R&D is new technology, which has been developing by iFun4all: system for the automatic game modeling, based on real world data, through the development and use of an intelligent testing agent and automation of the process of balancing the game. This system will be a result of the R&D project which is currently carry out by iFun4all. The R&D project started in March 2017 and it will finish in September 2019. The R&D project is co-financed by The National Centre for Research and Development (OPERATIONAL PROGRAMME SMART GROWTH 2014-2020).

R&D works will result in developing a system for the automatic game modeling, based on real world data, through the development and use of an intelligent testing agent and automation of the process of balancing the game. The System will be developed for the purpose of  producing one’s own video games building their functionality (scenario and game model) on the information taken from the real world and then adapted to the game environment – real world data. Games created with the designed solution will be directed to the players constantly seeking and expecting unique features in the game that will allow them to achieve the greatest immersion effect, or „submergence” in the game. The system will be the technological innovation on a global scale. There is no solution in the world, which would have the features and functionality convergent with the proposed system.

As you can see in the illustration, the system is based on a platform that will retrieve information from external sources of information, and then deliver it to the game environment. It will result in the modification of the scenario by the Agent. The Agent will make a decision based on a set of startup data implemented by designers and the data collected from users / testers.

iFun4all has experience in creating games using RWD in Serial Cleaner, the last produced game, only the following information from the real world will be used:

– Geolocation,

– Weather,

– Data from the social accounts (Steam, PSN, X-Box),

– Time and date.

However, Serial Cleaner has a few drawbacks resulting from the inability to use the designed technology or other, converging solutions. For instance, because of the need to connect with each resource separately, the data will not be updated in real time and will be collected at specific locations / time intervals. In addition, in the absence of the availability of sources of information, the only solution is to set the default configuration. The above-mentioned need to connect to each source individually is associated with the fact that a lot of information cannot be retrieved at once.

The lack of database and information platform does not allow to download the data in a different way than directly onto the device. At this point the problem of limited space and lack of access to historical data appears. A similar problem arises in the case of the player’s personal data, which will be taken only from their social account. Due to the lack of tools to help balance the game and testing, in Serial Cleaner only simple relationships between the data that can be tested manually will come into being.



The most important element of the designed technology will be creating an independent solution, based on artificial intelligence using a personally written set of features, that will automate the process of testing and designing scenarios of the gameplay in games that use RWD.

The agent will give the answer to how, in every single situation, to show or change the specific RWD or a combination of several factors, set different parameters of the game, including the settings and properties of individual characters, the structure of a level or to determine the level of difficulty.

For each event in the real world (rain, day / night, traffic jams, smog, exchange rates, events in the city as demonstrations or disasters, etc.) it will be necessary to prepare changes in the gameplay, which will affect the variable (lock / unlock some routes available for the player to reduce / increase visibility, improve / worsen the eyes and ears of opponents, change the number of opponents on the board, etc.). Any such change of the gameplay must be tested to ensure an appropriate balance of the whole game and eliminate obvious errors or such changes in the game, which would prevent continuing the gameplay, for example blinding or cutting off the opponents from the player, or in the range of the player, e.g. too dense crowd and too many vehicles that prevent access to the location.

The combination of parameters based on data obtained in the real world can be the result of several events which occur in the real world at the same time. For example, rain may occur at night in the time of increased smog. It will be necessary  to prepare the agent to respond to such a combination of events so it can modify the game environment, allowing the game to continue.

Currently, various designed versions of the gameplay are manually tested by people involved in the production process of the game. There are no solutions and tools on the market to automate this process. Considering the Company plans to increase the range of RWD used to design the gameplay, testing each single variant, and then modifying them would require a greatly increased workload of testers, programmers, level and gameplay designers, and perhaps, also graphic designers. It became necessary to develop solutions that would automate the testing process in such a way to find the best combination of parameters of the gameplay, maintaining the appropriate balance at the same time. As a result of the use of the automated testing agent for each variant of the game using the RWD, a specific set of parameters of the game will be obtained. The data will be saved and implemented into the game in an automatic way, without the use of human factors.

One of the major advantages is the fact that thanks to the use of artificial intelligence in the testing process, it becomes possible to test an unlimited number of variants in contrast to the previously used solutions in manual track testing, which was done by the staff of developers. In practice, that was limited to verifying only a pool of paths, as the whole process was very laborious. It is planned to use the following algorithms of ARTIFICIAL INTELLIGENCE: – A * algorithm, which is used to search for optimal paths, – Neural network algorithm based on which the agent will learn to play from the tester.

Another significant advantage of the solution is the UNIQUE NATURE AND NONLINEARITY created by the track agent. Because of its design and the ability to create an infinite number of variants, the agent, drawing  upon the collected real-world data, will each time be able to provide innovative solutions for the game, while in the case of creating test paths by employees of the development company, they are built in a stereotyped way, and therefore their number is limited.

When describing the technological advantages of this solution, one should not forget about the economic benefits arising from its use. Thanks to the INTELLIGENT TESTING AGENT, computer game manufacturers will be able to make savings associated with hiring additional personnel for manual track testing.

An additional economic benefit will be REDUCING THE VIDEO GAMES PRODUCTION TIME. Currently, the process of creating a complex computer game can take up to three years. The use of the intelligent testing agent will shorten the time needed to check all possible paths of the gameplay for at least 8 months compared to the situation if these tracks were checked manually by testers and game designers. Currently, based on the experience of the company, a person is able to test 8 tracks during the day. The designed solution will be able to test 40 paths a day.

