Interview With Jim Mummery Of Edge Case Games, Developers Of Fractured Space Part Two

I recently had the chance to sit down with Jim Mummery, the Chief Creative Officer of Edge Case Games who are working on Fractured Space. If you've yet to read the first part of this interview, you can do so here. In this second half of the interview, we discuss open development, lore, in-game upgrades, and more.

How has open development worked for you so far?

Mummery: It’s been fascinating. It’s hard to get away from the fact that when you start it’s quite terrifying. It’s everything you’re kind of told not to do. If you work with a major publisher, or you’re working on a major game, you’re often told not to go on the forums. You’re told not to talk about what you’re working on.

This is completely different to that. We have Q&As every week where guys ask us questions and we respond as well as we can with what our intentions are. We have Thursdays, so we have our playtest night those nights, and our guys will jump on TeamSpeak and talk to them about what we’re doing and what we’re planning on doing. We put updates out on not just what’s coming, but what we’re working on. I haven’t looked at Trello in ages—I’m not the guy that runs it—but the plan is to be very open about it. It’s very freeing because you can’t really say the wrong thing. You can’t really tell something about it because everyone already knows.

It’s been really good. We’ve got a whole group of hardcore players in our community that tune into every one, that always chase up the state of any specific feature, that query how new ships are going to work, that basically ask all those questions that you’ve got to make sure you’ve got in place.

I’d say my experience with open development has turned me around from being very scared about it to kind of being an evangelist for it. I think, if you’re going to put a game out there in Early Access you have to adopt it so that you’re not wasting all that information, talent, opinion, and educated points of view that your playerbase has. I think it’s good to have a vision of where the game should be that’s isolated from that, but I don’t think you could ever reach that vision of the game unless you’re listening to the people who are telling you about what you’re planning or what they would like to see in the game. They have to go hand in hand.

I saw on Trello that you are looking into providing AI, bots, and player-hosted servers. Are there plans to provide a way for players to play independently of the master server?

There are a couple of things. The concept of a “player server” is kind of required for a new player experience. So part of the experience of a new player coming in will be a number of tutorial-style levels where we teach them basic aspects of the game. Those have to be offline. You don’t want those on the server where you have to wait for the server to send down. You want to be able to jump into those things very very quickly. They’re not multiplayer games so you don’t need that server-client relationship.You want a local server so they can run that at any time.

The same is true of the evolution of those levels. It’s the concept of a “Proving Ground” where you can test out your ships and test out your weapons isolated from the multiplayer servers. So that’s aspect one.

About that part of Trello, the AI side is the need for being able to put ships into that environment where you can battle different ones. You might want to try a new ship you have or against a ship your friend has. We want that environment, the Proving Grounds, to allow you to do those things. But that requires the game to be able to play against you.

As you might know, AI in things like MOBAs is incredibly difficult because each champion has a completely different set of abilities that must be triggered at a completely different set of times or in combos, or in variations of combos at completely different times. So most of the AI and things used for training or PvE play in things like League of Legends, Heroes of the Storm, or Dota 2 is the AI is written for each champion. It’s very complicated but in order to do it properly we have to go down that line. We have to make a unique set of behaviors for the way the Hunter would work as opposed to the way a flagship, a sniper, or an assassin would work.

Those two things basically come together and allow us to make an interesting training experience for new players. Make an environment where you can play and test your ships yourself away from the game. All of these things help make you a better player. They help make your experience in-game better. Eventually we could start introducing the concept of cooperative or PvE modes that also become a part of that experience. So World of Warships runs through PvE before it reaches the main game; many League of Legends players play against bots before they go into the main game. I think helps give you that smooth movement from not knowing a lot about the game to being a competent player. So all of these things help that experience.

And then bots can also replace leavers?

Possibly, yes, but that’s also really dangerous in terms of getting a bot that good. Honestly, if the player has a lot of trouble, then the bot might be better. It’s a dangerous thing to try and do. If we could get our bots that good then that would be great, but our first port of call is ensuring that, when you get into a game, you’re in a well-balanced, well-matchmade game so that there’s no reason for you to leave.

If you leave, or you’re disconnected, we can put you straight back in that game. That’s the first thing that we have to do. Replacing people with bots in a live game, it’s tricky. If we can make AI that good, I think we’d be very, very happy and shouting from the roof about it. We’ll see.

