Dean Hall and his team at Bohemia Interactive may have missed their December release window for the enhanced standalone edition of DayZ, but it’s for a worthy reason. We spoke to Hall about his plans to completely reinvent our 2012 Game of the Year.
PCPP: Congratulations on receiving PCPP’s Game of the Year!
Hall: Even to be nominated, let alone win it, is really quite an honour. We’re really very humbled. It’s hard to describe how you feel about it, because there’s so much that goes into it, and there’s so much that comes out, that you don’t tend to look at it and say, ‘Wow, what a great effort!’ You say, ‘Wow, there’s a lot of responsibility that comes with it as well’. We’ve got to be very mindful of that going into the standalone edition itself.
What’s it like being back at Bohemia after a break?
There’s a lot of work to do. I’m getting straight back into it. But it’s good. When I was back home, that was something I’ve been wanting to do. After I had a little bit of a rest, I really wanted to go back and sink my teeth into it.
Has anything changed about the way you’re approaching development after having this break?
Yes. I think it’s a combination of the many events that have happened over the past month, as well as taking a break. It definitely changed my approach to the project going forward. I think, to summarise that, I feel like, not just me but the whole team, we’re more just focused on making the game as we think it needs to be, rather than worrying about some of the politics and all the other stuff that is going on around it.
We’re more just focused on making the game as we think it needs to be, rather than worrying about some of the politics
So we’re happy. We missed that target of December, and I think we really did need to do that date – but we couldn’t. So now, we just want to make the best game we can, and if less people end up playing it, then less people end up playing it.
What’s the state of the team at the moment?
DayZ is developed in sort of a cell nature. There’s a core cell which contains three programmers, myself, a production assistant, our executive producer who’s our company CEO, and we have another all-hands, able to do all tasks, production/designer. Running alongside that, we have various art cells, ranging in size from one person for our map designer, through to a cell of about eight artists.
These cells work semi-independently on their area, with everyone liaising together. It’s working not too bad, considering the project had no time for planning or anything.
Do you find yourself still programming? Have you shifted to more of an administrative role?
I’ve definitely stepped back. I feel like I’m doing more production work than I’d like. But that’s just the nature of where we’re at. In the last few days – maybe that’s why I’m so enthused going forwards – I’ve finally been able to step back and see the design ideas I had taking shape.
As we go forward, I’ll end up doing a lot more of using our own in-game scripting language, because now we can start actually building on the architecture that we’ve built. What we did, last year, about November, we saw that there was an opportunity to really redevelop the engine. There were two paths we could take: one, was to release DayZ as this, I think, adequate product, that was the mod polished and tidied a little bit, packaged and released. Or, we had this riskier, but much grander, option of redeveloping the engine from the ground up.
We took that risky option, and I think it was nervous, and we weren’t sure how that was going to work out. Not only did that cause us to miss that release date, but there were times where we thought we might have to scrap it and go back to where we were. But what we’ve learned from the last few days is that that option has worked. We now have this excellent architecture to build from.
Tell us more about how you’ve redefined the server architecture. In comparative terms, it functions a bit more like an MMO, with the server taking much of the workload, to combat the hackers?
There are two key aspects of the architecture now that have been proved. One of them is the server-client architecture. That is what the purpose of this technology test that we just announced is. That is to prove that this new architecture works. I think there’s going to be a lot of re-work to come out of that. I want to say it’ll be straight-forward, but that would be dramatically undermining that Andre, our lead programmer, is putting into that.
What that does is directly address issues like hacking. It addresses key performance issues. It addresses the number of players that we can effectively have per server. We can have a vastly increased number of players. It’s going to introduce a couple of little bottlenecks in terms of bandwidth and lag that we’re dealing with now, but none of these things are insurmountable.
Individual items being able to transmit disease – you loot a corpse, and pick their pants up, and that pair of pants is contaminated with cholera
The second lot of architectural changes, and one that really excited me at the moment, is around the inventory. We’ve completely revitalised the inventory system from the ground up. This is allowing us to implement a whole bunch of design ideas that were only dreams previously. Things like individual item durability. Individual items being able to transmit disease – you loot a corpse, and pick their pants up, and that pair of pants is contaminated with cholera. Additional weapon components, batteries – all these things are now possible.
Is this still a belt items and backpack system?
We’re looking at a very basic implementation. It ends up with containers. But we’re going to start with a very simple implementation and stick with where we are, get that working, and then we might have more advanced containers come along.
I took a lot of inspiration from XCOM, with the basic design, which actually went through a lot of revisions. It’s essentially been designed and constantly redesigned since September. The idea is to make it really intuitive, so that when you look at your inventory, you know where different items can be placed.
Is this still being represented with a graphical user interface?
We’re representing it in a user interface. You’re bringing up the screen, and you’re seeing your UI. Instead of having little pictures, we’re actually using 3D items, so that you actually get real feedback on what you’re looking at. The idea was to produce something very visceral, because inventory is levelling for DayZ. I think Zero Punctuation put it very well – in DayZ, there’s this great equaliser of your weapons. If someone else has a lot of equipment, and you don’t – but you have a pistol and they don’t, suddenly the world can equalise very quickly. Or vice versa! They can suddenly turn things back around on you. It really results in some interesting experiences, instead of working at the game for 21 hours, and being worse at someone who has been working at it for 30 hours.
