Review: Dawngate, Part 1

RD_PreviewFor the first time, I’ve decided to review a commercial lane-pushing game: Dawngate (Waystone Games, EA, 2014). It presents itself with the goal of usurping the “meta” that’s supposedly been a menace in other genre-games, and in practice aims to refine the wildly successful League of Legends formula. It implements a number of decisive design choices, many of which are ambitious and well-motivated. The game also earned a positive reputation for its emphasis on lore, with elaborate character introductions and the community participating in a supplementary webcomic.

Despite its best efforts, Dawngate announced its closure only 6 months into open beta, and its servers will shut down, presumably for the last time, on the 5th February 2015.

A simplified diagram of Dawngate’s terrain.

Dawngate‘s terrain is as full-featured as one would expect from a commercial lane-pushing game, so we’ve got quite a bit of dissecting to do.

Guardians:

The victory condition in most lane-pushing games is to destroy a particular building located in the deepest part of the enemy’s territory. The same applies in Dawngate, with the twist that this final building (called a Guardian) is equipped with five powerful abilities with which to defend itself. Its behaviour resembles a typical pattern boss: whenever enemies are nearby, it will randomly choose from its available abilities, play a short warning animation, and then execute the attack.


A Guardian, just as it’s about to fire its laser.

The Guardian is surrounded by five Cores: small buildings which are connected to the Guardian. Destroying a core will disable one of the Guardian’s abilities for the roughly five minutes until that core respawns. The cores at the back of the base (which are harder to reach) correspond to the most dangerous abilities.


All five Core buildings for the bottom team.

Unusually for a defensive structure, these abilities are designed for harming enemy heroes rather than troops, and the Guardian’s targeting priorities strongly reflect this. The abilities include a dangerously high-damage laser beam, mass-projectiles that stun or slow, and persistent damage-over-time in a line. The warning animations on each ability give enemy players a very brief window in which to dodge, which is manageable in isolation, but the purpose of the Guardian is clearly to give the defending team a teamfight edge, by providing that extra distraction or opportunity to snare someone inside a laser beam and turn things around.

The idea is an interesting one, and due to the wide-open space around the Guardian, the act of defending is less a matter of turtling in a choke-point while trying to zone enemies out, and much more a matter of hero versus hero combat. It feels like a step in the right direction.

Parasite:

Right in the middle of the map lies a major map objective: a boss creature called the Parasite, which is Dawngate‘s version of Baron Nashor from League of Legends. It requires the combined efforts of several heroes to kill, and grants bonuses to whichever team lands the finishing blow on it. The Parasite evolves over time:

  • When it first spawns at 5 minutes into the game, it will grant a flat gold and experience bonus to the killer’s team.
  • After 15 minutes, it also grants the killer’s team with a buff that improves regeneration and damage output.
  • After 25 minutes, it additionally spawns several waves of Striders: super-troops which march along the lanes and serve as a powerful pushing aid.


The three Parasite evolutions.

The close resemblance with Baron Nashor is disappointing, though the new central location makes it relatively accessible at all points of the game, as well as removing asymmetry concerns. Being located in an open space rather than a pit is a mixed bag: it makes stealing the objective easier (a plus for tension), but also reduces the commitment required for the objective, since backing off is easier. The Parasite should always be warded by both teams, so it’s nigh impossible to sneak one under the radar.

The Parasite has a few abilities that it casts, much in the same way that the Guardian does. These include throwing bombs at random locations, and an energy breath. The motivation is of course the same: to put teams under pressure should the enemy show up.

Spirit Wells:

A new objective in Dawngate are the Spirit Wells, a special type of control point. Each well generates workers over time, which harvest from the well and grant gold to the team that controls it. The longer a well stays in a team’s possession, the more workers it will spawn, and hence more gold it will passively generate (up to a maximum). There is one well tucked away in each of the map’s four corners.


Workers gather from three directions, making nuking them more difficult.

Standing on an uncontested Spirit Well for a short time will cause it to change hands, killing off any living workers (along with awarding their bounty), and causing it to become locked for four minutes. It will start generating workers for it’s new owners after 90 seconds. A locked well can’t change hands, but the workers it generates are always vulnerable. Killing a worker grants gold and experience, so “clearing” wells of their workers can be a valid source of both income and income denial. At the start of the game, all four wells are locked, with two controlled by each team, and they unlock after 15 minutes.


The white pattern along the circumference shows the remaining time before this well unlocks.

I like that the control points are interactive game elements. Hero abilities can be used to clear the workers, which feels satisfying. I also like that killing workers provides income, so there isn’t a deferred reward for their capture.

What I don’t like about Spirit Wells is that they are an isolated subsystem within the game. They can be pressured at any time, independently of the lanes or the general game state, and demand that someone respond within 15 seconds, in a game without teleports. Even if a defender arrives, the two heroes will be on an equal footing, since the Wells are in an open field. There is such a low level of commitment to capturing that players can simply walk away if the well is contested, gaining gold from the workers and wasting the enemy’s time. In practice, the interaction feels sporadic and forced.

