LEE- Oyun ve Etkileşim Teknolojileri Lisansüstü Programı
Bu topluluk için Kalıcı Uri
Gözat
Çıkarma tarihi ile LEE- Oyun ve Etkileşim Teknolojileri Lisansüstü Programı'a göz atma
Sayfa başına sonuç
Sıralama Seçenekleri
-
ÖgeArgent: A web based augmented reality framework for dynamic content generation(Institute of Science and Technology, 2020-07) Kurt, Gökhan ; İnce, Gökhan ; 636478 ; Game and Interaction Technologies ProgrammeIn the modern world, people are more and more interested in interactive technologies. Education, research and business habits are effected by this change and humans can be more efficient using interactive technologies. Augmented reality (AR), which is a novel addition to those interactive technologies, is especially effective in this matter. Through augmented reality, people can immerse more deeply with the subject experience and they can have enhanced interaction. Despite the usefulness of augmented reality, it may not always be efficient develop an AR application in terms of development cost. AR development still requires knowledge and experience with certain tools and frameworks. Such tools are usually programming and game development tools and they require programming and technical skills that is gained by long-term education and training. People experienced in design and content creation can be deprived of the ability to create and maintain AR applications. Nowadays, tools like Unity, Vuforia, ARKit and ARCore provide ways to develop AR applications without the need to have knowledge of low-level calculation and programming that is required for AR technology. Normally, developing an AR application would have taken years of research and development by large teams, but thanks to SDK and APIs provided by these tools, AR applications can be developed by small development teams easily and quickly. However, AR is still not easily accessible by all the tech-savvy people that may be interested in developing such applications. Majority of AR applications are developed using Unity. There are visual programming solutions in Unity, but they are not suitable to be used in AR applications. A Unity-based tool that allows people without programming skills to create AR applications, will be utmost useful. Such a tool would require features such as, creating an application without programming, optional support to do programming and scripting, real time updates and ability to ship without any build and packaging step, support for 3D object, image and video, the ability to modify objects and preview them in real time, and the ability to create user interfaces. The tool should also have a user friendly interface and experience. It should introduce the innovative features without changing the conventional workflows. Existing tools do not provide these features which are crucial for an ordinary person to create AR applications.
-
ÖgeImpacts of sensory visual effects on gamers' spatial exploration behavior(Graduate School, 2022) Öz, Bülent Koray ; Köse, Hatice ; 841873 ; Game and Interaction Technologies ProgrammeVisual effects are the basic visual elements that build virtual universes and computer games. How a scene will be perceived by the player, which emotions will be triggered, and the general theme of the game can be determined by visual effects. Visual effects have different meanings in computer games than the film and the photography industries. In the film and the photography industry, the manipulations on the camera lens or the visual adjustments made through a computer on the images form visual effects. On the other hand, in the game industry, the entire game environment is created virtually by a computer. Therefore, visual effects in computer games are visual elements with additional features. Visual effects in computer games are classified under 3 classes. Lighting effects are used to illuminate scenes and create shadows. Environmental effects allow real-world natural events to be simulated in the game environment. Finally, sensory effects define how the game environment will be perceived. These effects originate from the structure of the human visual system or camera lenses in the real world. Sensory effects have a great role in the formation of the perception of reality in games. There are two types of reality in games. One is perceptual reality and the other is photo-realistic reality. Perceptual reality is based on the characteristics of the human eye, and the scene is designed so that it can be seen with the human eye in real life. When photo-realistic reality is adopted, the created virtual environment is designed as if it is being watched by a camera. Effects such as blur, vignetting, dirty lens, and flare are sensory effects. Transferring the emotions such as pain and stress experienced by the character in the games to the player and making the player understand the depth in virtual reality games are some of the examples that are implemented thanks to sensory visual effects. As mentioned above, sensory visual effects are important game elements that determine the mode and convey emotions in games. Although the structures of these visual effects have been studied in detail, the impacts of the effects on the players have not been systematically studied enough. In this thesis, the effects of sensory visual effects on the players are examined. In particular, the unconscious effects of sensory visual effects on the players were investigated. Whether the players are aware of the effects, the recognition of the effects and their effects on decision-making mechanisms are taken as research points. While investigating the effects of sensory visual effects on the conscious and unconscious decisions of the players, the spatial exploration mechanics in the games were used. When people enter a new environment in real life, they first begin to explore and learn about the environment in its entirety. This exploratory behavior occurs instinctively and reflexively. Exploration mechanics are also used in most computer games. In the examination of unconscious decisions made by the players depending on the visual effects, the help of spatial exploration behavior was taken due to its reflexive nature. In the developed game, the relationship between sensory visual effects and exploration behavior was investigated. Two games were developed within the scope of this thesis. Both games are developed using the Unreal Engine game engine. First of all, it was necessary to determine a data collection technique that can be considered as a game, and quantitative data can be collected from the players without interfering with the gameplay.
-
ÖgeProcedural approaches in open-world games: Game artists' perspective(Graduate School, 2022) Özmen, Can ; Alaçam, Sema ; 713245 ; Game and Interaction Technologies Master ProgramProcedural Modeling (PM) or Procedural Content Generation (PCG) is both research and application topic for forty years. That is basically data amplification technique to create immersive media output. Also, the term includes many applications that run with specific inputs for desirable outcomes. Today, with the acceleration of digitalization, the interest in the game and the playing time per individual have increased, and the need for game content has increased accordingly. Since there are relatively large maps and universes in open world games, productive systems must meet this need.For theses in English, the summary in English must have 300 words minimum and span 1-3 pages, whereas the extended summary in Turkish must span 3-5 pages. A summary must briefly mention the subject of the thesis, the method(s) used and the conclusions derived. The main importance of Procedural techniques is labor-efficiency, realism and attractiveness in the product. There are several areas that PCG is using in games as modeling, texturing, narrative, simulation, sound and game design. Also, from the beginning of the game-making, Procedural approaches become more and more important in considering creating larger game spaces and adding layers into player experience. Commercial games use these techniques for quantity and variation of the game content to enrich player experience. Early examples of this approach, which has become widespread in digital game production processes, used randomness for story-based corridor production in role-playing games. Even though Rogue and Elite games were early applications, Procedural Content Production/Procedural Approaches were used in certain areas in the game production process. They could not pass into an advanced, general-use phase. Hardware (deficiencies in game development tools), software (limited algorithms), and a low number of experts can be listed among the main reasons why Procedural Content Production/Procedural Approaches did not spread rapidly when they first emerged. The power of data amplification offers the opportunity to create bigger projects with limited workforce in indie game companies. The main reason for interest in PCG is obviously the generation of a great amount of content. Human labor is neither fast, nor cheap in comparison with PCG. Since the game making has arisen, the quantity of workers has scaled up in particular projects and production times could last many years. Because of this, development costs rise up. The number of profitable games and the ability of the run budget in game making are affected negatively. In addition there is an ongoing argument which claims that there can be optimistic recruitment for game companies about cheaper and more rapid game productions. The argument is based on swapping particular workers who work in the design and art field, into Procedural or Generative Algorithms by keeping desirable game making standards. On the other hand, there is another perception in using PCG to broaden and work with artist creativity and make it possible even the small groups can handle a great number of contents in games. On the other hand, the independent developer model is the way companies produce with relatively little investment and employees. It is seen that the workforce is mainly limited in companies working with the independent developer model. In this context, it is thought that the data amplifier feature of Procedural Approaches creates an opportunity for game companies working with the independent developer model. Compared to Procedural Modeling outputs, the human workforce remains slow and expensive. One of the areas where Procedural Approaches respond in the game production process is optimization. Optimization techniques allow the compression and refinement of the game's partial or total data while supporting the production of games with relatively large file sizes. Thus, the desired geometry is produced at the desired time, and memory optimization is achieved. This optimization in computer hardware allows the effective use of visual data. Procedural approaches are used in many areas of Open World Game production. The literature data on Procedural Approaches, which form the basis of the study, are examined under two main headings: Procedural Content Production in Open-World Games and Procedural Content Production in Game Development, Actor and Processes. Procedural Content Production in Open World Games and Procedural Content Production in Game Development are examined under three sub-titles: Open World Components (Terrain, Plant, Built Environment), Content Types (Models and Textures) and Tools/Algorithms (Artist Friendly Tools, Procedural Game Engines and Plugins for Content Production). Although Procedural Approaches are used in many areas in game production, the scope of the study will focus on open-world environmental elements, modelling and texturing issues. In addition, current tools for Procedural Approaches from the artist's perspective will also be introduced. The main title of Procedural Content Production, Actor and Processes in Game Development is also examined under three sub-titles: Conceptual Design and Procedural Content Generation, Game Artist and Procedural Content Generation Relationship, Handcraft vs Automatization. A two-layer qualitative analysis method was employed in the study. These layers are pilot study and semi-structured expert interviews. Within the scope of the pilot study, open-ended questions obtained from the literature were directed to two professionals, and the data obtained from this constituted the main study scope. Interviews with nine professionals within the scope of the main study are analyzed in the conclusion part. Transcriptions obtained from the transcript of the interview texts, 1. Specification in the use of PA, 2. PCG Tools, 3.Standalone Need /Independent Tool Need, 4.Art Directable – Artist Friendly Tools, 5.Conceptual Design of Games within PCG, 6. Handcraft vs Automatization, 7.Customization in Narration, 8. Online-Offline Generation, 9. Need for Advanced Skills, 10.Opportunities for Indie – Small Sized Studios were examined in the light of lenses.