We felt that inventory epitomises that. And that’s why we’ve invested a lot of work into it. There is a lot more of a feeling of it being in-game, though. Instead of you interacting with piles of equipment, you’re actually interacting with the individual cans and the individual parts of the world. That was something we felt quite strongly about in lessons we’ve seen from a game like Minecraft, where we felt you were really interacting with the world directly.
Does the inventory component aspect apply to weapons, too? Can you scavenge a barrel, a slide, and a firing pin?
We haven’t gone as far as that yet. What we’ve done is confirm that the architecture of this works. It’s truly ground-breaking. It’s gone from the previous way ARMA stored weapons and magazines as strings. Now, that weapon is an object, and when you place that object on the ground, it doesn’t lose its information – it knows how many rounds in its magazine are left. But it can also store variables like its damage, what it’s come in contact with, who held it – all this information is suddenly persistent. So it allows us to have weapon customisation, such as attaching scopes. A lot of that requires us to do more work on them, but the fact that we can do it unlocks the development for the next twelve months for us to do everything you can imagine with inventory.
With this more complicated inventory system, do you feel it will be harder for players to achieve, for lack of a better term, an ‘endgame’ status, where they have the best weapons and spend their days joyriding in a helicopter around Chernarus?
I’d like to say yes, but in my heart I know the answer is no. Because if I’ve learned anything about players, it’s that they figure things out very quickly. I was watching on Reddit a person lamenting that they can’t figure out how the zombie aggro works in the mod – so zombies are scary for them. I was telling them that they should actually enjoy that, because once you figure out how to game a system – and you do, it doesn’t matter how much you don’t want to, naturally your mind will figure out the pattern – I feel like it loses a bit.
We want to have enough vagaries and softness around the edges, that it’s quite difficult for you to naturally figure out how to game the system
We want to put enough in there and have enough vagaries and softness around the edges, that it’s quite difficult for you to naturally figure out how to game the system. I think it’ll be a combination of how this new inventory system works, of the weapons as entities, continued revisions to the zombies for the next year, as well as some changes to the audibility and visibility system. Those things combined hopefully should make for a much more intuitive feel, which I’m hoping will make it less obvious to you how to game the system.
Do you want to bring back more of a threat to the zombies themselves and encourage players to cooperate against this common enemy?
Yes, we absolutely want to do that. That’s an area we haven’t done a lot on at the moment. We’re kind of leaving that. It’s a known thing, and it’s something we know we can fix pretty easily. It’s just a matter of our AI programmer like Andre to sit down and tweak it.
But we definitely draw the line at trying to force players to do something that’s good and that’s wrong. That’s maybe part of DayZ’s success, if I could be bold enough to choose a reason. It doesn’t force players to play in a certain direct way. Or, at least, it doesn’t come right out and declare that ‘this’ is the way you must play DayZ. If you want to be anti-social in DayZ, you can be anti-social. There isn’t any real consequence to that. There are some, but that’s how we want to keep DayZ. It’s part of the magic.
What we need to do is provide more options. Most people in-game are just going out and killing other players – because that’s all there is to do once you’ve figured out the relatively simple mechanics behind it. So if we provide more options, and more paths for people to choose, I think that more people will follow them.
Are you baking more social dynamics and mechanics into the game, or are you happy to leave those interactions up to the players?
Probably a bit of both. We will experiment – and that’s what this is all about. For me, and the game as well, people responded well to an experiment like Minecraft. But it’s no secret I follow Kerbal Space Program, FTL, and Prison Architect. Players responded really well to those experiments, so we want to continue like that as well. Social interaction is one of those areas we want to experiment with.
We’re looking at ideas, and we haven’t actually implemented these yet, but we’re looking at how we could have it so that some skills, maybe, are socially learned or enhanced. Maybe very advanced skills, like being a doctor, the ability to do very specific things like suturing, might be earned those ways.
We’re looking at how we could have it so that some skills, maybe, are socially learned or enhanced…like being a doctor
We’re still undecided about that, because we really want to have the game avoid things like specific skill trees and stuff like that. But having it socially developed as a skill, or something you get better at through repeated use, and more of those options for social interaction like suturing, setting fractures, these kinds of things are social options. The best example we have at the moment are blood transfusions, which seem to be well-received as a reason to socially interact.
If you do go forward this something like this, would it develop a greater sense of persistence per character life? Is that something you want to encourage?
Absolutely, but I don’t want to get people’s hopes up about a skill system or skill tree, so we’re more looking at how players develop the abilities to do certain things, and how players develop the ability to get better at doing things.
To provide a tangible example of something we know we want the player to do – we don’t want the player to start the game and choose how their character is going to be. That will actually come out as they play the game. An example would be wanting your character to wear a cowboy hat, camouflage jacket and black pants. You’re going to have to find those items in the world. Your humanity, and these kinds of things, will affect how you look. The way you play the game will affect your player. As we explore what the player can do, that is something we’re interested in, but it’s going to require a lot of experimentation.