Jungle:

There are also a plentiful selection of jungle camps around the map, some of which are specialised to give lots of gold, while others reward a buff for 210 seconds. The buffs, as in League of Legends, are an important resource for jungling characters, but not something we need to cover in any further depth in this review.

Map Layout:

We’ve now been through the major features on the map, which have included some interesting ideas in isolation. But looking at the whole picture, there is something fundamental that I identify as being wrong with this terrain: there are only two lanes.

Why should that matter? In any lane-pushing game, the lanes exist as a frame for everything else which happens: they are the structure inside which all other calculations of economy and strategy take place. Where the troops meet, both teams share mutual vision, and have an incentive to be in that space thanks to the income from troops dying. That incentive to be on-lane is the basis upon which all broader strategic decisions are made. Whether particular heroes are present, or have some reason to be absent, allows us to infer what the enemy plans might be, and formulate a response.

In a 5v5 matchup, two lanes simply don’t provide enough data points for teams to meaningfully infer each-other’s plans. But even worse, the only asymmetry in the two lanes which are there, is the minor detail that the positions of the Power and Damage Reduction buffs are inverted. Otherwise, the two lanes are identical. Consider the nuance of the safelane/offlane in Dota caused by the outer lanes having different shapes for each team, or the implications of being top/bot in League due to proximity to different map objectives. All that nuance that could have conveyed information about the enemy’s plans is also missing.

Another issue with having only two lanes is that map features and objectives end up outside the lanes. They divides the map into additional subsections, making it much harder to correctly gauge where an enemy hero might be.

Furthermore, consider that when objectives are outside the lanes, any hero which goes to tackle one will find themselves limited in their options of where to go next, as they are often only adjacent to one lane. Indeed, the Spirit Wells are tucked away in corners where there is almost no choice of where to go next at all: either back out of the corner, or teleport home. This is awful for the game’s fluidity, and makes Spirit Wells a chore to take or defend. I am sure Waystone struggled with trying to make players want to interact with Spirit Wells, and it is simply because they are placed in a corner where no-one wants to be.

Arguments in favour of having two symmetric lanes include putting less pressure on certain heroes to be in a particular lane due to the “meta”, and less arguing among teammates. We know those are the kind of things Dawngate aimed to do, and from that perspective Dawngate’s map layout is a great success! But in exchange, it loses a great deal of strategic texture.

Shapers and Resources:

The heroes of Dawngate are known as Shapers, and break some lane-pushing game design conventions.

First and foremost, they do not share a widely-used resource like mana. Instead, each shaper is treated individually, only being given a resource (usually a unique, shaper-specific one) if it fits their playstyle. Many shapers have no resource, and can use their abilities without cost (restricted only by cooldowns).

Consequently, most shapers are simpler to learn and play, while there is still room in the game for those more extreme designs which are counter-balanced by a resource. The shaper-specific resources tend to reward players for landing abilities or attacks, which is a friendly way for the designer to guide players towards the shaper’s intended playstyle. I like this philosophy of encouragement.

The overall feel of the game closely resembles League (instant turn rates, cast ranges, etc.). On a more granular level, the shaper skillsets are fairly conventional and don’t push any boundaries, but that’s not a point of complaint for a game that’s in beta. I do like that the majority of abilities are aimed.

Stats:

Another refinement that Dawngate makes is trying to simplify shaper stats. To illustrate:

  • Power: Increases a shaper’s total damage output.
    10 Power = +10% damage
  • Haste: Decreases a shaper’s cooldowns, and increases movement speed.
    20 Haste = 20% cooldown reduction, +8 movement speed

These two stats are shared by all shapers, and apply to both attacks and abilities. Hence, items with these stats are useful to every shaper. This avoids the situation often found in games like Dota2, where an item like Abyssal Blade has a useful active, but it’s coupled with an expensive +100 attack damage that isn’t efficient on most shapers. Giving items more broadly useful stats like Power/Haste decouples the item actives from specific build paths.

The system seems elegant at first, and players are definitely presented with simpler choices when it comes to item shopping. There is a trade-off though: if we offered percentage-based stats which improve “everything”, shapers would end up being able to do everything. They would get strong sustained dps from attacks, and be able to spam powered up abilities. The game would becomes less about a shaper’s playstyle, and more about which has the best numbers.

To resolve this, each individual shaper has assigned ratios which determine how much benefit they get from Power and Haste. Let’s consider the shaper Ronan:

  • His attack damage is increased by 1.0% per point of Power.
  • His attack cooldown is decreased by 1.0% per point of Haste.
  • His ability damage is increased by a variable amount with Power, detailed in each tooltip: “deals 35 (+30% Power) damage”.
  • His ability cooldowns are decreased by 0.15% per point of Haste.

