University: The Bartlett School of Architecture, UCL
MSc Architectural Computation
Project Description
An essential part of location-based games is the physical space where players engage in gameplay. Players navigate through the city, their game world, using a 2D map on their smartphones to make gameplay decisions and identify points of interest (POI). This research aims to utilize space syntax, which analyzes the city's configuration and how individuals move based on the geometry of spaces, to establish new rules and a revised logic for placing objects in the environment of location-based games. The player's visual field in the game environment from their current location can influence the decisions they make while navigating the city. While the game designer cannot alter the geometry of the real environment, they can leverage the visibility and spatial characteristics of existing spaces for various purposes and functions. Firstly, this paper examines existing location-based games, specifically Pokémon Go and Ingress, which employ a landmark-based logic, to analyze their distribution and underlying principles. The primary focus of this research is on selected areas in the central part of London. Initial findings indicate that a high percentage of POI in Ingress are situated in small spaces with low connectivity values, as well as spaces where the visual field is obstructed by obstacles. Based on these observations, new placements of POI are suggested, drawing upon spatial analysis theories, gaming principles, and existing literature. Space syntax measures serve as a foundation to prioritize certain locations within the game environment over others. These potential placements are evaluated by comparing them to the logic employed in the selected games. By considering the player's position and visibility in the gameplay location and its surroundings through visibility and street network analysis, a novel approach to object placement in the real environment is devised. Additionally, an alternative logic combining space syntax and landmarks is proposed, exploring whether it can offer additional value that the existing approach lacks.pace Syntax Symposium. 10th International Space Syntax Symposium. London.
Research Questions
Can the configurational theory of Space Syntax be utilized for the placement of points of interest (POI) in city gameplay, considering the player's position and visibility in the play location and surroundings?
Objectives:
Investigate the distribution of POI in Ingress and Pokémon Go games in central London, examining potential dependencies on parameters such as centrality, accessibility, connectivity, open space, visibility, obstacles, landmarks, and green spaces.
Explore the evaluation of game spaces for augmented reality (AR) games like Ingress and Pokémon Go through the application of Space Syntax spatial measurements.
Background
Location-Based Mobile Games (LBMGs) are smartphone-based games that merge digital experiences with real-world environments, with gameplay being influenced by players' physical locations (Spallazzo, Mariani, 2018).
In location-based games, game designers must adhere to the city's geometry as the gamespaces exist in the real environment rather than virtual structures (Ejsing-Duun, 2011).
Space Syntax defines spaces as voids, specifically streets, squares, and parks on an urban scale. These spaces may have obstructions such as walls, fences, or buildings that restrict access and obstruct vision. Each space is connected to other areas and holds relational significance (Al-Sayed et al., 2014: 7).
Significant correlations have been observed between spatial configuration attributes of an environment and pedestrian movement. Space should be regarded not only as gaps between buildings but also as potential for movement.
In Ingress, players are provided with a map of Portals, which are real-world locations transformed into POI for the game (Moore, 2015). Pokémon Go allows players to interact with three distinct point datasets: PokéStops, Pokémon gyms, and spawnpoints (Juhász and Hochmair, 2017).
Methodology
Existing location-based games were examined, analyzing their distribution and underlying principles. Specifically, Pokémon Go and Ingress, which utilize a landmark-based approach, were discussed. The central part of London was selected as the study area. Based on spatial analysis theories, gaming principles, architectural theories, and space syntax theory, new POI placements were suggested.
The research explores a new logic for placing POI in location-based games using visibility and street network analysis, guided by space syntax measures. Additionally, a combination logic incorporating landmarks and space syntax is proposed.
Methodology diagram
Step 1: Background and literature review: Analyzed location-based games and spatial navigation.
Step 2: Methodology: Chose Ingress and Pokémon GO games and investigated the distribution of their points of interest (POI) in the specified area. Ingress uses Portals, while Pokémon Go uses Pokéstops and Gyms as POI. Explored the landmark-based game logic.
Step 3: Results: Explored a new logic for placing POI in location-based games using space syntax measures, visibility, and spatiality. Proposed new placements and suggested a combination logic of landmarks and space syntax.
The spatial analysis measures used in this research and how they were used in the game logic.
Results
The player's visual field from their position in the game environment can influence their navigation decisions in the city (Summers, 2015). The study explores innovative ways to utilize physical locations as gameplay spaces, capitalizing on the unique opportunities they offer (Ejsing-Duun, 2011). Space syntax measures serve as a starting point for considering specific locations within the game environment over others.
New POI positions in location-based games are suggested, offering distinct features within the game logic based on the chosen spatial analysis metrics. The games can benefit from adopting this new logic and navigation approach, considering the city's characteristics. These metrics can also be combined to enhance distribution in the area. Furthermore, a combination of space syntax and landmarks is proposed as an alternative logic, potentially adding value to the existing landmark-based approach.
Pokémon GO Map with POI
Ingress Map with POI
POI that the players will come across in a selected a path, the shortest route, from an origin to a destination
Selected path (shortest path) and Pokémon Go Gyms (7 Gyms)
Selected path (shortest path) and Ingress Portals (32 Portals)
Two different visual fields, shapes and sizes of the isovist at Gordon Street.
Ingress portal locations (Points with orange color) extracted from Ingress Intel Map.
Points of Interest database (Points with cyan color) extracted from Digimap.
