. There’s incentives to push down lanes, to grow stronger as quickly as possible, and as vaults become less safe: lots of encouragement to hunt enemy heroes and try sneak the boss.

The Second Stage

Once a team is in a controlling position with keys, they have the option to try and build an advantage before dropping the barrier, at which point either team will have an equal shot at contesting the Book of Sealing. There’s lots of ways to build an advantage that can be converted to a (hopefully) decisive victory… though no promises on anything going smoothly! Of course, building an advantage can happen during the first stage as well. Let’s examine some of the options.


There are two capturable towns on the map, which will spawn extra troops for whichever team owns them. A town is captured by destroying a modestly durable building in the center of the town; the other buildings are invulnerable and simply produce troops. Any 2-3 heroes who show up and focus the building are sure to succeed in taking it.

Once captured, the building will respawn for its new owners at about 20% maxlife, so it’s still very contestable if the capturing team isn’t committed. The ‘town’ building has regeneration (it will heal to full in 100 seconds), so it can endure unfocused fire, and returns to being a bit harder to contest if a recapture isn’t attempted immediately.

The towns differ in value; the ‘strong’ town to the west produces good troops and will hold itself indefinitely; the ‘weak’ town in the east will hold, but can sometimes flip back and forth without hero intervention. Having both towns under your control is a great springboard for victory.

The strong town’s Supply Depot is… a throne packing a flamethrower. Who knew?


Of course, the best thing about capturing towns is they make pushing easier, and in Keys of Sealing losing a barracks means losing troop production on that lane. Thus, it’s possible for the second stage to extend for a while if a team feels they can force the enemy into a permanent lane disadvantage, before dropping the barriers. Once that happens, the game will be focused around the Book of Sealing, and allocating time for counter-pushing becomes more difficult.

There are still options though: players can hire mercenary units, including siege specialists that can quickly recapture a town and provide a foothold on that lane. The other mercenaries are a building-repair unit, basic ranged attackers, a caster, and a big tanky unit. Their restock time is quite long, so building up an army would have to start early.

Another option for getting a troop advantage is paying for some of the troop upgrades available in the Altar, which follow the usual trend of +damage/+armour with high cost, and long cooldowns on each upgrade. I’ve written about this implementation for troop upgrades before, and I’m not a fan (though they’re less impactful in KoS than other maps).

Gold Income:

Close to each team’s base is their gold mine; sporting a long line of busy workers running back and forth harvesting. Despite appearances, they have no impact on player income at all, but can still be raided to pick up some last-hit gold (if you have AoE damage; otherwise they’re cumbersome to clear out).

Clearing out the enemy worms, for a profit.

What does have an impact on a player’s income is their wages! Every in-game day, heroes are paid by their faction based on their performance so far. The formula gives more gold to higher level heroes, more again for total player/unit kills, and subtracts for each death. (For this reason, gold mine workers can be used as a slightly more efficient way to inflate unit kills and increase a player’s wages.)

The idea’s cute, though this system hugely rewards early kills with both experience (that leads to levelling and higher wages faster) and a kill which will pay dividends for the rest of the game. Similarly, each death causes a small but permanent reduction in wages.

In games where losing gold on death is a mechanic, wages (or generally, deferring rewards) can help balance situations where two heroes kill each-other, but one hero benefits more because they died first (losing little to no gold) and then got a kill (awarding bounty), while the other gets a kill first but then loses that bounty with their death. Generally, “dying first” isn’t supposed to be a good thing, and waiting before paying out means that the death/kill is what counts, not the order in which it happened.

While they’re nice for flavour, neither of these gold mechanics contributes to overall gameplay. Numerically, neither system offers anything better than the gold for last-hitting troops and heroes, so their existence is largely a flavourful one.


Keys of Sealing has many quests around the map, taking the form “fetch this item for me to get a reward”. They often involve fighting hostile monsters. Most quests can only be completed once, so players might end up racing or interrupting each-other’s quests if they know the map well and feel the rewards are worthwhile.

