If you haven't already, please check out Part 1 which goes into the different ways FromSoftware and Deathloop handle their invasion systems. In this article, I will be highlighting more specific problems with these invasion systems as well as potential solutions to these problems.
Poor Host Risk/Reward Structure
While Deathloop provides an excellent risk/reward structure for the host, as killing an invader not only provides the host with weapon and power upgrades in addition to replenishing the host's respawns, FromSoftware titles struggle when it comes to giving the host an incentive to be invaded. Even though invaders do provide resources when it comes to leveling up, the amount is both relatively small and does not justify the risk of invasions. Furthermore, even if a host successfully defeats an invader, the host is typically left with fewer healing resources than they would have if they were not invaded. As a result, simply being invaded, even if you come out on top as the host, can feel like a loss should it drain your healing supply.
Luckily, this issue has a simple solution: provide better rewards for players when they win in an invasion and give hosts their supplies back should the invader be defeated. In Dark Souls, invading players could drop upgrade materials and replenish the host's Estus Flask when defeated by hosts in addition to the normal soul gain. While this method may improve host incentive to be invaded, it may also encourage griefing behavior by the host, such as setting up "ganks." For ganking in specific, you could have invaders drop greater rewards when hosts fight them without any cooperative help.
Invasions Encourage Passive Play
An issue that exists in both Deathloop and FromSoftware titles is that invasions encourage both parties to stay in a defensible position. Hosts are encouraged to hide behind summoned allies or traps while invaders benefit greatly from hiding behind various enemies within the level. This results in neither party wanting to actually engage the other instead of waiting for the other to grow impatient and overextend. Furthermore, complex level design can result in players being unable to find each other should they wish to fight head-on. As a consequence, you can simply end up with invasions where neither party makes any progress until the invader gets bored and leaves.
There's likely many ways to help alleviate this issue, but I have two in mind. The first is to push the invader to actively engage with the host with some sort of timer. An organic way to accomplish this is to have invader health slowly drain over time, eating away at their resources should they wish to wait in ambush. This also necessitates giving the invader consistent, accurate information regarding the location of the host at any given time as well as additional tools to combat host-only options such as summoning help. An example of the latter could be the ability to lure an enemy towards the host that scales in effectiveness for each summon the host has.
The second solution is to change the invasion setup entirely. Instead of the invader seeking to kill the host, both the invader and the host may have alternative goals. This could be killing a special enemy, collecting a certain number of items, reaching a location, etc. Whatever the goal is, both players have an active goal to pursue, providing pressure on the other party to succeed first. There certainly have been attempts at this with concepts such as the Mad Phantoms in Dark Souls 3 and Rat King Covenant in Dark Souls 2, these have never been the primary method of invasion.
Opt-in vs Opt-out Invasions
One issue that plagues FromSoftware titles is that invasions are often a form of punishment for players looking to engage in cooperative multiplayer. Should a player wish to reliably fight invaders by themselves, they often need to use a specific item to opt into play. However, this item needs to be used repeatedly the majority of the playerbase will likely never use it. This produces a rather small host pool for invaders, which can make invasion matchmaking rather slow, especially when combined with area and level-based matchmaking.
Instead of having invasions be an opt-in system where hosts need to actively pursue invasions with multiple uses of an item, invasions should be an opt-out system. All players should be potential hosts by default, and should a player wish to avoid invasions, all they would need to do is opt-out to disable invasions until they turn the option back on. Players are simply more likely to choose not to opt-out than they are to choose to opt-in, increasing the potential host pool. This is what Deathloop does, and it works quite well there. If this system would be implemented in a FromSoftware game, it could be activated by interacting with a particular location to maintain the more diagetic nature of those games.
Concluding Thoughts
While invasions will never be something everyone enjoys, it's worth looking at ways to improve the current experience. Invasion multiplayer is criminally underexplored and underdeveloped as a concept that I hope gets seen in future titles from various developers.
Next time, I'll be switching focus over to tabletop games, discussing different dice systems and ways that they are useful.
Comments