Arma 3: Interview with Ivan Buchta, Creative Director of Bohemia Interactive

[Butulino 10]

This interview on Everyeye.it is the English version of a conversation with Ivan Buchta about Arma 3, the next installment in Bohemia Interactive Studios' upcoming military shooter sequel. Topics include the structure within the game's open world, AI improvements, their disinterest in adding a cover system, underwater combat, vehicles, weapon changes, DirectX 10/11 differences, multiplayer modes, and more.

[ziiip 1]

Cool, reading.

[dmarkwick 261]

I note this little snippet:

We only consider allowing the player to experience the helicopter flight model from Take On Helicopters based on the game difficulty.

Which reads as though realistic helicopter flight modeling might be linked to difficulty settings.

[Mr. Charles 22]

Better move this into the press coverage thread.

It was a nice and informative read, thx Ivan!

[Windexglow 10]

The social aspects of multiplayer gaming are certainly interesting and an increasingly popular part of network gaming. We would like to make clan support more intuitive for casual players. Ladders and achievements are among the features considered, but there are still a lot of open questions.

Dedicated servers including a Linux server will be present, of course. The in-game interface for the dedicated server may receive some improvements, and we constantly improve the engine's multiplayer protocol based on user feedback from Arma 2 and Operation Arrowhead.

People don't play arma mp to be competitive and look at points. What other uses for a ladder are there, aside from comparing yourself to others? Achievements also seem a bit odd, out of place.

The rest, great information.

[dmarkwick 261]

People don't play arma mp to be competitive and look at points. What other uses for a ladder are there, aside from comparing yourself to others? Achievements also seem a bit odd, out of place.

The rest, great information.

It will be like everything else: an optional feature.

[dale0404 5]

It will be like everything else: an optional feature.

You certain?

[Maio 293]

You certain?

Maybe it will be only for TvT , PvP , etc.

I don't see it affecting co-op in any way ;) .

[nikiller 18]

hi,

nice interview thanks for the link :)

cya.

Nikiller.

[metalcraze 290]

I like what I'm reading.

Especially the bit where Ivan Buchta is dismissing boring cover systems as unrealistic and not viable for PC games where it's about freedom of movement and goes like "hey let's talk about improving AI instead".

The bit about focus on coop is also what I like. Guess that's why the article is so relatively full of AI discussion. This means SP will be improved too.

And I guess by 'improving AI movement' he means that they are trying to fix the stance bunny-hopping present since OFP? (e.g. AI lies down on the ground, stands up, lies down again - repeat)

Now if only BIS managed to actually improve AI like they promise and make them properly use interiors of buildings - then it will be a real step forward for the series as a battlefield simulation.

[Big Dawg KS 6]

After reading this I have mixed feelings. It seems BIS is still trying to be a little ambitious (which was a big problem in ArmA 2). I was hoping they would focus more on polish this time around, but so far there's no info on whether or not this is the case.

[wolfbite 8]

Damn blocked at work :(

[the_antipop 10]

After reading this I have mixed feelings. It seems BIS is still trying to be a little ambitious (which was a big problem in ArmA 2). I was hoping they would focus more on polish this time around, but so far there's no info on whether or not this is the case.

But since ArmA 2 BIS have acquired 3 new studios... Maybe they have the man power to make this game what we all want it to be, and what it should be! :)

[dmarkwick 261]

After reading this I have mixed feelings. It seems BIS is still trying to be a little ambitious (which was a big problem in ArmA 2). I was hoping they would focus more on polish this time around, but so far there's no info on whether or not this is the case.

You seem to be naturally conservative in your A3 hopes :) Not a complaint BTW just an observation, I like you in general but I get confused when you don't wish for say editor advancements but do wish for polish, I struggle to think what kind of form polish would take if not improvements to existing features.

I think ambition is what kicked off BIS's success in the first place, and I should hope that they continue to be ambitious. Myself, I didn't notice too much advancement of ArmA2 over ArmA1, in my mind it was polish.

I'm getting the general feeling that A3 will be a very polished affair, and also ambitious. That will no doubt cause some problems but I am confident that A3 will be:

At least playable out of the box.

Quickly patched to be pretty well perfect, which despite what I read on the A2 forum happened with A2. It always worked well for me anyway.

[metalcraze 290]

Eh Ivan Buchta actually talks about polish.