Specific metrics provided by DepthmapX were utilized to analyze properties of isovists, such as size, shape, local potential for exploration, control of the visual field, and areas significant for movement. The objective was to propose alternative rules for distributing points of interest (POI) in location-based games.
New Placements and Evaluation
Ingress Portals and Pokémon Go Gyms and Pokéstops were used as examples to test these new logics for potential positions. Graphs were generated to display the highest metric values in specified areas. Rhino, Grasshopper, and C# were employed to create these graphs.
The previous POI placements were categorized into groups based on their position and distance. Then, the points within each group with the highest metric values were selected. The suggested POI for these location-based games are represented by black-colored points on the new graphs. Additionally, these new placements were combined with landmarks, denoted by cyan-colored points. Rhino, Grasshopper, and C# were used to generate these graphs.
Street network analysis of Central London
Street network analysis, Segment map, Choice 2km Rank
Space syntax techniques can identify places with higher or lower levels of urban flow and accessibility. Choice measures the movement flows through spaces and represents the distribution of movement. Choice indicates how frequently a street segment appears on the shortest path from all origins to all destinations (Hillier and Vaughan, 2007).
Street Network Analysis - High Values
Streets with high choice values are the most connected streets, facilitating high player movement.
Choice combined with landmarks (Points with cyan color).
The new placements were placed in the most connected streets.
Street network analysis, Segment map, Integration 2km Rank
Integration reflects the accessibility of a street and measures the number of directional changes required to reach all other segments. Higher integration values indicate fewer turns (Ibid.).
Street Network Analysis - High Values
Streets with high integration values are more frequently visited by players with less effort.
Integration combined with landmarks (Points with cyan color).
The new placements were placed in streets where players would end up more often.
Visibility analysis of Central London
Visibility analysis, Connectivity
The connectivity graph illustrates the number of visible points from a specific location in an area, indicating the visible space from that point.
Visibility Analysis - High Values
Choosing the highest connectivity values, these placements can be designated for easily discoverable objects. For example, in Pokémon Go, these placements can be for the most common Pokémon.
Connectivity combined with landmarks (Points with cyan color).
The new placements have connectivity values higher than 600 and are located in highly visible positions in the city for players.
Visibility analysis, Isovist Min Radial
The Isovist Min Radial metric represents the minimum distance from a point to obstacles that form the isovist, which is the shortest view ray. It indicates how close a person is to obstacles like walls and buildings (Koutsolampros et al., 2019).
Visibility Analysis - High Values
Choosing the highest Isovist Min values, these placements can be suggested as meeting points or events in Ingress, where many people gather or work in groups, and should be located in open and spacious areas. In Pokémon Go, these positions can be suggested as possible Gym locations where players battle each other, attracting a large number of people.
Isovist Min combined with landmarks (Points with cyan color).
The new placements are positioned at least 20m away from obstacles, ensuring they are not located in narrow streets but rather in squares, parks, and wide roads.
Visibility analysis, Through Vision
Through Vision metric can be used to identify locations most likely to be traveled, and these areas are important for movement.
Visibility Analysis - High Values
These positions are most likely to be passed a lot by the players and can be for objects which are easy to find. For example, in Pokémon Go, these placements can be for the most common Pokémons.
Through Vision combined with landmarks (Points with cyan color).
Visibility analysis, Isovist Occlusivity
Isovist Occlusivity shows the parts of the perimeter of the isovist that are not blocked by obstacles. Specifically, it is the length of the “open” edges.
Visibility Analysis - High Values
Isovist Occlusivity combined with landmarks (Points with cyan color).
Visibility analysis, Isovist Max Radial
The Isovist Max Radial detects the longest line of sight from a generating point. These points are essential positions in the city configuration. Wiener states in studies about the virtual environment that participants chose the option that features the longest line of sight. (Wiener, et al., 2011).
Visibility Analysis - High Values
These placements can be for objects which the player can locate from a distance. For example, in Pokémon Go, these placements can be PokéStop locations, that you can discover from a visual distance.
Isovist Max combined with landmarks (Points with cyan color).
References
Al-Sayed, K., Turner, A., & Hillier, B. (2013). Space Syntax Methodology, The Bartlett School of Graduate Studies, University College London, UK.
Ejsing-Duun, S. (2011) Location-Based Games: From Screen to Street, PhD Dissertation, Aarhus University.
Hillier, B., Vaughan, L (2007) The city as one thing, Progress in Planning, 67 (3), 205-230.
Juhász, L., Hochmair, H., H. (2017) Where to catch ‘em all? – a geographic analysis of Pokémon Go locations, Geo-spatial information science, vol. 20, no. 3, 241–251
Moore, K. (2015) “Playing with portals: rethinking urban play with ingress”. In: UPR: http://analoggamestudies.org/2015/11/playing-with-portals-rethinking-urban-play-with-ingress/#note-3 (visited on 02/09/2019).
Koutsolampros, P., Sailer, K., Varoudis, T., and Haslem, R. (2019). “Dissecting visibility graph analysis: the metrics and their role in understanding workplace human behaviour”. In: Proceedings of the 12th International Space Syntax Symposium. 12th International Space Syntax Symposium. Beijing.
Spallazzo, D., Mariani, I. (2018) Location-Based Mobile Games: Design Perspectives, Cham, Switzerland: Springer.
Summers, A. (2015) Cultural cognitive differences in the spatial design of three-dimensional game environments, In: Proceedings of the 10th International Space Syntax Symposium. 10th International Space Syntax Symposium. London.