Skip to content

Changelog

Jump to bottom
Julius Bañgate edited this page Apr 17, 2023 · 78 revisions

GAMA 1.9.1 is out

The GAMA development team is pleased to announce the release of GAMA 1.9.1

This version, while maintaining the power, stability, expressiveness and ease of use of GAMA, brings new capabilities and openings to the platform, making it even more intuitive to use by modelers and even more versatile in terms of applications.

This major release of GAMA contains many new features and fixes, including:

  • A much more fluid and powerful IDE, offering support for all the latest technologies, from HiDPI displays to JDK 17 and Apple Silicon processors.
  • A new server mode of GAMA, offering a clear and extensible exchange protocol, which completely revolutionizes the way to interact with the platform from R, Python or any web client.
  • Increased model exploration possibilities thanks to new calibration and optimization methods, also directly usable in the server mode.
  • The addition of the two new data types field and image, which make it even easier to load, analyze, visualize and produce raster data
  • A much more powerful graph manipulation than previous versions, but still easy to couple with agents
  • A focus on urban mobility applications, with skill advanced_driving and pedestrian, which make it much easier to produce realistic large-scale models.
  • The possibility to simulate physical interactions between agents thanks to the new skills static_body, dynamic_body and physical_simulation, which rely on the native bullet library.
  • New and faster display capabilities, offering more intuitive handling of agents and organisation of display surfaces, making it easier than ever to build interactive simulations, serious games or advanced scientific visualisations.

Comparison chart

  Gama 1.8.1 Gama 1.9.1
Java and environments Java 11 and x86 (intel architecture) Java 17, x86 and ARM architectures (notably Apple Silicon)
Server mode - Headless server / connection with Python and R
Model exploration Exhaustive sampling and calibration Several new sampling methods (e.g. latinhypercube), sensitivity analysis (e.g. Sobol) and calibration
Physics modeling Limited Extended features with native bullet library and influences/forces computations
Mobility modeling moving and driving skills moving, advanced_driving (non normative traffic) and pedestrian (Social force model) skills
Raster data integration Limited with grid (bad performances above 500x500) New field and image types allow larger sizes and better performances
Graph integration Programmatic with fixed layout Import / export to 6 graph file formats (e.g. .graphml, .gml) with various spatial layouts rendering

Major changes from 1.8.1 to 1.9.1

GAMA Server mode

gama-server is a new way of running GAMA experiments. It consists of an instance of gama-headless that, once launched, waits for commands sent through websockets and executes them. These commands follow a clear and extensible protocol, enabling its use in many contexts, from the definition of experiment plans in R to the design of dashboards in JavaScript. See the corresponding wiki page to setup a server instance of Gama.

Modelling improvements

field type

A new variable type (field) to support the management (import and use) of large raster geographic data. It allows in particular to:

  • import large mono/multi-band rasters
  • simply access / modify values of spatial grids as simply as before, but with very high performance improvement

Try out:

image type

  • Easier to work with images

Try out:

pedestrian skill

A new plugin has been integrated in GAMA that allows to simulate pedestrian movement. This plugin uses Helbing's social force model as a basis to support pedestrian walk and offers tools to reconstruct paths from an open environment and obstacles. This two new features are identified by a skill (pedestrian) and an operator (generate_pedestrian_network) respectively. You can find examples in the models below.

Try out:

advanced_driving skill

The driving skill has been completely redesigned in order to offer a more realistic representation of driver behavior (by explicitly using the Intelligent Driver Model and Lane-change Model MOBIL) and by allowing to take into account multi-lane vehicles - this allows for example to simulate mixed traffic composed of motorcycles and cars. Besides, the behavior of drivers can be custom to represent non normative behavior, such as dangerous take-off, disrespect of signals, signs, speed limit or road direction and lanes.

Try out:

  • An abstract representation of vehicles size (bus, car, motorcycle) and free use of road lanes and direction (Drive Random.gaml)
  • An abstract representation of vehicles managing cross section, with collision avoidance, priority, etc. (Simple Intersection.gaml)
  • A very small road system with stops to simulate congestion (Following Paths.gaml)

Physics extension improvement

Physics plugin has been completely rewritten and allows to use native implementations of the bullet library in a redesigned framework (where physical agents can coexist with non-physical ones).

Try out:

  • Interaction between static (skill static_body) and dynamic (skill dynamic_body) 3D objects (Eroding Vulcano.gaml)
  • Manage 3D objects movement based on a Digital Elevation Model (Flow on Terrain.gaml)

Experiment

Batch methods

Batch experiments have been rework to better distinguish simulation exploration and model calibration. On the first hand, modelers should engage in simulation exploration if they want to launch many simulations across the parameter space, better understand the contribution of stochasticity and evaluate the specific contribution of given parameters to output variability. On the other hand, modelers should use calibration methods if they want to find parameters values of the models, so the simulation outputs are as close as possible to desired ones. A detailed description is provided in this wiki page.