-
ÖgeReinforcement learning in fighting games(Graduate School, 2022) Uğursoy, Muhammet Sadık ; Sarıel, Sanem ; 529171012 ; Game and Interaction Technologies ProgrammeReinforcement learning is one of the most popular learning methods used on games because of its similar nature to competitive play. Winning the game and the means to win the game can be used as rewards easily, which enables us to create a reasonable benchmark. The field has many algorithms and approaches that can solve simple Atari games and robotic problems, however, it still has many unexplored areas with difficult problems to solve. After the introduction of Deep Q-Learning (DQN) by DeepMind, learning from pixel data become popular and applied to many other games. Agents could reach and exceed human level play in simple games. But for more complex games like Montezuma's Revenge, different approaches such as hierarchical DQN is needed to search the huge search space of the game. Furthermore, the classical +1 reward for win, -1 reward for lose strategy is not always enough for complex games. As the complexity increases, the algorithm and model should change and adapt. Even though Atari games look simple, they are hard problems to solve for an AI agent. The most recent work on Atari games published in 2020, claims to outperform humans on all Gym Atari games. However, there are still many difficult games to solve that requires novel approaches. The work on this thesis focuses on reaching a human-like play at the end stage boss fight in the game called Megaman X. Existing RL algorithms have been tested with different replay buffer types, parameters and exploration strategies and their performances were compared. To make a better comparison of the algorithms, a simple game called Super Mario World and a fighting game with similar characteristics to the main game called Ultimate Mortal Kombat has been tested as well. We proposed new game specific methods to make the agent play better, including reward shaping and feature extraction methods. This thesis shows the all the results of those trainings and analyses the results. In order to get better results from the trainings, reward shaping and feature extraction methods have been suggested and tested. For feature extraction, CNN based methods and auto-encoder frameworks have been tested and in addition to that, direct data read from the RAM such as character and enemy positions. Reward shaping is applied for the main focus on this thesis, the game Megaman X. Variables such as the charge status of the weapon, distance between the enemy and the agent, health and time are used as reward shaping parameters. Both Q-learning and policy gradient methods are tested. In addition, the latest exploration focused methods and hierarchical methods, which are said to be enhancing exploration, are tested. Also, human players who are familiar with platform games are also played the game and their experiences are recorded in a survey. In this thesis, all those methods and results are analysed.
-
ÖgeAnalyzing player engagement in western action role playing games using user reviews and achievements(Graduate School, 2022-03-17) Hacıtahiroğulları, Ziya Volkan ; Ovatman, Tolga ; Tinç, Hüseyin Kutay ; 529181020 ; Game and Interaction TechnologiesUser reviews are one of the leading factors of judging player engagement and popularity of digital video games. They not only provide sentimental feedback but also give insight on what the player likes or dislikes about a product. However, it is confusing to see this segment not reflected in tools of player engagement such as game achievements. Game achievements provide objective data related to the successes of a player in terms of engagement. We believe we can compare these two user outputs to create a correlation between what players say and what players do in relation to player engagement. A sentiment analysis can be done using the frequency of words used in reviews and review recommendation to understand why players use the words they use. In this thesis we will be focusing on western action-based role-playing games, comparing Steam user reviews with Steam achievements by categorizing each based-on type, and consistency. Game achievements will be categorized based on player experience, engagement, and enjoyment aspects. User reviews will be ran through a stemmed word frequency query, manually selected and divided between predefined categories. Data gathered from these categorizations, such as word count and completion rate was then compared with reviewer recommendations to analyze the sentiment regarding each defined player engagement aspect. Furthermore, k-nearest neighbors algorithm (k-NN), a non-parametric classification model, was used to create review recommendation level clusters to analyze and player reviews for other games of this genre. In conclusion, we have found that a correlation between frequency of words used by reviewers and average achievement unlock rates of games can be made. We have also noticed correct applications of different aspects of game engagement may have an impact on the other aspects, as a game with engaging gameplay might lead to lower rates of game completion while a poor implementation of exploration mechanics that lack engagement can lead to increased game completion rates. We have also found the correlation between our two main datasets can be used to create a classification model to analyze review sentiment rates of other games moving forward. Going forward, more research on different video game titles or genres might be necessary for understanding why increased success in one aspect of engagement leads to decreased interest on other fields.
-
ÖgeDrone wars 3D: an interactive simulator for drone swarms(Graduate School, 2023) Karadeniz, Gökhan ; İnce, Gökhan ; 779468 ; Oyun ve Etkileşim Teknolojileri Ana Bilim DalıThe utilization of drone swarms, characterized by their formidable destructive capabilities and broad range of potential applications, has led to the pressing need for effective countermeasures to mitigate their potential threats. Additionally, the relatively low cost of drones has further amplified the need for robust defense strategies. Although consumer-type drones can be neutralized by electronic countermeasures or microwave weapons, military-grade drones are produced to be protected against such attacks, leaving the physical destruction as the best choice. Within the scope of this study, a simulation environment was developed in the Unity3D game engine in order to measure the effectiveness of defense systems against drone swarms and to find effective defense tactics and swarm formations. Unless otherwise stated, swarms with the same default values were used in the tests. The actor types in the simulation were 1) drone, 2) drone swarm, 3) machine gun, 4) laser weapon, 5) anti-aircraft gun, 6) air defense missile launcher. In the tests of defensive drone swarms against offensive drone swarms, two enemy swarms (identical in everything but their formations) were created. Then attack drones destroyed were inspected by applying all possible formation combinations. Results show that, the drone swarms have an average success rate of over 90% in destroying enemy drone swarms. Though, in order to achieve this success rate, the defending swarm must be located on the attacking swarm's target approach path. Another noteworthy finding was that each formation was most effective in defending against the enemy swarm having the same formation of attacking party. A strategy to copycat the offensive swarm's shape when possible seems to be viable given the fact that identical swarms are more effective against each other. This finding was also supported by the results of drone spacing tests. When two swarms of the same number and formation were used against each other with different drone spacing, the most effective defense was obtained when the drone spacing of the two swarms were equal.