Much of the game is framed in terms of quests: collecting the keys is technically the main quest. The second most important quest, which these days would be called a map objective, is killing the enemy’s reserve hero. This is a max-level NPC hero camped outside each team’s base, usually stacked with strong items and auras. The hero will sit in place, waiting patiently for two possible outcomes:

  1. To be killed once by the enemy players, and laid to rest for all time.
  2. For the barrier on the Book of Sealing to fall, at which point reserve heroes gain the ability to revive when killed. As we’ll soon see, this turns out to be a big deal.

Sitting in reserve for now, but those auras will make a difference later.

Finally, I’ll mention that there’s creep camps scattered all over the map so farming these is also an option for building an advantage (en-route to quests or otherwise). However, their returns aren’t great.

The Third Stage

The keys are assembled, the barrier falls, and the victory condition of “a hero channels in front of the book for 2 minutes without dying” becomes open to all takers. But as the stakes get higher, the factions up their game as well:

  1. A new lane opens up: air units will start flying directly to the Book of Sealing at the center of the map to contest it.
  2. The goalie hero and reserve hero will leave their posts, and fight their way to the book to channel on behalf of their team.

This definitely raises the level of excitement! The air lane has some spawning rules: the troops stand guard at the book instead of continuing to the enemy base, and troops won’t spawn if there’s already a wave on guard. (This seems fair.)

Trying to channel, but the enemy air superiority won’t let that succeed.

The second point, about NPC heroes being willing to channel, is a critical saving grace for the game. If a human player was forced to sit and channel, I would be very unhappy about it. Instead, NPC heroes can do it (but they have to be escorted to the book along the mid lane, as they’ll try to pick fights along the way). Players can still volunteer to channel themselves, and that might be strategically the right choice in some cases.

With this in mind, the reserve heroes are clearly a big deal. The extra max-level hero with all those auras; and willing to channel the book even when your goalie hero is occupied fighting off the west lane? Very helpful. If you have key control in stage two, taking out the enemy reserve hero is a top priority. That said, teams still have their goalie hero no matter what.

The air units can get distracted by enemy troops in their base, so if the mid or left lanes have lost their barracks and no-one’s holding them off, that can impact how many fliers make it to the center. Thus, even though taking barracks doesn’t lead to taking a main base, it still contributes an advantage in the third stage.

I think this works out as a pretty exciting victory condition in cases where there’s an even fight. In an uneven fight where one team has complete map control (two towns, reserve hero advantage, upgrades, possibly up a barracks…), it may be too hard to contest and turn things around. Well, put those wages to use and don’t fall behind!

Heroes and the Safe Zone

It’s not so clear from the map, but the team bases in Keys of Sealing don’t have any fountains for friendly heroes to heal at. This is because heroes revive, heal, and shop in a separate dimension from the main battlefield, which I’ll call the safe zone.

To leave the safe zone, players step onto a teleport pad matching their preferred lane (or the base), and are instantly teleported to the appropriate exit point. There are separate exit points for each team. All exit points can be used by either team as re-entry points, and re-entry requires a short channel.

The safe zone, with pads for each lane and a separate one for the base.

It’s nice to be able to revive and immediately jump back to your lane of choice, and once the game reaches the third stage, having a reliable central access point is great. (Revive times are  For general get-around purposes, there are also waygates in the corners of the map which connect opposite corners.

There are 30 heroes available for each team. They’re customised for flavour, and later versions have a couple of triggered spells. The items for sale are weak compared to other games, though the numbers all-round are low enough that +6 to a stat feels like it makes a difference. In general, heroes will be leaning on their abilities more than their item builds.


One of the things I like about Keys of Sealing is that it recycles mechanics from the first two stages (the goalie heroes, taking barracks, and the keyseeker) and keeps them relevant during the third stage. That aspect of the design feels really clean to me.

I also like that the lanes have a clear hierarchy: the strong town is the most important and will push the fastest, the weak town is useful and contestable, and the mid lane is hardest to push with, but gains importance over time as the boss becomes viable to fight, and controlling it allows NPC heroes to reach the book. The addition of a fourth lane in the final stages is a nice touch.

If the economic system were tightened up in a few areas (consolidate last-hits and such into wages, drop exploration quests for a central shop, give gold mines a clearer purpose), I think the game would have felt really sharp and been suitable for competitive play and development.

The lesson from KoS is that alternative victory conditions are a rich space to explore, and having the ‘form’ of the game evolve with new lanes and objectives is an area with lots of potential.

Download: Here