Developers of Caliber, a third-person shooter, discussed the development of the game, technical details, and challenges they met along the way.
80.lv: Could you tell us about the studio?
Albert Zhiltsov, Creative Producer: The studio as a team began to exist in 2006. Like most, we started making mods for other games and soon grew to have our own development. The first big project was Rise Of Flight, the most realistic flight simulator of the First World War. Now in the studio, we have more than 100 people and 2 main projects in development — IL-2 Sturmovik and Caliber.
80.lv: What inspired you to create Caliber? What references have you been using?
Albert Zhiltsov: We came from flight simulators where teamwork is the basis if we are talking about multiplayer. Therefore, we have always had the desire to make team games. And the way of working with the history of aviation from 1915 to 1945, of course, determines that you love and understand military style. It is only necessary to replace the genre and we get Caliber, a team, and tactical shooter. The fact that it would be a shooter was rather a desire for us to work in as different as possible directions. Literally different worlds — sky and ground.
Alexey Komarov, Art Director: It is difficult to talk about the idea. First of all, I wanted to make a game. Before that, we were making simulators with complex systems, without any conventions. It is difficult to start playing such projects right away, you need to study a lot of documents before starting. That is why we had a desire to create something available to everyone, something that you could start playing without studying a lot of educational information, something that you can easily download, install and immediately play. At that moment we were all very interested in games related to shooting as well, we always played something together. And somehow it also happened that WG had thoughts to do something like that, and so we started.
We did not have any unambiguous vision at once, it all appeared in the process, at that moment there was only a direction. We agreed to go firstly from how users play, to check everything experimentally. A framework began to form. Gradually, new people were joining the team, new ideas were appearing, all this was intertwining with each other, we started to get new directions for testing. Somehow we ended up with the project we have now.
But I would say this is still a long road, we have not planned to stop yet, and there is still a lot we would like to implement. So we are still underway.
80.lv: How did the development process start? What is the size of Caliber's world map?
Alexey Komarov: Initially, the project was in the first person on large scale. Gradually we began to understand that this format was not very suitable for the goals that we had set. It did not sound like a game that was easy to start playing. First-person made it difficult to navigate in space, large proportions greatly increased the requirements for hardware and time necessary to finish it. As a result, we changed the concept a lot and continued to move towards the original goals. Apart from the scale, the maps were more complex logically. As it has turned out, unnecessary complexity is not useful, excessive variability confuses, and the lack of variability is annoying.
In the end, we came to some kind of balance and those sizes of maps that we have now.
Albert Zhiltsov: Like any normal project development it began from scratch — with loads of ideas, attempts, and mistakes. We knew what the result should be, but we did not know how to get to it. We understood that we needed to do technically complex things, MMO, plus game mechanics, plus service-content system, and so on. It is good that we were aware of this task, and therefore completely wrong ideas and solutions were quickly rejected. We were making prototypes, lots of prototypes, sometimes very different "1st and 3rd person" ones, for example, changing the entire perspective for the player. Sometimes the prototype could only have a new UI or mission logic, but otherwise, be a copy of its predecessor.
80.lv: How were the game-levels created? Are they different for each class? Was it hard to follow your initial idea?
Alexey Komarov: The creation of game levels, I think, happens like with most developers. First, the level is assembled on blocks, cubes. Roughly speaking, a sketch is made in space from primitives. The main points of appearance of players, opponents are determined, other positions are set up. After the sketch, a more detailed development already begins, filling with assets. At the very beginning, the cubes were as simple as possible, without any standards. In the course of work, various requirements, rules, types of objects were formed. As a result, now there is already something like a basic simple constructor of different sizes and types of objects from which the object is assembled. There are typical objects. With such objects, both development and getting used to new game locations, understanding what is happening are easier.
If to talk about the difficulties, I can note the specific work with space with a third-person camera. For example, since the camera is located just above the character, you cannot make ceilings and doorways below a certain level. Otherwise, the camera begins to behave incorrectly, awkwardly, and not ergonomically. Roughly speaking, in real life, most of the things are created to make the average person comfortable. If a person is very much out of these average sizes, then many things may be uncomfortable for us. Everything stops being designed for us.
Now, if all these spaces in the game are made the same as in reality, then a third-person camera will feel approximately the same as a 2-meter person somewhere in Japan, where he will most likely not fit even into an ordinary bath.
Therefore, I had to adapt the world a little to the way the camera is positioned to make it convenient.