-
ÖgeGenerative models for game character generation(Graduate School, 2023-06-13) Emekligil Aydın, Ferda Gül ; Öksüz, İlkay ; 529191006 ; Game and Interaction TechnologiesGenerating visual content and character design for games is generally a time-consuming process and is carried out by designers. The design process can be both costly and time-consuming for small businesses and independent developers. Working in this field requires a detailed understanding of visual aesthetics, creativity, and technical skills. It is important for the characters and visual content used in games to be compatible with the game's story, atmosphere, and gameplay. Designers and artists work to create original visual content and characters that align with the game's objectives and target audience, considering these requirements. Due to these reasons, content creation for games is a challenging process. Automating the design process helps to save time and budget. Many game companies and developers use procedural methods to automate the design process. Procedural content generation involves automatically generating game content using algorithms and rules. This approach offers significant advantages in generating repetitive content and enables developers and designers to create content faster. However, the visual content generated by these algorithms may have certain limitations in terms of diversity. With the advancement of technology and the progress of deep learning methods, approaches incorporating deep learning models have also started to be used instead of procedural methods. Examples of such methods include Generative Adversarial Networks (GANs) and Latent Diffusion models. In addition, in the studies presented in the thesis, the transfer learning method has been used in conjunction with generative models, and its success has been evaluated compared to these methods. In order to perform machine learning, a large amount of labeled data is typically required. However, it is not always possible to have access to a large labeled dataset, and obtaining and labeling data can be costly and time-consuming. The transfer learning method has been proposed to reduce or eliminate this requirement. When applying transfer learning, a pre-trained machine learning model is selected, which has been trained on a significant amount of labeled data. This model is a deep neural network that has learned general features from a large amount of labeled data. For example, a pre-trained classifier model trained on a popular dataset like ImageNet can be used. The initial layers of the selected model contain useful information about learned general features, while the top layers are not applicable to the target task. Therefore, some or all of the layers of the pre-trained model can be frozen, and only specific layers (usually the classification layers) can be retrained on the target dataset. This way, a much more successful model that is tailored to the target task's dataset is obtained. Transfer learning can be used in situations where the dataset is small, just like we have tried. Since a pre-trained model is used, the training process is much faster compared to training from scratch methods. Pre-trained models have learned generalized features from datasets that contain a wide variety and a large number of examples. This gives the models more generalizability, making them applicable to a wider range of domains. In summary, the transfer learning method involves transferring knowledge gained from previous experiences to a new task. It provides benefits in terms of speed, reduced need for labeled data, and improved model performance. Pretrained models trained on diverse and large datasets are used to apply this method effectively. Generative Adversarial Networks (GANs) can generate highly successful results for image generation and are also used in game character generation. GANs are composed of two distinct deep learning models: they are the Generator and the Discriminator. The primary role of the Generator network is to generate synthetic images, while the Discriminator network determines whether the generated images are real or fake. These two artificial neural networks compete with each other during the training phase. The Generator tries to deceive the Discriminator by generating images that are close to reality, while the Discriminator tries to identify the images generated by the Generator accurately. Feedback obtained at each iteration is used for training purposes. Latent Diffusion modeling method is a deep learning approach that involves generating synthetic data, denoising, and noise estimation. This method is based on capturing the temporal evolution of data points. It creates a latent space network in the training data, and through this distribution, it iteratively performs noise estimation and noise removal operations, allowing for the generation of synthetic, high-resolution, and impressive images. The U-Net architecture is used for the denoising model. To achieve this, word embeddings are utilized. Word embeddings are fed as input to all layers of the U-Net. The complexity of the U-Net model increases as the size of the input image increases, necessitating dimensionality reduction. Variational Autoencoders (VAEs) are used to reduce the dimensionality of the input image. By iteratively generating the latent vector, high-resolution images can be obtained. Latent Diffusion models can capture more complex data distributions and achieve more realistic and successful results that align with the real world. However, compared to other generative models, the training process of Latent Diffusion is much more time-consuming and challenging, and it also has higher computational costs. As a result, its implementation can be more demanding and resource-intensive. In this thesis, visual content generation for games is addressed in two different studies. In the first study, six different GAN models were trained using visual image datasets of two different RPG and DND characters. In 3 out of 18 experiments, transfer learning methods were used due to the small size of the datasets. The Frechet Inception Distance (FID) metric was used to compare the models. The results showed that SNGAN was the most successful in both datasets. Additionally, it was concluded that transfer learning methods (WGAN-GP, BigGAN) outperformed the training from scratch approach. In the second study presented in the thesis, a different dataset containing images of 2 different animals and fruits was used. Stylegan, and Latent Diffusion methods were employed. In the training of StyleGAN, eight types of fruit images and three types of animal images were used as conditioning inputs, and conditional learning was applied. In the Latent Diffusion method, the datasets were labeled with descriptive sentences about the images and fed into the model. FID scores were calculated for the generated outputs, and these outputs were transformed into a web game and played by 164 players. The results showed that the Latent Diffusion model performed well in the animal dataset according to the FID score, while StyleGAN performed well in the fruit dataset. In terms of the overall evaluation, the Latent Diffusion method yielded better results. According to the scores obtained from the players, the Latent Regression method also achieved better overall rankings. This indicates the consistency between the results obtained from the FID score and the player evaluation. Both studies demonstrate the feasibility of generating game characters or synthetic artistic visuals using deep neural networks and have produced consistent and continuous results.
-
ÖgeGame-based learning in architecture education: Consolidating visual design principles in freshmen(Graduate School, 2023-07-21) Karakaya, Mert ; Kanan Çekmiş, Aslı ; 529201016 ; Game and Interaction TechnologiesUsing games to teach people concepts has always been an attention-catching topic in the area of education. That is the main premise of game-based learning. Providing the players with manipulation tools such as the ability to change scale etc. to change the game environment by giving them more flexible options further facilitates their learning through trial and error, and encourages more involvement. With this aim, it was decided to create an artificial intelligence (AI) evaluated free creation video game aimed to consolidate visual design principle (VDP) knowledge in freshmen architecture students. VDPs are the underlying patterns that are present in designs people usually find appealing. The artificial intelligence that was used in the game was trained to detect some of these principles. There are more than only three main elements, however the artificial intelligence is trained on 3 main principles and their derivatives, which end up making up 9 sub VDPs. These VDPs are emphasis (color, isolation, shape), balance (symmetric, asymmetric, crystallographic), rhythm (regular, progressive, flowing). Artificial intelligence was decided to be used in the game because it can detect the trained VDP with success, as well as create a reliable checkpoint before students present their ideas to their instructors, which is time-consuming and might induce stress on some students. The usage of AI aims to eliminate some of that stress and the gaming nature of game-based learning is expected to motivate students further. The game welcomed the players with a menu screen in which they could choose to learn about VDPs with several example pictures and explanations. Following that, they could play a quiz mini-game where they were asked to choose a picture out of three pictures that held the VDP that was asked from them. All 9 sub VDPs, derived out of 3 main design principles, the artificial intelligence can detect were asked in order. Once they were done with the quiz, the game moved on to the "creation" section where the players were given an objective to create a composition with an emphasis on a specific VDP. The composition was managed by placing and manipulating design elements (basic shapes such as square, triangle, hexagon etc. or object textures) on a canvas with a grid. The design elements could snap onto the grid so that it would be easier to create more calculated compositions. When the players were happy with their composition, they submitted their compositions to the application in the background where artificial intelligence resides and that artificial intelligence evaluated that composition, giving a score to the player. The score showed the top 3 VDPs the composition had and the percentages of how much they governed the composition. Once players were done playing and creating their compositions, they were asked to fill out a questionnaire. Players' experiences about the game were obtained via the questionnaire, which based its scientific validity on the research named "Assessing the Core Elements of Gaming Experience (CEGE)". The questionnaire consisted of 24 statements in which students were asked to grade their level of agreement to using a 5-point Likert scale. 43 students participated in the questionnaire and all 43 of them answered every question. There was only one group and no control group. Following the questionnaire, a follow-up survey was given to 10 students. The follow-up survey consisted of 7 open-ended questions aimed to further understand what elements of the game were liked or disliked. According to the results gathered through that questionnaire, the majority of players stated that playing the game helped them consolidate their VDP knowledge, that they would want to see the game used in VDP education, and that they would want serious games to be used in education more. In addition, the results of the follow-up survey were helpful to see what was done right in the serious game, and what else could have been added to improve the game. This result points to the need for further research on game-based learning in architecture and shows some students' wish for more game-based learning in architecture education.