Try out:

  • A walkthrough of all provided methods to explore, method exploration, and analyse the sensitivity of your model, including a tool to decide method stochanalyse or method sobol (Exploration.gaml)
  • A walkthrough of minimal way to setup calibration, including the new PSO algorithm (Calibration.gaml)

Headless batch

We implement a way to launch Gama batch experiment in headless with a simple command line, using the gama-headless.sh bash script with -batch option. For more information, see the related (wiki page).

Reproducibility and random number generation

  • Great effort towards tracking and limiting the use of random generators outside the ones built in GAMA
  • Addition of several new random number generators

Displays

OpenGL improvements

Great improvements have been done on the displays and specifically on opengl ones. Key points are:

  • Lot faster (2 times) on geometries
  • Rendering of large-scale images, grids, fields or matrices using the new mesh layer, with several colouring options
  • More flexibility:
  • camera statement to specify the dynamic movements of the camera
  • light statements to specify the lighting(s) of the scene
  • rotate statement to specify the rotation of the full screen
  • Better and more accurate rendering of texts (with 3D, etc.)
  • Possibility to choose between several predefined cameras, to save cameras, etc.

mesh layer

  • display large rasters

layout improvements

  • Allow to easily split or compose the displays
  • Possibility to define borderless displays

User Interface

Support of HiDPI

  • HiDPI and various "display zooms" are now supported natively. Displays, text and icons scale up and down accordingly. Only issues remaining is that the text and icons can be blurry and pixelised on some configurations (Windows 10, Windows 11 with 150% zoom, etc.)

Support of dark mode

  • Light and dark modes are also now supported out of the box. Preferences allow GAMA to impose its own theme or follow the one defined in the OS. A new syntax highlighting theme for dark mode is accessible from the preferences too.

User Interaction

Addition of wizards and dialogs

Try out:

Addition of events

  • new events can be defined as display layers: #arrow_down, #arrow_up, #arrow_left, #arrow_right, #escape, #tab, #enter, #page_up, #page_down

Clipboard

  • the clipboard can be written and read using the copy_to_clipboard(value) and copy_from_clipboard(type) operators

Advanced programming usages

Additions to GAML

  • on_change: facet can be added to attributes and parameters to trigger any behaviour in response to a change of value. Particularly useful for defining interactive parameters.
  • abort statement can be defined in any agent (incl. global and experiment) and executed just before the agent is disposed of.

thread skill

The new thread skill allows to run actions in a specific thread. In particular, this skill is intended to define the minimal set of behaviours required for agents that are able to run an action in a thread.

File manipulations: copy, zip, delete, save improvements

  • One can now completely manipulate files directly in the gama models with dedicated copy_file, delete_file, rename_file(which can be used to move a file), zip and unzip operators.
  • save accepts more file formats and provides a hook for developers to develop ISaveDelegates

network skill improvements

To increase the integration between Gama and other applications we improved a lot the network capabilities:

  • The communication with web-services is now easier with the possibility to execute post/get/update/delete HTTP requests directly in gaml with extensions of the send action of the networking skill, as described in the HTTP POST.gaml and HTTP GET.gaml of the Plugin models library.
  • Adding support for the websocket protocol in the network skill
  • General work on the network skill with communication outside of Gama in mind

Graph improvements

Shortest paths

Integration of new algorithms for computing shortest paths in graphs.

  • BidirectionalDijkstra: default one - ensure to find the best shortest path - compute one shortest path at a time: https://www.homepages.ucl.ac.uk/~ucahmto/math/2020/05/30/bidirectional-dijkstra.html
  • DeltaStepping: ensure to find the best shortest path - compute one shortest path at a time: The delta-stepping algorithm is described in the paper: U. Meyer, P. Sanders, $\Delta$-stepping: a parallelizable shortest path algorithm, Journal of Algorithms, Volume 49, Issue 1, 2003, Pages 114-152, ISSN 0196-6774
  • CHBidirectionalDijkstra: ensure to find the best shortest path - compute one shortest path at a time. Based on precomputations (first call of the algorithm). Implementation of the hierarchical query algorithm based on the bidirectional Dijkstra search. The query algorithm is originally described the article: Robert Geisberger, Peter Sanders, Dominik Schultes, and Daniel Delling. 2008. Contraction hierarchies: faster and simpler hierarchical routing in road networks. In Proceedings of the 7th international conference on Experimental algorithms (WEA'08), Catherine C. McGeoch (Ed.). Springer-Verlag, Berlin, Heidelberg, 319-333
  • TransitNodeRouting: ensure to find the best shortest path - compute one shortest path at a time. Based on precomputations (first call of the algorithm). The algorithm is designed to operate on sparse graphs with low average outdegree. the algorithm is originally described the article: Arz, Julian & Luxen, Dennis & Sanders, Peter. (2013). Transit Node Routing Reconsidered. 7933. 10.1007/978-3-642-38527-8_7.