Can you talk a bit about what the second pass at Proving Grounds will be like?

At the moment there are a couple of isolated levels that you play. The plan is that you will move through these levels that introduce concepts to you. Eventually it’ll be much more like training grounds in World of Tanks, to a certain extent, where you have a familiar map that you can spawn into. Ideally that you can spawn enemy bots into and attack and try out various weapon systems and attacks against different ships. Those bots will start out very simplistic, but the hope would be to develop quite competent ones where the bots know how to use individual ship systems.

Beyond that, we would like Proving Grounds to be something that can access an external server and you can bring friends into it. So you could set up custom matches. The community’s kind of done that in our servers. We have server selection at the moment. When matchmaking’s off you can select a server. So a lot of the guys would all just jump into it and try out some crazy stuff. There are some great videos of all-Destroyer matches, with Destroyers battling out across the lanes, and a whole load of similar kind of odd versions of the game. They’re all awesome and we want to give you the ability to do that in the game through the Proving Grounds. You could invite your friends and have a custom match, pick your own ships, or set your own rules within the concept of the Proving Grounds.

That’s way off yet, but we want that portion of the game to be where you go to test stuff, where you go to learn stuff, where you go for a laugh, where you go for quick games outside of the main game environment.

How are in-game upgrades planned to be implemented?

We’ve been talking about this a lot lately. We have a habit of doing things in phases. There’s a first implementation where we get stuff in, we test how it works, we see how it goes, and there’s a further implementation that requires a lot of content. Currently, when you level up in-game, we adjust a series of parameters on your ship. Depending on what your ship is, it could improve its main health. It could improve its cooldown. It could improve your refire rate. It could improve your capture rate, the amount of damage you take from an enemy base.

All of these parameters mean that your ship gets better as you move through the game. So the more mines you capture, the more kills you get, the more assists you get, you’re pushing your team level up. Your team is getting better. But all of those modifiers are applied to your ship automatically at an owned base. That works pretty well, but what’s missing is the player’s ability to adapt their ship and their team to the enemy team.

So the idea of allowing players to select upgrades would mean, at certain points in the upgrade process, you would pick a path from one of two options that would allow you to improve maybe your offensive capability of your ship, your defensive capability of your ship, your utility of your ship, or your movement of your ship. You’d pick these paths at, say levels three, six, and nine, and they would allow you to go down specific paths that would make your ship faster, stronger, or harder to kill. Because all of the ships are unique, those paths have to kind of be unique between ships. So there’s a lot of work in building that system, even if all we’re changing is modifiers on the paths.

So implementation one is taking the system we have at the moment and adding options. So you’re picking offensive, defensive broad stroke options down a tree and maybe going all offensive at the risk of everything else, focusing on movement so you can take more captures, or focusing on utility because your ship has systems that would benefit. That’s implementation one.

Implementation two is a whole other system that involves a mass amount of content work where, within the context of those options, you’re altering the system that we use on your ship. If you’re picking offensive, you would see your guns change as you move through the offensive ladder, you would see the way your cloak works change or the way your blink works changing. That has a lot more work involved because we have to duplicate systems and create more work for each of the systems.

It’s something we talked about a long time ago where you’re adapting your loadout to the team rather than the modifiers on your ship. Once we’ve got the modifier option in, we’ll look at the practicality of building all the options for the loadout swap. That’s kind of the two-tier version of how in-game level ups would work.

So it’s going to be more Heroes of the Storm and less Dota?

Absolutely. We love the way that Heroes of the Storm approaches level ups. Heroes of the Storm has made a lot of interesting decisions in terms of trying to simplify the MOBAs that existed. I think there are two things that we took away from it.

One of them was the way they’re dressed. Obviously, they don’t have a shop, but the way they have upgrade options, as opposed to a shop. We never intended on having a shop. We’re in 3D. We have multiple ships. There’s enough complexity in the game already that we never intended to have a shop, but we did want players to alter their makeup based on the enemy and the original idea was to allow them to swap loadouts and have loadouts level up. Now that’s an incredibly complex system to add in and the more ships we add, the more complicated it gets.

The idea behind allowing the modifier level up is that it allows us to put in a system where you have a branching structure, set up the UI, set up how powerful ships can get, how different the branches can be, while still being noticeable but not breaking the game. And then, on top of that, we can support weapon changes that support it.