-
ÖgeExploring the role of game mechanics in generating spatial compositions: Snaris case(Graduate School, 2023-08-02) Özvatan, Ozan Can ; Alaçam, Sema ; 529201018 ; Game and Interaction TechnologiesAs the world becomes increasingly digitized, the study of virtual environments and user agency in it has emerged as an important field of research. Video games, in particular, serve as an influential subset of these digital spaces, presenting an array of dynamic, complex, and interactive worlds. These game spaces are experienced by users regularly and also raise interesting questions about human interaction, cognition, and experience within virtual realities. In a game, the player acts as both the subject, engaging with the gaming system, and the object, receiving responses based on their input. Their actions and decisions are interwoven into a complex system of cause and effect, where each decision leads to a change in the state of the virtual environment, thereby influencing their subsequent actions. This cyclical process organizes all decisions made during the gaming session, and the game space emerges as a result. This research is about user interaction with virtual spaces, exploring the impact of game mechanics -specifically risk and reward mechanics- on the player's agency in shaping the virtual environment, thus establishing a discussion on exploration at the intersection of game studies, architecture, and human-computer interaction. More specifically, the goal of this research is to answer the following questions: How does the involvement of risk and reward mechanics impact the spatial outcomes generated by participants using a 3D puzzle game? In which ways do the risk and reward mechanics affect the player's role in shaping virtual spaces? What is the impact of more challenging situations on the spatial composition in a virtual environment? How do software limitations influence the generation of spatial elements? To investigate the impact of risk and reward mechanics on generating spatial outcomes, we designed and developed a two-state digital application called Snaris, a 3d puzzle game, in which we can isolate various mechanics in two distinct modes. We compared the spatial compositions created under these distinct conditions by comparing scenes created in Play Mode which harbors risk and reward mechanics with scenes created in Build Mode, which lacks said mechanics. We proposed a method for qualitatively assessing spatial features of scenes created in Snaris. This method employs 12 criteria for evaluating unique 3D spatial compositions generated by the application. We collected data by conducting playtests with 20 participants with each submitting a scene from both modes of application. We observed that where the risk and reward mechanics exist, participants were usually more preoccupied with being able to shape the spatial outcome deliberately. Thus, creating more random and incoherent structures. On the other hand, the absence of risk and reward mechanics and a clear, unobstructed game environment allowed players to engage in unconventional actions, and create familiar topologies, in a setting that encourages exploration and experimentation throughout the session.
-
ÖgeGeneralized game-testing using reinforcement learning(Graduate School, 2023-10-17) Önal, Uğur ; Sarıel Uzer, Sanem ; Tinç, Kutay Hüseyin ; 529201019 ; Game and Interaction TechnologiesThe gaming industry has experienced significant growth and evolution, becoming a prominent sector in entertainment and technology. This growth has led to increased consumer expectations regarding the quality and complexity of games, prompting developers to explore innovative solutions to meet these demands. To meet these demands, one of the pivotal approaches adopted by game developers is the game testing process. Game testing is an incredibly resource-intensive procedure, demanding comprehensive evaluation of all aspects of a game through actual gameplay. To address this challenge and alleviate the associated workload, this thesis proposes an innovative approach to game testing. This method integrates a generic environment framework with reinforcement learning (RL) models, facilitating seamless communication between any game and an RL model under specific conditions. The framework optimizes the game testing process by capitalizing on the efforts of game developers. It relies on developers to compile and transmit essential information, such as state and reward data, to the generic environment. Subsequently, this data is processed and harnessed within the RL model, allowing it to learn and play the game in accordance with developers' intentions, while simultaneously generating valuable data for game testing purposes. This method also capitalizes on the beneficial aspect of game-playing AI agents trying out various actions in different states as they learn to play games. Game testing entails the creation of diverse scenarios by implementing different actions in various in-game situations. These scenarios are observed, and, when necessary, actions are taken in the game development process based on these observations. Therefore, as the situation where game-playing agents experience various scenarios closely resembles game testing, we can utilize not only the actions performed by agents during testing but also their behaviors during training as part of the game-testing content. The experimental phase of the study involved the deployment of six distinct builds of the same game, each serving as a means to test the functionalities of the generic environment and observe their impact on the behavioral patterns of RL models. These builds were thoughtfully crafted to uncover various aspects of RL model behavior and the diverse methods of representing game states. These builds can be summarized as follows: - Basic side-scroller: This build's purpose is to test the seamless communication between the generic environment framework, the game build, and the RL model. It features a simple reward system designed to guide the player to a target point, an action space consisting of three actions, and employs a state image as the state information. Exploration-oriented side-scroller:} Designed to encourage the player to explore the entire game area, this build incorporates a comprehensive reward system. It boasts an action space comprising four actions and utilizes a state image as the state information. - Exploration-oriented side-scroller with colored textures: This build serves as a variant of the exploration-oriented side-scroller build, with the only alteration being the modification of game textures. Its purpose is to investigate the impact of texture changes on the training of RL models. - Goal-oriented side-scroller: Sharing the same action space and state information as the exploration-oriented side-scroller build, this build primarily aims to observe the effects of reward system modifications. It employs a detailed reward system to guide the player toward specific objectives and a goal. - Exploration-oriented side-scroller using no image: With an identical action space and reward system structure as the exploration-oriented side-scroller build, this build seeks to examine how using a state array as state information influences the RL model's behavior. - Exploration-oriented side-scroller using image and array: Being similar to the exploration-oriented side-scroller build in action space and reward system structure, this build aims to maximize its impact on the RL model's behavior. It achieves this by employing both a detailed state array and a state image as state information. - Arcade: This build aims to demonstrate how the generic environment framework will perform in a completely different game. It has both exploratory and goal-oriented structures. It features a moderately complex reward system and an action space consisting of five actions. It uses both arrays and images as state information. The investigation into the communication system between the RL agent and the game build yielded valuable insights. It became evident that the generic environment framework played a crucial role in achieving positive and efficient outcomes. Nevertheless, the research also pinpointed areas ripe for enhancement, particularly concerning the reduction of the workload on game developers and the resolution of issues stemming from external factors. The logging system integrated into the generic environment has proven to be a valuable asset in the realm of game testing. It leverages the total reward accrued in each episode, efficiently guiding the selection of episodes meriting closer scrutiny. Furthermore, the supplementary information provided by this system offers exceptionally insightful data, greatly enhancing our comprehension of the actions taken in various gaming scenarios. Our proposed approach holds significant potential in the realm of game testing. It enables AI agents to adjust their behaviors by utilizing dynamic rewards and extensive state information from arrays and images to meet specific criteria. Moreover, successful game-testing outcomes have been consistently observed throughout both the training and testing phases, where agents adeptly exploit game vulnerabilities and uncover unforeseen features and bugs. In spite of the apparent successful outcomes, the implementation involving both a state image and a state array exhibited a notable reduction in training speed and encountered a substantial level of system load attributed to hardware constraints during the training process. When evaluated in accordance with the objectives of the thesis, it can be concluded that, overall, the proposed method has achieved successful outcomes in the game testing process and holds promise for future development potential. Further endeavors aimed at enhancing system performance may yield positive results concerning the broader applicability of game testing.