We don’t want the player to start the game and choose how their character is going to be
A couple of examples we thought of were very advanced skills, such as advanced repair of aircraft and vehicles. That’s not a skill that everybody has. And various doctor-related skills. They aren’t things that everybody just has; they’re things that maybe people can learn, but they’ll require practice and maybe some social interaction.
What are the plans for introducing additional maps? Are you still exploring a microtransaction model for official maps? Can community maps be integrated?
I haven’t done a lot of thinking – but I think probably all of the above. I would really like to see if DayZ got really successful, then very shortly after release, we would actually start clobbering together some resources to produce some official maps.
Maybe DayZ is so successful that we can convince all those associated at Bohemia Interactive that we could actually provide those for free, or as part of the product. Or, maybe, we would do it as a paid kind of micro-DLC. Or, maybe, we would do a community-based pack. That’s worked very well for games like X3. I loved how Egosoft pulled together some really good community content.
The time to review that will probably be in about three months, once we’re out there, once we’ve got a good grounding on how much work there is. I suspect that community maps are going to be heavily involved then, just because of how popular they’ve been. But we need to solve some issues associated with modding to get there.
Do the changes in architecture make it more difficult for the community to create mods and maps?
We’re definitely not I’m going to say, solving, but taking the right steps in terms of dealing with hacking-related issues, unfortunately kills modding outright for a start. It’s something we really want to work back towards. And we can – because the architecture doesn’t inherently disable modding completely, but it’s going to initially make it virtually impossible to mod or script the game, which reduced certain elements of hacking – not everything, but some of it. It also places a lot of responsibility for how the game runs on the server, which reduces a significant portion of hacking that exists at the moment.
Taking the right steps in terms of dealing with hacking-related issues, unfortunately kills modding outright for a start
It’s a double-edged sword. By reducing hacking, we’re reducing that user-created content.
Does user data still report to a central server?
Yes. We’ve implemented that and are testing it. We’ve completely changed the way that architecture works. Initially, we’re releasing it with a central hive. Maybe we might have several separate hives, but if they were separate they would be with large, very trusted networks, because we want to reduce the opportunity for hacking.
But we will definitely allow people to run their own servers. Perhaps, at least initially, these servers will be much more centralised. I think as we go on, naturally we’ll end up with a situation that we have now, even if we didn’t provide the architecture, I’m quite sure people will reverse-engineer it. It’s a natural progression for the product, and we just need to do it sooner rather than later.
It’s not like we’re going to be charging hosting companies for this stuff. It’s a symbiotic relationship. We don’t get money from renting out the servers, but they take away a massive risk for us because they’re competing with each other to provide the ability to rent servers. And they’re holding all the risk of player numbers. Once we make the game, it can survive forever so long as there’s a market for that.
Any plans for a boxed retail release?
There are no plans at the moment. I honestly believe that this is the end of the beginning. The standalone is going to come out – and this is an expectation I really need to manage going forward – that when it comes out, it’s done. But that’s really just the start of development.
This is the end of the beginning
We just want to have the architecture working with some base level assets – and then we want to iterate on that, much like Minecraft did. I think it’s going to be 12 months before we have something we can sit back on and say, we’re really happy with this. From there, maybe we do the occasional content update. But I’m hoping around the end of next year, we’ll be able to sit back and say we’re really happy with it – DayZ is where we need it to be. And that would be the time, if there was going to be a box release, that’s when we’d see it.
The problem with box releases is they cost so much to make. It’s not necessarily the publisher’s fault that the developer gets so little money from it. It costs a lot to make CDs, and boxes, and you’ve got to hold stock.
So you see the first release of the standalone not as the game as it will finally be, but the first alpha milestone release in that Minecraft style release model?
It’s hard to define it. I chose the name ‘Foundation’ at one point – but we’ve abandoned that as it’s been borrowed elsewhere. I don’t think it’s really true to call it an alpha, because DayZ builds off ARMA II. In many ways, it’s not an alpha, because it’s a post-release product that has evolved into something else. But it’s behaving as an alpha. So I don’t really know what to call it.
I guess the safest is to call it an alpha, because people understand that. So that, I guess, is probably what we’ll do. Because people then know that they’re buying into something that’s very changeable.
You’re climbing Everest later this year. What possessed you to do this?
The uncle of a friend of mine, who I’m still in contact with, came to my primary school and talked about Everest. I’ve always been interested in mountaineering ever since I was a little kid. I did a lot when I was at university, but fell off the table more recently. So, obviously I financially did well out of DayZ, and the only thing I could think of really wanting to do was Everest.
I’m not done with the survival genre
I really want to explore some interesting game ideas, and it feels to me like I’m not done with the survival genre, if I can call it that. I want to look at other survival options other than running around shooting people. So I guess it’s a mixture of me wanting to explore it as a hobby, and me wanting to explore it to see if there is a good game design there. It’s a real challenge I’ve always wanted to do – and do it while I’m still relatively young.