We can see that Ronan’s abilities almost ignore stat bonuses, but his attack is improved greatly. This is reflected in his abilities, which centre around mobility and boosting his attacks. So if you pick Ronan, you’re playing an attack-damage carry.


Laning with Ronan.

For contrast, let’s look at another hero called Amarynth:

  • Her attack damage is increased by 0.2% per point of Power.
  • Her attack cooldown is decreased by 0.2% per point of Haste.
  • Her ability damage is increased by a variable amount with Power, detailed in each tooltip: “deals 35 (+75% Power) damage”.
  • His ability cooldowns are decreased by 0.60% per point of Haste.

So we can see that Amarynth gains a mere fifth of what Ronan would get towards her attack, but she makes up for this in her abilities. Because these ratios are fixed, she will never achieve good dps with her attack.

(Note that building exclusively towards attack or ability damage is still possible, thanks to a few items with special abilities that specify only improving attacks/abilities. However, there are no item special abilities that provide attack rate, so the ratios still hold a lot of sway.)

My conclusions about Power/Haste are as follows:

  • They simplify a lot of tooltips.
  • They decouple item special abilities from attack or ability focused builds.
  • In exchange, shapers themselves become coupled to attack-focused or ability-focused playstyles

We can conclude that Dawngate’s approach is to present shaper designs with a relatively rigid playstyle, and offer lots of ways to build upon or explore that playstyle (items, spells, loadouts, roles). If you trust the designers to provide a good selection of shapers, I think that system can work and achieve a good degree of diversity. It does rely on players investing the time to find the right shaper(s) for them, and consistently being allowed to play them, but these are relatively small issues and easy to overcome.

Wards:

Shapers have an innate ability to place a ward every 3 minutes, at no cost. Generally, they are placed inside tall grass to scout ganks, or at the Parasite. Since there’s no way to detect or remove wards (read: no counterplay whatsoever), the mechanic does nothing to improve gameplay, and instead adds a bunch of unwelcome guesswork. This implementation is strictly terrible.

Continue to Part 2

Advertisements

One thought on “Review: Dawngate, Part 1

  1. I’ve been waiting for this review, because I find that Dawngate has done a lot of things right in the sense that it’s a relatively polished and complete game, while still taking some risks in the form of features that aren’t in the mainstream. I really wanted to see what you had to say about itemization, as well.

    Spirit Wells: I think the notion of spirit wells that generate income and can be harassed is a decent one, and I can understand how the pitfalls suffered came about. I can imagine developers discussing it as being ‘like tides of blood vaults, but relevant throughout the game and with less snowballing’ – sadly this wasn’t the case, however, I do think they could make minor adjustments to make them viable. Dawngates distinct lack of global ultimates *in addition to* lack of teleports makes them a bit silly. I also think that the ‘locking’ phase was a bit of a noob trap and/or could have been more clear. Other potential options for adjustment would be removing the zone control element and instead reducing worker respawn rate – this would probably make a marginal reduction in snowballing to Dawngate, which I think would be welcomed. Major changes to gameplay mechanics, like pausing lanes from spawning for n minutes every m minutes could make the zone control element more dynamic. I don’t necessarily think any of these are better solutions than just having things like global ultimates and teleport, however. Disclaimer: I think giving spirit wells static defenses and making them die, like vaults in ToB, would be the most interesting option for Dawngate.

    Map Layout: I think having two lanes is interesting, and I can understand the desire for symmetry in a genre full of angry adolescents who want fairness. I think two lanes, with some creativity, could easily be made more interesting, with or without asymmetry. Having more minions (satiation), or more powerful minions (like the dota franchise) could make paired or even triplet lanes worth consideration.

    Shapers, Resources, Stats, Items: This one is dangerous. I have mixed feelings about the stat system. I think it’s interesting that Dawngate developers wanted to improve clarity for beginners by ensuring item tooltips are as brief as possible, but I think it’s terribly misleading that stats work differently on different shapers, and terribly one-dimensional that most shapers can only be optimized via a very small set of item permutations. If you want to improve item clarity over league of legends’ item system, you should (a) remove (or at least make them ‘special’) stats like attack speed, armor penetration, magic penetration, and movement speed; (b) simplify build options by reducing the number of base items for any one stat (e.g. longsword, pickaxe, bestfriend sword), and (c) reduce the number of high tier items in general, especially ones with (i) active effects and (ii) passive effects that provide little gameplay with respect to tooltip complexity. Disclaimer: I don’t think Dawngate should have done any of these things, as I think the level of complexity for league of legends items is at a pretty nice level.

    Beyond the obvious: I stopped playing dawngate for 2 reasons: Firstly the minimum graphics settings felt laggy on my machine, without providing anything better than other, similar games. Secondly, I couldn’t find any shaper with enough depth to suit my desires in the moba genre. Could I have learned a lot about timing and teamwork by playing Dawngate? Yes. But I can do the same in League of Legends, while playing champions that are more interesting for myself.

    Looking forward to the next part.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s