We thought the options that they give in Heroes of the Storm for level up were a really cool simplification whilst allowing the player to make decisions based on their performance. We really liked it.

How are you planning to implement lore?

Lore is something that comes down to me to do usually. I’ve been kind of busy trying to get the game together to a certain extent. But also, my experience with development has been that, especially with something like open development, whilst the community is being very vocal about wanting more detail on the universe, there are things about the game you want to evolve outside of the restrictions of lore. As soon as you lay down what the universe is and how it exists, you have to start obeying the rules that exist.

For example, we originally wanted to have a context for everything. We wanted capture to be marine shuttles that would fly from your ship to the station. We wanted healing to be drones that repair your ship. We wanted mining to be mining drones. We still intend to do all of these things.

But the first thing that happened when we started talking about adding a new healer ship was that the guys wanted us to make a heal beam. It was something that went against the plans that we had because a beam that heals a ship doesn’t make a whole load of sense. But we had a beam system, we had a heal warhead. These things were easy to put together. If we’d had lore in place, I think we would have hesitated about building that ship, which is the Corvette Protector. And that would have prevented us from making a very disparate type of healer from the Disruptor. So there are points when your game is flexible and evolving where lore can get in the way.

That said, very, very early on, I wrote down the basic structure of the universe. Factions, manufacturers, the evolution of the universe. At the moment, we’re actually in a human-only universe and there’s a point not far down the line where we’ll move into a more galactic community and bring in alien manufacturers. There is a basic structure set up, but what I didn’t want was that to get in the way of making the best game we could.

However, we’re at a point now where the game is relatively stable and I’ve handed over all of my stuff to another writer who’s going to take it on and build up our universe using the information I’ve given him. We’re really looking forward to getting that going.

As to how the lore exists in the game, I think the lore cannot get in the way of the player experience where you’re able to jump into a battle, play a game, and create your own narrative to a certain extent, your own history. It will be on websites. Hopefully, we’ll build a codex within the game so there’s a nice place in the frontend where you can drop in and discover the history behind the technology, the ships, the manufacturers, the factions, and the crew members, to a certain extent. Most of the crew out there actually have bios that were written for the voice over artists.

So we’ll get all that stuff into the game in a shape. That aspect of it wouldn’t stand between you and having a game, but it would be there if you choose to do it. As with everything else, there’s a step beyond that. I spoke about this before on the team chats we have. There’s an ambition we have that, once the lore is in the game and everybody understands the universe, what we want to do is have it evolve season by season. The ambition is that we will use any metric we can take from the game, because obviously, we need to track how ships do, weapons do, teams do, how quick games can last, how quick games can be. Basically anything we can do metrics for.

As a result we’re also aware of the crew people take into games and ships people take into games. We want to set up a system where all of those metrics feed into how the universe evolves to a certain extent. That’s quite a big thing to say, but how this works is that our manufacturers in the game at the moment aren’t sides. They aren’t factions. They’re corporations that sell ships to, frankly, almost anybody. But above that that does set a group of factions that exist both in the human version of the universe and the human-alien version of the universe. Those are some groups at war for reasons I won’t quite go into now.

But the corporations themselves and the crew will all automatically have relationships with either multiple factions or some factions. So when you’re taking your crew and your ship into the game the results of that game apply to those relationships. So if you have a really really good game, your crew will have a really really good game, your manufacturer will have a really really good game, and that will feed the relationship to the factions that exist in the universe. So those metrics would push how those factions do in the concept of the broader universe. We want a galaxy map to support this.

The idea is that, season by season, these relationships would change and you’d see the player behavior change the way the map works. Not in a direct way. Not in an “I pick faction X today because I feel like it” way, but in picking the ships and the crews I like to use. Those would influence the narrative from season to season. At the end of a specific season, we would take the way the world has changed and break it into a narrative that explains what happened to cause this change between last season and this season.

The idea is that every game has an impact on the evolving universe and I imagine that, when we put that system in and play with it, there will be a lot of iteration on trying to make it interesting enough and have a big enough impact into the universe for people that care.

--

You can play Fractured Space on Steam now. The Early Access phase requires a minimum buy-in of $9.99 for the time being.