Because fixing/improving AI, using PhysX for vehicles instead of a clunky system we have now and using one customizable weapon instead of spamming a hundred weapons in a box is exactly that

[Tonci87 163]

Nice Read, thanks!

[4 IN 1 0]

Well as long as its not "ambition but rubbish" i'd be OK with that

[purepassion 22]

Damn blocked at work :(

there you go ;)

Everyeye:The revamped singleplayer looks very promising. We understand that you implemented some kind of quest-system in the sandbox environment. Would you give us an example of what this means? How do you obtain them?

Ivan Buchta:We decided to create a more open structure, although we use the means already present in the game since Arma 2: tasks, conversations system and FSM scripting techniques. With the flexibility our engine and tools provide, we are able to employ them in a new way and on a larger scale.

In the game, the player can obtain either "quests" directly related to the story, or some "side-quests" which would help him to get something special: equipment, information, guerrilla support, vehicles and so on. For example, your main objective may be to destroy an enemy base, because a friendly force is threatened by the enemy presence. By following some hints, you may eventually get a local guerrilla chieftain to distract the OPFOR in order to weaken its defenses or receive some close air support, all of which can make your effort much easier. Maybe you would even stumble upon a piece of Intel which would make the friendly command reconsider their intent. To achieve such changes, you would probably have to do a favor for the guerrilla commander or to prove the need for CAS by delivering some extra Intel - simply to accomplish a "side-quest".

I would also like to note that we are not abandoning the scenarios (missions), which are much simpler and shorter providing the element of instant fun. I am sure this kind of gameplay will be attractive for many players including the seasoned Arma veterans.

Everyeye:Two problems of the full-scale battles in ArmA 2 were that the enemy AI was really good at spotting your unit and that your soldiers were too good at shooting them and some times you won an entire mission without shooting once. Are you addressing these in the new AI system? On what exactly are you focusing, beside those?

Ivan Buchta:The case you mention illustrates how the mission design works in our games. Arma has never been player-centric, and the things may always proceed towards a certain set of conditions which indicate the mission end: we do not ask the player to fulfill an objective, we rather ask whether an objective is accomplished. Imagine yourself as a Special Forces guy behind enemy lines tasked to blow up an enemy ammo dump. You may either do it yourself, or you may ask local resistance to do this task for you. In both cases, the ammo dump would eventually get blown to pieces, and you win. However, lots of testing and balancing will be done in order to avoid putting the player in the role of a mere witness. In the campaign, there are a numbers of unavoidable tasks or decisions.

Regarding the AI improvements in general, we would mainly like to achieve more natural movement of the AI soldiers. The Micro-AI system already makes the AI entities formidable opponents, but there is a lot to improve in terms of the visuals. Also, we put a lot of effort into "teaching" AI to use the new features, e.g. underwater movement, first aid routines or customizable loadouts.

Everyeye:Will you ever introduce some kind of cover-system or blind-fire?

Ivan Buchta:Most probably not. The blind-fire (probably meant as shooting without aiming from behind a cover to suppress the enemy) is probably nothing a real soldier would do on a regular basis. Also, these features are quite common and important in console shooters, where they compensate for the less precise controls. with Arma 3 being developed exclusively for PC, we don't feel any need to implement such features at the moment.

Instead, it would be much more interesting to make the AI use more suppressive fire and smoke grenades, as well as to make the suppressive fire more accessible to human commanders.

Everyeye:Another big new feature will be the underwater combat. Will you introduce larger ships to board and sabotage, submarines perhaps? Will sea currents be implemented as well?

Ivan Buchta:So far we plan only the smaller vessels, but the sabotage missions involving placing charges is something we would really like to introduce.

We don't plan to implement sea currents, as it would only complicate matters for the AI. However, such feature can be probably scripted in case we'd need it for a particular situation in the campaign.

Everyeye:Speaking of vehicles, which level of realism do you want to achieve? We saw Take-On Helicopters for example and it sure takes a lot of practice to master such a complex model. We wonder how driving a tank would be...

Ivan Buchta:Tank controls shall be the same as in the previous titles of the Arma series. We already offer a vast number of features, therefore we don't want to complicate the vehicles controls any further. We only consider allowing the player to experience the helicopter flight model from Take On Helicopters based on the game difficulty.

However, the tank driving will be a lot better experience thanks to the improved physical simulation of driving and collisions.

Everyeye:What changes to the weapons behaviour are you introducing in ArmA 3 (i.e. the new wind system physics etc.)?