-
ÖgeTransforming structures into interactive experiences: Generative frameworks(Graduate School, 2024-06-27) Nardereli, Duygu ; Alaçam, Sema ; 529211005 ; Game and Interaction TechnologiesUnderstanding the complicated relationship between player experiences and game components is essential in the evolving game design landscape. This study explores this dynamic by offering practical and interactive game design exercises. These exercises, which are based on the Mechanics, Dynamics, and Aesthetics (MDA) framework, a key tool in game design, are designed to make them more accessible and actionable for game designers, emphasizing their crucial role in shaping player experiences. Gathering feedback at specific thresholds is crucial in the intricate game design and development process. Today's novice learners favor various learning sources, rapid information access, clearly defined objectives, and the ability to use technology to visualize the consequences of their decisions. This study offers a structured yet adaptable game design approach catering to these diverse learning preferences. The primary objective is to enhance the comprehensibility and practicality of the MDA framework for game designers by redefining its core components—mechanics, dynamics, and aesthetics. This endeavor is anticipated to establish a new language that will prove advantageous in workshops with individual and collaborative game designers. By introducing these concepts early in the game production, designers can understand the design perceptions necessary for crafting captivating games. The MDA framework plays a crucial role in bridging game design, game development, and technical background, enhancing comprehension of the design process. According to MDA, one of the critical components, 'aesthetics,' was designed as a destination, much like a compass. It guides the gameplay dynamics and mechanics, helping us map out a route for the emotional responses we want to evoke in the player. Mechanics are the only aspect of the game that designers have complete control over when it comes to fulfilling its emotional purpose, or aesthetics, by creating dynamics. Understanding all these experiences is crucial: It shows where the team should focus their efforts and how it can impact the player's experience when interacting with the game. This study uses board game mechanics as a foundation to avoid confusion in defining mechanics and ensure that selected mechanics evoke specific dynamics. The study began by examining the necessity of creating academic systems for creative processes, followed by an investigation into existing frameworks for design, game design, and serious game design. Secondly, common points and objectives are identified, allowing for a comprehensive analysis and redefinition of the foundational concepts of the MDA framework. The foundational concepts of the MDA framework—mechanics, dynamics, and aesthetics—were analyzed and reconsidered in conjunction with game design literature. This analysis helped to develop a theoretical foundation on how these frameworks could be utilized as content in game development processes. After establishing the theoretical foundation, board game mechanics were employed to transform MDA into a new generative experience. Using the BoardGameGeek (BGG) database and an academic study's classification, the definitions of BGG mechanics were revised. A scoring system was developed to determine the emotions a game mechanic could generate, creating a network of relationships visualized based on the "8 kinds of fun" in MDA. The proposed framework follows a card game structure with 28 different board game mechanics and components. These elements evolve into exercises focusing on various design issues requiring an iterative design process. The mixed-method approach validates the framework, where six expert game designers participate in a design workshop to evaluate MDA exercises and applicability with several parameters. This study aims to provide game designers with a half-structured tool for creating engaging and emotionally resonant games by redefining the MDA framework and employing board game mechanics. The innovative approach ensures that designers, whether novices or experts, can systematically explore and implement effective game design strategies. The next generation of game designers will produce breakthroughs in player experience, designing games with unique play mechanics and thinking beyond the existing parameters. This generative approach can help foster innovation and give game designers a solid process for exploring unusual questions about gameplay and player experience possibilities.
-
ÖgeOptimizing artistic process: Exploring efficient environment creation workflows in gaming industry(Graduate School, 2024-07-05) Özçiçek, Emrah ; Gül, Leman Figen ; 529211008 ; Game and Interaction TechnologiesAlthough there is no specific data about when researchers demand to study video games, it is understandable to experience an industry which reach 200B in just two decades and making 400 million in just 24 hours with a single title. The technological innovations and hardware improvements in early twenty-first century bring huge productions which eventually made consoles, computers and games more affordable and increase number of players in all groups, ages and genders. Game industry growing fast which pressure academicians and studios make more research about every aspect of developing, marketing, player and experience. Despite many research focusing why, only few of them describe how. This is most likely video game industry is relatively young and to develop certain tools, software and plugins for production; it is required to make research about why we need it, why we develop it. But asking why in many research raise another question which is how. Demand for comprehensive research becomes evident, not only to understand the underlay motivations of development but also explore practical methodologies and how to do. This need for practicality is mirrored to environment art realm where artist and designers are tasked to crating virtual worlds that all digital stories drive. Game development is consisting of four main disciplines. The first discipline is programming where developers create codes and blueprints to drive gameplay and mechanics like elements. The second discipline is design where designers craft gameplay and create room for interactions and mechanics that fit player experience. The third discipline is technical art and design where first two disciplines connect in a way to make overall gameplay, mechanics, player experience and interactions flow as expected. The fourth discipline is art which is visual representation of all other disciplines on top of story and narrative that pull the player into game world. Art in video games involves of concept art, prop art, texture art, lighting art and more. Environment art can be defined as combination of all art disciplines in a video game. Each environment is a synthesis of location, culture, history, theme, lifestyle and atmosphere like elements and has a purpose of answering questions like where gameplay located, who lives there and how they live, what is their culture and lifestyle or what happened there. Overall, an environment is all visual elements of digital world that game characters and player live and interact. Environment artist are specialized artists responsible to make environment scenes. These artist uses various techniques and methods to create desired looking that pull player into game world. In this creation process, they follow various environment art creation steps. These steps are initial planning and referencing, conceptualization, blackout and proxy meshing, modeling and sculpting, texturing, set dress, lighting and finalization. Environment artists use various methods, strategies and approaches as their workflow to create an environment. Workflow as a term means a repeatable pattern or series of steps to produce an activity. Workflows in game context means the sequence of steps from initial planning and concepting to final game art creation process. The purpose of a workflow provides a repeatable process from beginning to end; therefore, the process can apply to multiple projects in multiple time frame. It is essential to note that while environment art creation involves steps to the medium or underly technology, these steps are distinct to be considered as a workflow or a base. Conversely, a workflow in environment art means an optimized or refined variations of art creation steps. Therefore, the creation of an workflows mainly refers to a need, to refine one or more environment art creation stage to possibly reduce time, cost or increase art quality and efficiently. The purpose of this thesis was to understand how game environment art workflows are being used, improved, and developed by artists with a focus on well-organized, optimized and high-quality work to create narratives that enhance player experience. By analyzing approaches and methods used by artists and literature, this study identifying efficient practices and strategies for the creation of video game environment arts. The final purpose of this thesis is presenting finding in a format that is accessible to artists, studios, and educators working in the video game industry whom aims to implement these practices and strategies into their environment art production pipeline. In the introduction part of this research, we discussed the purpose of our work and significance of thesis following by our research questions and scope, we finished this part by defining research method. The second section of this research paper started with literature overview. In literature section we discussed two main questions. These questions were what is environment art in video games and what workflows actually mean. We discussed these two main terminologies Following by detail clarification of workflows stages specialized to environment art, these were important because a workflow mainly refers to a need, to refine, optimize, shorten, increase one or more environment creation stage in terms of time, cost or quality. This section of research continued with background research, particularly previous works and history of environment art in video games from very first two-dimensional(2D) graphics to this moment. This part included examples of environment art and industry developments. We finished this section with current environment art workflows trends and previous studies made for this topic. In the third part, we