Downloading RWD from various places in the world and at different times of the day, the player will be able to shape the environment in the game in an indefinite way, each time creating an unconventional and unique gameplay, under the influence of various factors. Relying on real phenomena, characters and events, will increase the realism of the game, by adapting the game environment to the environment in which the player lives every day.

To sum up, the first functionality, proving the innovation of the designed technology, will be to develop and use in the process of producing your own video games an intelligent testing agent, which will have the ability to test and create an infinite number of paths faster than a man, which will save time and cost of video games, and from the perspective of players, provide them with a unique design, able to surprise the user at every step.


Another functionality of the designed solution will be the module of automatic balancing the gameplay.

Balancing the gameplay in video games should be understood as a concept and a set of practices to adapt the gameplay to the player, and at the same time preventing the improper functioning of all components of the game. The consequence of improper balancing the gameplay can be cancellation of the entire set of rules of the game, which can lead to preventing its completion.

In previously released games which used RWDs, the extent of their use was limited only to the time and the weather, which simply meant that only a few variants of the game (2-3) were available. Such a small number of variants to be tested were always checked out manually, without the need to automate the process. This means that the foregoing balancing of the gameplay was also done by hand through the game developers who set the necessary parameters to maintain the balance of the gameplay independently.

Video games designed by iFun4all, will focus on the use of all possible data taken from the real world, not only limited to weather conditions, but diversifying the range into communication factors (traffic jams during rush hours, accidents), economic (exchange rates, quotes on stock exchange) or environmental (smog, air pollution). It clearly multiplies the set of variants of the gameplay, that require testing, to hundreds of variants, instead of two or three variants of the gameplay, like before.

For instance, a variant in which the game takes place during the day, with no precipitation and the impact of additional factors, will require a different matching of the level of difficulty and the behavior of opponents than a variant in which the player has to do with the time of night with heavy rainfall and the negative impact of ecological factors.

There is no such a solution available  on the market that allows the automatic balancing of the gameplay in relation to the actual data, thus allowing to reduce the costs associated with the production of a game, mainly due to testing hundreds of possible variants, which is time consuming.

The project will produce a module that will be an automated system of collecting test results and prioritizing the values of individual variants of the gameplay and modifying them automatically, without the need of human involvement.

The main advantage which brings this solution for manufacturers of games, is the abovementioned cost saving in terms of time and cost to the manufacturing process of games, as well as in terms of automation of the updates of the finished product, as opposed to the solutions used in sports games. Another advantage is the ability to create unlimited variants of the gameplay by means of RWD, which will allow adapting all essential data from the real world, without disposing of a substantial part of the above, due to the inability to manually balance the gameplay by means of them.

The advantage of this solution will also be brought to users of the final product, i.e. a video game created by means of the used solution. Players looking for realistic gameplay will be able to fully enjoy the product which adapts all the data downloaded from the real world to the game environment, maintaining an appropriate balance of the game, which will allow smooth and uninterrupted gameplay.


The last and at the same time crucial stage of RWDs use in the designed solution, from the player’s perspective, is to use them to modify the gameplay in real time.

In games that are currently available on the market, the writers were those who assumed the game environment at a certain stage of the game. If the plot required certain weather conditions and time of day / night, they were assigned to the game stage, without the ability to complete it in another configuration, which also determined the behaviors and attitudes of opponents and other characters in the production.

The innovativeness of the designed solution comes down to the ability to modify the gameplay and the world of the game in real time under the influence of collected and then processed and implemented RWD into the game environment.

This type of modification, allows to create a multi-dimensional game, which, in practice, will refer to the game engine to receipt feedback from the system on current information from the player’s area, and then run the appropriate variant in real time. For example, when in the player’s area it starts to rain, after receiving the information from the system the game will adapt rainy weather to the game environment that changes not only the visuals, but also the model of the behavior of other characters, including adversaries, who under the influence of rainy weather have worse sight, or they seek shelter. Moreover, heavy rains could lead to flooding the board, and thus the total changing the location of the game, which launches a whole new scenario.

It is worth mentioning at this point about the relationship between the intelligent testing agent and the module of automatic balancing of the gameplay. The testing agent, in addition to help in testing the game, also supports the process of setting the parameters of the influence of the real-world data when creating the game. The module of automatic balancing of the gameplay is started with the game. When data from the real world are collected, it examines the configuration that the system has for this data and automatically influences the game world seen by the player. To sum up, the testing agent is used during the process of creating the game, the module of balancing of the gameplay works in the product.

This functionality brings numerous benefits for players who will be the final recipient of the products. Firstly, implementing into the game real-time aspects such as: – Weather, atmospheric conditions, – Time (day / night) – Traffic information – Economic information (data from the stock exchanges, the currency market) – Important events of political nature (rallies, protests);

This will allow creating the world  the game in an infinite number of configurations, and thus being able to complete it by one person manifoldly, because under the influence of adapting above factors to the game, the behavior of characters, enemies, and the appearance of the location will change, which in the case of some types of games, will create a completely new model of the gameplay on the basis of the same title.

Secondly, the multidimensionality of the game will allow the player immersing in the gameplay and destroying the „fourth wall”, i.e. closing the gap between the real world and the world of the game. This functionality is particularly important in producing adventure and stealth games. The latter genre, rewards the player for hiding and using deception to circumvent or defeat enemies. Products of this kind allow players to remain undetected by stealth, using disguises or avoiding noise. The use of elements taken from the real world, will not just make the course of the gameplay more real, but will also allow players better adaptation to the role of the character they will play.