Ivan Buchta:We mainly focus on adding the customization options: with various accessories and optics, a single weapon can turn into anything from a night-ops special forces rifle to a marksman rifle with anything in between.

Addition of more complex windage simulation is not planned. We already have bullet drop and weapon zeroing capability both working nicely even for the AI; anything more complex would probably only complicate the matters (weapon control, performance) without adding much to the gameplay.

Everyeye:Why did you choose DirectX 10 for the new game engine, while the new DirectX 11 libraries are out?

Ivan Buchta:The game will be released only for DirectX 11, but the engine is still capable of running in the DirectX 10 environment. The new possibilities offered by DX11 still wait to be fully explored by our programmers.

Everyeye:We would like to know, if possible, which of these technical features you are planning to introduce out-of-the-box: SLI / Crossfire, Nvidia Vision / AMD 3DHD, Nvidia Surround / AMD Eyefinity?

Ivan Buchta:The engine has been capable of running on a multi-GPU setup since Arma 2; however, things work a bit differently: the hardware manufacturers maintain compatibility with the games, not vice versa. Regarding the new technologies, NVIDIA Vision and Surround as well as AMD 3DHD and EyeFinity are partially supported since Arrowhead, and we hope to be able to further enhance and expand the support.

Everyeye:About the Multiplayer, are you planning any new modes?

Ivan Buchta:We would like to focus mainly on cooperative gameplay, but it is too early to reveal any details. I can safely say we will make sure to use the unique possibilities Arma games have over the competitors: huge environments, simulation of large numbers of AI entities, various vehicle types including aircraft, capable AI or scenario parametrization.

Everyeye:With the increasingly number of players after the success of ArmA 2 will there be any kind of in-game clan support or ladders? Finally, what about dedicated servers?

Ivan Buchta:The social aspects of multiplayer gaming are certainly interesting and an increasingly popular part of network gaming. We would like to make clan support more intuitive for casual players. Ladders and achievements are among the features considered, but there are still a lot of open questions.

Dedicated servers including a Linux server will be present, of course. The in-game interface for the dedicated server may receive some improvements, and we constantly improve the engine's multiplayer protocol based on user feedback from Arma 2 and Operation Arrowhead.

Everyeye:We are seeing great things forthcoming for the FPS genre and ArmA 3 is one of those titles that will most likely make the difference. You already have a solid fan-base and that's certainly a great opportunity for you to make it even bigger within the ArmA franchise. What are the ArmA 3 key elements and innovations that, in your opinion, would appeal to a larger gaming public?

Ivan Buchta:Most shooting games keep focusing on the action-packed cinematic experience, emphasizing the little details at the cost of the player's freedom and gameplay possibilities. We would like to deliver a completely different experience, offering player true and unparalleled freedom of choice in an environment which reacts to his actions. In other words, Arma 3 will hopefully be more demanding on one's brains than fingers, rewarding the creative players with a proper response to their tactical efforts.

Everyeye:In the shooter history Operation Flashpoint marked a milestone no doubt. How does their creators look at the new releases of their first warfare brand?

Ivan Buchta:We will have to take a look. :)

[maturin 12]

Two problems of the full-scale battles in ArmA 2 were that the enemy AI was really good at spotting your unit and that your soldiers were too good at shooting them and some times you won an entire mission without shooting once.

Learn to play.

Instead, it would be much more interesting to make the AI use more suppressive fire and smoke grenades, as well as to make the suppressive fire more accessible to human commanders.

Score.

anything more complex would probably only complicate the matters (weapon control, performance) without adding much to the gameplay.

Fail fail fail. Sniping without windage is a joke once you have experienced it. Give us a well-optimized system for scoped weapons, and let the modders tweak the configs to extend it to all weapons if they so choose. And disable it for AI.

[gossamersolid 155]

After reading this I have mixed feelings. It seems BIS is still trying to be a little ambitious (which was a big problem in ArmA 2). I was hoping they would focus more on polish this time around, but so far there's no info on whether or not this is the case.

I kind of agree.

Don't complicate things BIS, ArmA 2/OA were good, just bring some polish and a few new features.

We don't need GAME2 here :)

[Big Dawg KS 6]

You seem to be naturally conservative in your A3 hopes :) Not a complaint BTW just an observation, I like you in general but I get confused when you don't wish for say editor advancements but do wish for polish, I struggle to think what kind of form polish would take if not improvements to existing features.