introduced core part of this work where we explain our methodology for this thesis. We adopted grounded research method for our research. This method has been used in many areas including sociology, psychology and business and become very effective for its ability to generate new insight and theories that based on real world data. This method is especially suitable for dynamic research fields that has fewer existing resources and has a need to construct ideas based on evidence itself, as research field has low literature references, it was essential to benefit from real world data and external contributors. Our methodology consists of three data set and study type; preliminary study, experimental study and main study. These studies created and used data sets, preliminary data, experimental data and semi-structured interview data which later on called as main data. Environment art field evolve rapidly whereas due to dynamic nature of research topic, some of literature finding may already replace with new methods while some of them may completely obsolete over time. This created a problem, because investigating replaced or unused methods were be unpractical for research goals. Therefore, a study for ensuring literature being reviewed is relevant data was needed. This study was preliminary study which only examined recent literature. In this study, we further detail recent literature and observations to identify active literature while eliminate obsoleted part of it. This study only focuses each environment art creation stages because workflows mainly refer to a need, to refine one or more environment art creation stage. Therefore, only focusing environment art creation stages was a logical method for such purpose since any changes on environment art creation stages would dictate workflow and method itself. Also, this approach would be efficient and practical to ensure literature being reviewed was relevant and up to date. This approach allowed us pinpoint areas where refinement, change or obsoleted happened. The study in general showed us; 2 dimensional, manual and time-consuming methods are greatly replaced while more automated, algorithm or AI based workflows has a trend. Furthermore, this study provided strategies and methods for efficient practices visible in literature for each environment art creation stages. The end result of this study created preliminary data. The dynamic nature of research topic created yet another problem. As the research topic was extremely dynamic, some of preliminary study data may be too new and only have theorical background without practical base. Experimental study was a practical study for this need. Therefore, we created an experimental project based on preliminary data. In this study, we created a dedicated environment art project named ''Living in the Bubble of Love''. This study focused preliminary data and measured the practical base of it. Therefore, some of preliminary study findings like procedural workflows and modular workflows has a practical base while AI based workflows may need further definition, tool, learning resources and most importantly regulations for copyright. This study created our second data set which is experimental data. Preliminary study and experiment project study created two data type, these were preliminary data and experimental data. These two data sets used as a base for preparing a semi structured interview. Therefore, based on previous studies we prepared a semi structured interview framework which included an extensive survey along with minimum 4 industry artists and compare their approach for each step of environment art creation pipelines. The end result of this interview created last data set which is main data. This data, preliminary data and experimental data used for final main study. The main study was final analyzing and writing of all data sets using grounded research method to generate theories for literature, research goals and environment art field. The main study started with our goals and scope following a detailed explanation about our data collection and analysis methods. We used Grounded Theory as both analysis methods data collection method. This section of study started with semi structured interview framework, including participant selection criteria like relevant field and experience; recruitments methods like professional networks and referral methods. Semi structured framework setup continued with confidentiality procedures, data sets, final interview layout and interview questions. The section ended with main study analysis and findings. The analyzing used open coding, axial coding, selective coding, theorical sampling, theorical saturation and memo writing to analyze all three data sets. This analysis and finding included discussion about methods and workflows we interviewed and experimented during this research. The final theoretical sampling and memo writing groups created various theories as follows. 1-Base workflows, 2-Tools and compatibility, 3- Prototyping, 4-Feedback and collaboration, 5-Importance of planning stage, 6- Artist Role, 7- AI workflows and 8-Automated workflows. In the last part, we concluded with contribution of this research and future prediction. The thesis specially focused artistic process of game environment and it presented in a format that is accessible to artists, studios and educators working in video game industry, this thesis filled an important gap in literature and proving a valuable insight for those looking to improve their environment art workflows. Furthermore, this research analyzed approaches and methods used by artists working in game industry, identified efficient practices and strategies for creation of game worlds and aid to advance in game environment art workflows. The presentation on this study could be used to inform best practices and aid new developments in industry. Meanwhile, this study brought a deeper understanding of game art in video game and highlight the importance of art in game development and the needs for a greater focus on creative side of game development. This could lead to a greater appreciation for the role of artists in game development and encourage researchers to study artistic perspectives in game development. Research topic has bright feature in computer graphics. Finding can be used for related research or an upper education level like PhD project. These projects could be like Workflows Developments and managements various AI implementation projects and AI training projects for environment creation purposes.
-
ÖgeAssessing the influence of color blindness on player engagement and emotional experience in digital games(Graduate School, 2024-07-11) Tillem, Merve ; Gün, Ahmet ; 529201022 ; Game and Interaction TechnologiesColor blindness is a visual impairment that primarily affects male individuals. People with color blindness cannot see colors in the same spectrum as normal vision, leading to visual inaccuracies and misunderstandings. Color blindness can negatively affect many areas of life, including the experience of digital games. In digital games, colors are often used to transmit important information, such as distinguishing enemies from allies or indicating targets. Individuals with color blindness may have difficulty interpreting these visual cues, making the game less enjoyable and even affecting their ability to play. Previous research on accessibility has rarely concentrated on the intersection of color blindness and digital gaming. This study aims to examine the relationship between game design and color blindness through a comprehensive literature review, find gaps in current research, and evaluate the gaming experiences of individuals with color blindness regarding performance, emotional responses, and engagement. This study employed a group experiment to understand the experience of color blindness. To achieve that we developed a game called Color Quest, which was played silently by 5 color-blind and 8 participants with normal color vision. The participants were then asked to fill out a feedback form. Additionally, color-blind individuals were asked to play the game a second time with sound. Players were instructed to record their screens during gameplay, and their faces were simultaneously video-recorded. Facial recordings were analyzed using Emotion AI to capture real-time emotional changes. The data was output in CSV format and visualized in PowerBI. The analysis revealed that color-blind individuals did not react less than those with normal vision, but their reaction rates increased when playing the sound-supported version. This suggests that auditory support is a significant factor in enhancing player engagement. In the feedback forms, color-blind individuals mostly rated the silent game experience as 'Neutral,' while the majority of normal vision individuals selected 'Very connected.' In other words, individuals with normal vision reported a more positive visual experience compared to those with color blindness. When playing the game with sound, most color-blind individuals chose 'Connected,' indicating that sound enhancements support the emotional connection to the game. In the silent version, color-blind participants had an average completion time of 275.4 seconds, compared to 248.3 seconds for normal-sighted players. Color-blind individuals made an average of 9.4 wrong moves throughout the game, while normal-sighted individuals made 7.6 wrong moves. These findings shows that color blindness negatively impacts player performance. These findings underscore the crucial role of visual design in games, impacting both player performance and experience. While the video analysis of color-blind individuals did not reveal a distinct negative trend, the feedback clearly indicates that color blindness does have a detrimental effect on the game experience. Importantly, the study also demonstrates the positive influence of auditory support on player experience, a key insight for game developers and researchers in the field of accessibility and game design. It is important to acknowledge that this study was conducted with a limited number of participants, and individual differences and experiences could have influenced the results. However, these initial findings provide a strong foundation for future research. By involving more participants, designing more comprehensive games, and implementing standardized conditions, we can expect more definitive and insightful results, further advancing our understanding of game accessibility and color blindness.