The levels do not differ between the roles, all roles exist at the level at the same time. However, at each level, we try to provide different zones for each of the roles. There are places for snipers to shoot into the distance, and for medics to shoot close range from shotguns at the enemy.
80.lv: Please tell us about the game's mechanics, modes, and features.
Alexey Komarov: There are many modes, and in general we want to do more of them, but there are three main directions:
- PvE – modes against bots
- PvP – modes against players
- PvPvE – modes where players fight both bots and other players
The distinctive feature is probably the last mode. It turned out to be something like a MOBA, only in the genre of third-person shooters. We want to develop this very direction above all for the simple reason that all characters can realize themselves in this mode. Some characters are better at destroying bots, some are better against players. In PvP-only or PvE-only modes, let us call them the "sterile" modes, only certain characters feel good. And in PvPvE you can play as anyone, all the mechanics begin to work, everything lives like a single organism. Unfortunately, since this mode is the most complex, it is still the most difficult to work on. We can say that at first we made different parts of the constructor and now we have just started to combine these parts into one whole.
Albert Zhiltsov: After working on complex documentary games about technology, I wanted to make it accessible first of all. That is it, minimum requirements for the player in the main gameplay. The players themselves must decide whether they want to improve or not their shooting, knowledge of maps and modes, and, most importantly, finding friends for the game. We are pushing for this. We have chosen player interaction as our main challenge. We have decided for ourselves that we will offer players a mandatory co-op game, not a single-player multiplayer that we usually see in shooters, but a team one. Yes, the weapons have a recoil with a characteristic pattern, and the duel will always be won by the one who handles his character and his weapons better. But flanking is flanking. From the flank, all masters are equally vulnerable. This applies to both PvP and PvE. In PvE, players also need to interact. AI does not stand still and after a while starts attacking the players. The number of resources is limited for the entire mission. They must understand that all of them must fulfill their roles, otherwise, lose. There were moments when we showed missions with a victory rate of 16 out of 100, an average for the entire audience. We experimented; I am sure we will continue to experiment, for the sake of the interest of the audience, of course, and we will develop PvPvE, the most interesting and the most competitive mode for me.
80.lv: Could you tell us about each character, its history, and what they do in the game?
Alexey Komarov: Oh well. It will take too long to talk about each character. Although there are only four roles:
- Assault — breaks through the defenses, goes to the rear
- Support — takes the enemy's fire on himself, distracts the enemy
- Medic — responsible for the team health
- Marksman — team balance, works on especially dangerous targets from afar
But within each role, there are so many completely different characters. Some characters clearly reflect the role, some are slightly biased towards others.
For example, there is the character Sokol. It is a classic marksman, with a bolt rifle, his single shots hit targets at a very long distance, but are weak at close range. This is the most pronounced character as a marksman. And then there is Strelok armed with an automatic rifle. His goal is the same, to be the balance of the team, to destroy especially dangerous targets, but he is played more like assault, a character with a machine gun.
Some operators feel good close to bots, while others, on the contrary, work on long-range targets against players. In general, in all this diversity, you can definitely find someone to your liking, with the gameplay that is closer to you.
Each operator is a mini-game. You can find someone who you will be useful as without shooting at all, and it will still be fun and interesting.
Albert Zhiltsov: The most important thing about the roles is that they exist. The role determines your basic gameplay. Shooting only = marksman, shooting and moving = assault, shooting and monitoring the battlefield and allies = support, monitoring the battlefield and allies = medic. They use different parts of the interface, and what is important to the medic is sometimes not important to the assault, he does not even control his health, let alone his allies’. We do not have a role migration, i.e. medic is not self-healing assault. This changes the gameplay a lot, even in a shooter.
The Art of Gamedev
80.lv: Let's discuss the game from the artistic point of view. How did you work on the game-world? What tools were you using?
Alexey Komarov: First of all, work on the game world continues. We are a game service still, and we constantly add new locations. If we talk about the approach, then I do not think that it is somehow different from other studios.
In the beginning, there is always a search for references, topics, the study of some nuances. Perhaps some peculiarity is that we have agreed that the game world is a collective image. Still, the entourage must be believable. Therefore, we always try to preserve the identity of any area, territory, but not copy everything as the real world and real events are one thing, and a game is a game.
As for instruments, I believe our main tool is Unity. For convenience, some small things we purchase in the asset store, others we make ourselves. There are many different things, I do not want to make a list.