I don't consider enhancing the editor to be polish. Polish is all about presentation; fixing the minor bugs/oddities that show up here and there & making the game look more completed. Mr. Buctha sort of made it seem like presentation is kind of a low priority.

I think ambition is what kicked off BIS's success in the first place, and I should hope that they continue to be ambitious. Myself, I didn't notice too much advancement of ArmA2 over ArmA1, in my mind it was polish.

The campaign in ArmA 2 was quite ambitious and because of that it was very far from polished upon release. A lot of people here probably don't care about the campaign, but the SP campaign usually drives the development of other features (like conversation system) which were also lacking some polish.

And ambition is not something BIS has to worry about; there's still no competition for their type of games. And I would say it's not only the massive scope of their games that made them sucessful, but their attention to detail as well. Polishing up what they have would do more for me than adding major features.

I'm getting the general feeling that A3 will be a very polished affair, and also ambitious. That will no doubt cause some problems but I am confident that A3 will be:

At least playable out of the box.

Quickly patched to be pretty well perfect, which despite what I read on the A2 forum happened with A2. It always worked well for me anyway.

If you look at the CIT for ArmA 2 and the sheer number of minor bugs that are simple tweaks/easily fixible you'll see that a couple of weeks/months dedicated to polishing would have been very beneficial. Unfortunately there are still a large number of these issues that have never been addressed.

[wolfbite 8]

It sounds very interesting... The campaign sounds a bit like the Enemy engaged style campaigns mixed with oblivion :P Looking forward to see how it works out.... And anyway if worst comes to worst... We still have the Editor to do things our way :D

I wonder if you will be able to stop players from switching sights/scopes with a script? I find that a bit of a random addition... Nice but not really good for zeroing IRL...

P.S Thanks a lot Passion much appreciated!

[dmarkwick 261]

I don't consider enhancing the editor to be polish. Polish is all about presentation; fixing the minor bugs/oddities that show up here and there & making the game look more completed. Mr. Buctha sort of made it seem like presentation is kind of a low priority.

Because he didn't specifically mention it? I wouldn't say that represented seeming like a low priority. It's his job to push new features when interviewed I guess.

The campaign in ArmA 2 was quite ambitious and because of that it was very far from polished upon release. A lot of people here probably don't care about the campaign, but the SP campaign usually drives the development of other features (like conversation system) which were also lacking some polish.

Hmm you're right I never touched the campaign so I cannot make any judgement on how polished it is. Or isn't :)

And ambition is not something BIS has to worry about; there's still no competition for their type of games. And I would say it's not only the massive scope of their games that made them sucessful, but their attention to detail as well. Polishing up what they have would do more for me than adding major features.

While playing the new OFP patch today I remarked on how even now, 10 years on, OFP still represents more ability than just about any game released to date. I also remarked on how clunky the interface is :D we definately have more polish today. So I guess, speaking for myself, I'd like to see more features. I've always managed to get around various bugs and foibles, which get fixed/modded most times in any case.

If you look at the CIT for ArmA 2 and the sheer number of minor bugs that are simple tweaks/easily fixible you'll see that a couple of weeks/months dedicated to polishing would have been very beneficial. Unfortunately there are still a large number of these issues that have never been addressed.

I'd rather an early release than a later, polished release. I know that's not the general view, but it is my view :)

[gossamersolid 155]

I think you are alone on the "early buggy release". I'd much prefer BIS to take 2 months extra to polish the game.

There should be an early point set about 1-2 months before release in which new features/content maybe stops or slows down quite a bit so the crew can be focused on fixing the issues in the game/polishing it.

I don't want another buggy release from BIS, they have no excuse at this point. Their studio is much larger and they've had plenty of feedback from us over the last 10 years.

[dmarkwick 261]

I think you are alone on the "early buggy release". I'd much prefer BIS to take 2 months extra to polish the game.

I'd just like to clarify that the quote "early buggy release" are your own words, not mine :)

There should be an early point set about 1-2 months before release in which new features/content maybe stops or slows down quite a bit so the crew can be focused on fixing the issues in the game/polishing it.

I don't want another buggy release from BIS, they have no excuse at this point. Their studio is much larger and they've had plenty of feedback from us over the last 10 years.

Maybe I'm more tolerant of obstacles, but I always found BIS products to be emminently useable on release. OK so I never play the campaigns, and I guess that's where most bugs reside, but for my own uses I never see anything that stops me enjoying the product.