-
ÖgeExploring the factors that affect game production process in casual mobile games(Graduate School, 2024-07-16) Yayla, Yavuz Selim ; Gün, hmet ; 529211016 ; Game and Interaction TechnologiesVideo oyunu, sanal bir ortam sunmak için etkileşimli oyun ve bilgisayar teknolojisinin kullanımını içeren bir dijital medya biçimidir. Video oyunları kişisel bilgisayarlar, oyun konsolları ve mobil cihazlar da dahil olmak üzere çeşitli platformlarda oynanabilir. Oyunculara bir eylemlilik duygusu sağlama yetenekleri ve etkileşimli yapıları ile karakterize edilirler ve genellikle oyuncuların becerilerine ve problem çözme yeteneklerine meydan okumak için tasarlanmıştır. Video oyunları, oynanış mekanikleri ve diğer özelliklerine göre çeşitli türlerde sınıflandırılabilir. Video oyunu türlerinin bazı yaygın örnekleri arasında aksiyon, macera, rol yapma, simülasyon ve strateji yer almaktadır. Aksiyon oyunları tipik olarak hızlı refleksler ve el-göz koordinasyonu gerektiren hızlı tempolu oyun ve zorlukları içerir. Macera oyunları genellikle hikaye ve keşfi vurgular ve oyuncuların problem çözme becerilerini kullanmalarını gerektiren bulmacalar ve diğer zorlukları içerebilir. Rol yapma oyunları oyuncuların bir karakter yaratmasına ve kontrol etmesine izin verir ve genellikle karakter ilerlemesi ve özelleştirme unsurlarını içerir. Simülasyon oyunları gerçek dünya sistemlerini veya faaliyetlerini taklit etmeye çalışır ve genellikle strateji ve kaynak yönetimi unsurlarını içerir. Strateji oyunları oyuncuları stratejik kararlar almaya ve gelecek için plan yapmaya zorlar ve kaynak yönetimi ve birim kontrolü unsurlarını içerebilir. Oyun geliştirme, bir video oyunu oluşturma sürecidir. Çok çeşitli görevleri içerir ve karmaşık ve zaman alıcı bir süreç olabilir, ancak aynı zamanda oyun yaratma konusunda tutkulu olanlar için inanılmaz derecede ödüllendirici bir deneyim olabilir. Oyun geliştirme süreci tipik olarak oyunun kavramsallaştırılmasıyla başlar. Bu, oyun için bir fikir bulmanın yanı sıra hedef kitlenin, türün ve oyun mekaniğinin belirlenmesini içerir. Geliştirme sürecinin geri kalanına rehberlik edeceği için bu aşamada oyun için net bir vizyona sahip olmak önemlidir. Oyun için konsept oluşturulduktan sonra, bir sonraki adım oyunu tasarlamaktır. Bu, oyunun sanat stilini, karakter tasarımlarını, seviyelerini ve diğer görsel unsurlarını oluşturmayı içerir. Ayrıca oyun için bir hikâye veya senaryo oluşturmayı da içerebilir. Oyun tasarlandıktan sonra, bir sonraki adım oyunu programlamaya başlamaktır. Bu, oyunun amaçlandığı gibi çalışmasını sağlayacak kodun yazılmasını içerir. Oyunun karmaşıklığına bağlı olarak, bu süreç nispeten basit veya oldukça karmaşık olabilir. Bir oyunun programlanması tipik olarak aşağıdakiler gibi birkaç alt görev içerir: - Oyun motorunun kurulması: Oyun motoru, oyuna güç veren ve grafik oluşturma, ses çalma ve kullanıcı girdisini işleme gibi görevleri yerine getiren yazılımdır. - Oyun mekaniğinin uygulanması: Bu, karakter hareketi, çarpışma algılama ve oyun fiziği gibi oyunun çeşitli oynanış unsurlarının kodlanmasını içerir. - Oyun mantığının uygulanması: Bu, seviye ilerlemesi ve oyun bitti koşulları gibi şeyler de dahil olmak üzere oyunun akışını kontrol eden kodun yazılmasını içerir. - Varlıkları entegre etme: Bu, oyun için oluşturulan modeller, dokular ve sesler gibi çeşitli varlıkların içe aktarılmasını ve entegre edilmesini içerir. Oyun programlandıktan sonra sıra test etmeye gelir. Bu, oyunun düzgün çalıştığından emin olmak ve herhangi bir hatayı veya aksaklığı tespit edip düzeltmek için oyunu oynamayı içerir. Oyunun son kullanıcı için mümkün olduğunca cilalı ve eğlenceli olmasını sağlamak için kapsamlı bir şekilde test edilmesi önemlidir. Bir oyunun test edilmesi tipik olarak aşağıdakiler gibi birkaç alt görev içerir - Oyun testi: Bu, insanların oyunu oynamasını ve oynanış, kontroller ve oyunun genel keyfi hakkında geri bildirim sağlamasını içerir. - Hata testi: Bu, oyundaki hataların veya aksaklıkların belirlenmesini ve düzeltilmesini içerir. - Uyumluluk testi: Bu, oyunun çok çeşitli sistemlerde düzgün çalıştığından emin olmak için çeşitli farklı donanım ve yazılım yapılandırmalarında test edilmesini içerir. Oyun test edildikten ve gerekli değişiklikler yapıldıktan sonra sıra oyunu piyasaya sürmeye gelir. Bu, pazarlama materyalleri oluşturmayı, dağıtım kanalları kurmayı ve oyunu hedef kitleye tanıtmayı içerebilir. Özetle, oyun geliştirme, kavramsallaştırma ve tasarımdan programlama ve test etmeye kadar çok çeşitli görevleri içeren karmaşık bir süreçtir. Yaratıcılık, teknik uzmanlık ve detaylara gösterilen dikkatin bir kombinasyonunu gerektirir. Zorlu bir süreç olsa da, başarılı bir oyun yaratmanın sonucu, oyun geliştirme konusunda tutkulu olanlar için inanılmaz derecede ödüllendirici olabilir. Çalışma kapsamında, oyun geliştirme sürecine ilişkin daha önce yapılmış çalışmaları incelemek üzere sistematik bir literatür taraması gerçekleştirilmiştir. Tarama çalışması Web of Science, Scopus ve ACM Digital Library gibi kapsamlı veri tabanları üzerinde gerçekleştirilmiştir. Bu veri tabanları "Oyun Geliştirme" ve "Süreç" anahtar kelimeleri kullanılarak taranmıştır. Şirketlerin çalışan sayıları 3 ila 10 yıl arasında değişmektedir ve nadir de olsa 50'yi aşar. Ürün geliştirme ve oyun tasarımı departmanları rol oynarken, Yazılım ve Sanat ekipleri esastır. Şirketler, özellikle sanat ve seslendirme konularında genellikle dış kaynak kullanırlar ve bazılarının kendi altyapıları vardır. Oyun sektöründeki şirketler projelerini yönetmek için Waterfall, Agile, Scrum, Kanban, Scrumban ve PPP gibi çeşitli proje yönetim yöntemleri kullanmaktadır. Sanat içeriğindeki tutarlılık, deneyimli ekiplerin daha iyi çalışmasıyla deneyler yoluyla elde edilir. Oyun mekaniği ve yazılımı yinelemelidir, bazı ekipler daha büyük değişiklikler yaparken diğerleri daha küçük değişiklikler yapar. Proje yönetim yöntemine bağlı olarak kontrol ve değerlendirme toplantıları haftalık olarak yapılır. Çok oyunculu özellikler yaygın değildir. Analizler tasarım sürecinin sonlarına doğru kullanılır ve önemli analizler yumuşak lansman aşamasında yapılır. Oyunlarda kullanıcı geri bildirimi önemlidir. Şirketler atlas ve içerik boyutu gibi kendi optimizasyon yöntemlerini geliştirir. Hikaye ve anlatım genellikle basit mobil oyunlarda kullanılmaz. Bir oyun geliştirme projesinin yönetimi tipik olarak CEO ve ürün müdürünü içerir ve her 1 veya 2 haftada bir kontrol toplantıları yapılır. Daha esnek çalışan şirketler genellikle görevleri açıkça yazmazlar ve bunları uzatabilirler. Sektörde geliştiricilere, sanatçılara ve oyun tasarımcılarına ihtiyaç duyulur; muhasebe, finans ve İK gibi roller ise yönetim kademesindeki çalışanlar veya dış kaynak hizmetleri tarafından yerine getirilir. Çizelgeleme genellikle katmanlıdır ve ekipler zamanı kendi aralarında daha küçük görevlere bölerler. Bazı görevler zamana tabi değildir ve görevler zamanında ulaşmadığında, daha az önemli özellikler kaldırılır. Şirket kültürü, görev çeşitliliğinde önemli bir rol oynar; belirli durumlarda sabit atamalar kullanılır ve iş tanımları gerektiğinde değişir. Test yaklaşımları farklılık gösterir; bazı şirketlerde sorumlu belirli kişiler bulunurken bazılarında bulunmaz. Yerelleştirme ve seslendirme çoğunlukla dış kaynaklıdır, bazı şirketler ise pazarlama ve muhasebe için dış şirketlerle çalışır. Para kazanma modelleri, kitleye bağlı olarak oyuncu satın alımları ve reklamların kullanılmasıyla değişebilir. Hedef kitle çok önemlidir, çünkü daha küçük şirketlerin özellikle hukuk ve pazarlama alanlarında süreçleri uyarlamaları gerekebilir. Şirket kültürü önemli bir rol oynar; bazı şirketler daha sabırlı ve esnekken, diğerleri daha esnektir ve daha verimli küçük projelere izin verir. Kanunlar da sektörü etkiliyor. 2019'da dünya çapında kişisel verilerle ilgili yapılan değişiklikler araçları kullanılamaz hale getiriyor ve şirketlerin kendi analiz araçlarını oluşturmalarını gerektiriyor. Şirketlerin çoğu şirket ve çalışan deneyimine öncelik veriyor ve on şirketten dördünde ilk sırada yer alıyor. Bu da şirketlerin süreçleri oluştururken know-how'a öncelik verdiğini gösteriyor. Çalışan devir hızı da önemlidir ve iki şirket bu kategoride ilk sırada yer almaktadır. Bu faktöre öncelik veren şirketler, daha uzun süreli bir çalışan deneyimi sağlayarak kurumsal kimliklerini koruyor. Bir diğer önemli kriter ise para kazanma modelindeki hedef kitle, kullanıcı geri bildirimi kontrol noktaları, oyun türü, özellikleri ve mekanikleridir. Bu faktörler sektörün gerçeğidir ve geliştirme sürecini değiştirir. Şirketler vizyonları veya bakış açıları ne olursa olsun bu faktörleri göz önünde bulundururlar. Şirket büyüklüğü, dış kaynak kullanımı stratejisi ve şirket yaşı gibi kriterler birçok şirket tarafından daha alt sıralarda yer almaktadır. Dış kaynak kullanımı stratejisi ve şirket yaşı, süreç inşasında daha sonra dikkate alınır. Şirketler süreçleri oluştururken on yaşındaki veya 20 yaşındaki şirketlerini göz önünde bulundurmalıdır, çünkü her zaman aynı yaklaşımı izlemeyebilirler.