As for the process, at the initial stages, the stages of prototyping, we used whatever came to hand for efficiency. Some basic things we made ourselves, some things we bought or ordered. The main goal was to form an image, to play it. So it did not matter what we used or where it came from.
When we had already moved from prototyping to the formation of something that we were going to show to the players, then we decided to do almost everything ourselves. In our case, it was more convenient as sometimes you can spend more time on adapting someone else's purchased model for the necessary game goals, rather than taking and doing everything yourself from scratch, especially since then the requirements for the models were just being formed. We have worked and are still working with shaders. Some new elements are constantly appearing that require some special solutions, for example, we added long hair. We will gradually continue to work on adding some similar visual cool things.
80.lv: How were the characters created?
Alexey Komarov: As the Art Lead, the main thing I look for is the base image. Each character should be easily recognizable and have a distinctive color key. Considering that each character is based on real photographs of the units, the main difficulty is finding references, unique and colorful features of each unit. There is also some division in detail, attention that we pay to certain parts of the character. I think Alexey Konzelko will tell you better about some more complex details and nuances.
Alexey Konzelko, Lead Character Artist: The hardest part, in my opinion, was to assemble a team of artists willing to work in the office and capable of creating characters that would be competitive on the level of modern games. Starting work on the first sets of characters, it was necessary to take into account what the requirements would be on the market at the time of the project's release. We managed to attract talented artists with extensive experience in the film and game industry, for whom there was no difficulty in working with anatomy, sculpting faces, and elaborating equipment at any level of detail. The limitations were only related to polygon budget and resolution and number of textures. In general, now we can create a character of any level of complexity and detail.
80.lv: How was the game optimized for PC? Were there any technical difficulties with the multiplayer?
Konstantin Afanasyev, Technical Director: All the complexities of creating an online game are by no means related to technology. Technologies are known to everyone, we did not make the first online game in the world, and not even the first in the CIS.
The main task that needs to be solved is the definition of the necessary-sufficient architecture, the choice of optimal technologies/services, and the forecasting of risks and ways to solve them in the shortest possible time in the event of an avalanche-like increase in the number of players.
After that, there are not so many options left on the shortlist, and the main problem, in this case, is not to slip into perfectionism and to comply with the set deadlines. In this case, it remains only to implement the planned plan, which we successfully managed to do.
Albert Zhiltsov: I, as the Producer, basically demanded the problem to disappear and scolded it.
80.lv: Could you tell us about your marketing strategy?
Albert Zhiltsov: Nothing successful enough to be boasting of has happened yet. We created the right technologies, learned how to make regular updates, and work with the player community. Let us see if these efforts bear fruit.
80.lv: What challenges have you encountered during the development process? What advice would you give to other developers?
Alexey Komarov: I think that the most important challenge is the lack of the ability to do everything and check everything right away. You need patience and a sound assessment of the available resources and opportunities. So, to stay within the framework of this understanding, keep calm under the waves of criticism and not fall into the abyss of attempts to make everything perfect at once — these are the main difficulties.
The most important things before setting sail are:
- Find a reliable team.
- Find enough food or have an understanding of how to get it and where.
- Mark the course.
The most important thing in sailing is to simply stay on course.
Albert Zhiltsov: We have faced all the challenges and problems, and all developers will face them. Therefore, the only sensible advice is: collide and overcome, and try to do it with pleasure.
Konstantin Afanasyev: Creating a good online game is not only an interesting plot, beautiful graphics, and not even fast servers, it is a whole infrastructure that players do not know about, and young teams often do not even think about it.
Our main challenge was how in a very short time (and we had about 3.5 months from the development team, who are just creating the product) to switch to the full-cycle team, those who are not only the developer but also the publisher of their game. To do this, we had to very quickly solve a lot of problems, and find answers to questions that are not at all obvious to the players. For example, where the servers would be located, what team was needed to maintain them, where the users would play geographically, how the version of the game would get to them on their computers, how to migrate user data, through which payment system to build the online store, in what languages would the support team communicate with the players and how many employees needed for this, how to build a flow for testing and accepting all services, and many, many other problems and issues, without the solution of which, even the highest quality product would forever remain unknown to a wide range of players.
Thanks to our excellent team of professionals, we managed to cope with all these tasks.
Do not be afraid to take risks, but at the same time clearly understand the limitations of your capabilities as a team/availability of resources.
If you manage to find a balance between these extremes, everything will work out for you and there will be no impossible tasks.