-
ÖgeDynamic difficulty adjustment by changing enemy behavior using reinforcement learning(Graduate School, 2024-07-25) Akşahin, Burak Furkan ; Sarıel, Sanem ; 529201003 ; Game and Interaction TechnologiesDynamic difficulty adjustment (DDA) systems are essential in modern gaming to provide to the diverse skill levels of players. These systems ensure that games remain challenging yet enjoyable by automatically adjusting the difficulty based on the player's performance. Traditional fixed difficulty settings often fail to provide an optimal experience for all players, leading to frustration for less skilled players and boredom for more skilled ones. Implementing DDA systems aims to enhance player engagement and satisfaction by maintaining an appropriate level of challenge throughout the game. Various techniques have been explored to implement DDA systems. One common approach is dynamic scripting, which involves adjusting the game's rules and parameters in real-time based on the player's actions. This technique allows for a more responsive and adaptable gaming experience. Other methods include player modeling, which uses data from the player's performance to predict their future behavior and adjust the difficulty accordingly, and machine learning algorithms that continuously learn and adapt to the player's skill level over time. Reinforcement learning (RL) has emerged as a powerful tool for developing DDA systems. In this approach, Artificial Intelligence (AI) agents are trained to play the game and learn optimal strategies to maximize their rewards. These agents can then dynamically adjust the game's difficulty by modifying the behavior of non-player characters (NPCs) or the game's mechanics based on the player's performance. This method allows for a more nuanced and effective DDA system that can adapt to the player's skill level in real-time. In this thesis, a DDA framework was created to be used in various gaming environments. Three different game scenarios were developed to demonstrate its effectiveness: a basic shooter, a basic action game, and a complex action game. Each of these scenarios provided a unique set of challenges and complexities, allowing for a thorough evaluation of the framework's adaptability and performance. The developed framework is capable of analyzing the performance of AI agents against human players and suggesting new difficulty levels accordingly. All parameters for these difficulty adjustments can be modified in the editor, providing game developers and designers with the flexibility to tweak the system to suit their specific needs. This capability ensures that the DDA system remains effective and relevant across different games and player demographics. The BrainBox plugin, developed for Unreal Engine, is a versatile tool designed to facilitate the creation of environments for DDA systems. It seamlessly communicates with a Python backend, managing the complex interplay between game environments and AI training processes. The plugin efficiently handles the creation and management of game environments, executes player and agent actions, calculates rewards, and xxii implements difficulty change procedures. This integration ensures that game developers can easily implement and tweak DDA systems, enhancing the gaming experience by maintaining an optimal level of challenge. A Python backend was created for training and evaluating the RL models. This backend communicates with the game environments created in Unreal Engine using Transmission Control Protocol (TCP), facilitating seamless integration between the training process and the game. The backend is responsible for managing the training data, running simulations, and updating the models based on the results, ensuring a robust and efficient training process. In the complex action game scenario, models were trained and evaluated to determine their effectiveness. The models were ordered by their median rewards across 20 episodes and mapped into difficulty levels. This process allowed for a detailed analysis of the models' performance and provided insights into their learning capabilities and adaptability to different levels of game complexity. A test case was conducted with 20 participants of varying game experience and skill levels. In this test case, the DDA was benchmarked with multiple testers. All sessions were logged, and a comprehensive analysis was performed on the collected data. This analysis provided valuable feedback on the system's performance and effectiveness in real-world scenarios, highlighting areas for improvement and potential future developments. In conclusion, the DDA demonstrated a robust capability in tailoring game difficulty to individual player needs. Its ability to adapt in real-time, guided by both player performance and feedback, highlights its potential to enhance gaming experiences significantly. The findings suggest that the DDA system not only improves player engagement and satisfaction but also offers a scalable solution for balancing difficulty in a wide array of games. Future implementations could benefit from refining this system to further optimize player retention and enjoyment, ensuring the game remains accessible and rewarding for all players regardless of their initial skill level.
-
ÖgeDeep learning for game genre classification from game posters(Graduate School, 2024-12-06) Güneş, Batıkaan ; Uzer Sarıel, Sinem ; Game and Interaction TechnologiesThis thesis aims to classify the genres of video games using a deep learning based model. The models, which were developed based on the ResNet50V2 and InceptionV3 architectures in the Keras library, were trained with game capsule images taken from the Steam platform and genre labels determined by game publishers. In addition, an online survey was conducted in which a group of participants classified games according to their genres. The results were used to comparatively evaluate the ability of both human participants and deep learning models to infer genres from the image content. The survey participants performed this task with more success than the ResNet50V2 model. Among the genres analyzed, it was observed that the prediction scores of the "indie" and "action" genres were higher than the "RPG" and "strategy" genres. This difference was also observed when the survey results were compared with the performance of the model using ResNet50V2. While the predictions made by survey participants show less variation between genres, the low performance of the ResNet50V2 model can be attributed to the different proportions of genres in the dataset. Therefore, it is recommended that the study be repeated with a more extensive data set, especially for these species. The variation in the survey results may be due to the ambiguity in the definition of RPG and strategy games or, in other words, the broader definition of these genres. The discrepancy in the ResNet50V2 model could be caused by the problem's structure or dataset size. Consequently, we trained two more models with the same number of data for every genre and one genre for every game. Again the ResNet50V2-based model performed poorly across most genres, and changing the dataset and issue format had no considerable effect on performance. Nonetheless, the model constructed using InceptionV3 demonstrated noteworthy advancements, especially with an RPG success rate of 82\% and an overall accuracy of 52\%. These findings imply that InceptionV3 outperforms ResNet50V2 in this challenge, suggesting that it is a more effective design. When the games that the ResNet50V2 model and the survey participants performed well and poorly were divided into two groups, it was observed that action, adventure, indie and casual game genres were more prominent in the well-performed group. It was stated that the images in the poorly performed group usually consisted of only the game name written on a simple background and that the common opinion that these images were incomprehensible was a positive result for the "providing design feedback" use case of the model. Likewise, the occurrence of cases where the model and survey participants marked the same genres but did not match the game genres on the Steam platform indicates that the model can be used for the "genre validation" use case. A noteworthy point between this study and previous studies is that there is no fixed standard for determining game genres. To overcome this problem and to find hidden relationships between games, unsupervised classification studies can be carried out in the future, and genres can be grouped into clusters without naming them, and the consistency of these clusters can be examined. Based on the responses from survey participants, the game images were reclassified, and new models capable of generating genre-specific game images were created using the stable diffusion 1.0 model. Subsequently, these models were employed to generate sample images for imaginary games, and a new survey was conducted. In this survey, the same pool of participants was asked to match the images with the corresponding genres. Unlike the initial survey, the independent game genre, which defines the organizational structure of the game developer team, and the casual game genre, which describes the complexity of the game, were excluded from this study based on the feedback received. The results of the second survey were similar to those of the first survey. When compared to the base model, the genre-specific image generation model produced images with higher genre predictability. As a result, using a deep learning models in recommendation systems based on game genres will improve the performance of the recommendation system and genre-specific image generation model has potential to assist designers. This thesis study contributes to game development and marketing processes by revealing the power of game images to reflect genres and the difficulties in accurately determining genres.