Input/ouput

You can now load / save your graph into dedicated file format such as .gml, .dot or .gefx to build your graph.

Try out:

Layout

Non spatial graph can be rendered using operators to locate nodes on a circle, as a grid lattice or considering connection as forces.

OS and computing environments

GAMA 1.9.1 has been tested on:

  • Windows 10 and 11 on Intel processors
  • MacOS Monterey, Ventura on Intel & Apple Silicon computers
  • Ubuntu 20.04 and 22.04 on Intel processors

Note that this version drops the support for 32 bits architectures.

Support of JDK 17+

Gama 1.9.1 brings compatibility with JDK17+ and should remain compatible for the following JDK versions.

Support of ARM processors

A specific version of GAMA is now built for Apple Silicon processors on macOS. Even if no specific version is produced for the ARM version of Windows, reports show that it works well in emulated mode.

New installers for Windows, Mac (brew) and Linux (aur, deb)

Gama 1.9.1 comes with a dedicated installer for every platform, so it's easier for newcomers to get it working. In addition, the macOS version is now fully signed. Linux and macOS users can also benefit from CLI installers.

New versions of native libraries: SWT, JTS, GeoTools, bullet, JOGL, JGraphT

All the major libraries on which GAMA is relying have been bumped to their latest versions, except GeoTools (version 25) and JGraphT (version 1.5.1).

Changes that can impact models

🔴 Errors 🔴: concepts that cannot be used anymore

  • gama.pref_lib_r, gama.pref_lib_spatialite, gama.pref_optimize_agent_memory, gama.pref_display_triangulator have been removed
  • In experiment, the method statement exhaustive and explicit does not exist anymore. Use exploration instead, see the related documentation on batch.
  • the material type (and the corresponding material: facet in draw:) does not exist anymore and has not been replaced.
  • the built-in equation types (SIR, etc.) do not exist anymore and have not been replaced.
  • field cannot be used anymore as a species or variable name.
  • image cannot be used anymore as a species or variable name.
  • to_list cannot be used anymore as a species or variable name.

🔴 Errors 🔴: concepts that need to be written differently

  • timeStamp() in SQLSKILL does not exist anymore. Use machine_time instead.
  • dem(...) operators do not exist anymore. Use a combination of field and mesh layer to load and draw a digital elevation model
  • event ['k'] should be rewritten as event 'k'.
  • generate_complete_graph, generate_barabasi_albert, generate_watts_strogatz, and as_distance_graph now take different arguments. Please refer to their documentation.
  • load_graph_from_file has been removed and replaced by the use of the corresponding graph file types (graphml_file, etc.)
  • simplex_generator has been removed and replaced by generate_terrain

🟠 Warnings 🟠:

  • grid + lines: is deprecated and replaced by border:
  • save + type: is deprecated and replaced by format:
  • display + draw_env: is deprecated and replaced by axes:
  • display + synchronized: is deprecated. synchronized: should now be defined on output:
  • display + camera_pos: is deprecated. Should be replaced by location: defined on a camera statement inside the display
  • display + camera_interaction: is deprecated. Should be replaced by locked: defined on a camera statement inside the display
  • display + camera_up_vector: is deprecated. Not used anymore.
  • display + camera_look_pos: is deprecated. Should be replaced by target: defined on a camera statement inside the display
  • display + focus: is deprecated. Should be replaced by target: defined on a camera statement inside the display
  • display + ambient_light: is deprecated. Should be replaced by intensity: defined on a light #ambient statement inside the display
  • light + position: is deprecated and replaced by location:
  • light + update: is deprecated and replaced by dynamic:
  • light + color: is deprecated and replaced by intensity:
  • light + name: now takes a string and not an int
  • light + draw_light: is deprecated and replaced by show:
  • light + type: now takes a string among #spot, #point or #direction
  • user_input is deprecated and should be replaced by user_input_dialog
  • draw + empty: is deprecated and replaced by wireframe:
  • image (layer) + file: is deprecated and replaced by the direct use of the file name as the default facet
  • event now takes a string for its default facet (preferably the defined constants like #mouse_move, #left_arrow, etc.)
  • event + action: is deprecated as the definition of the action should directly follow the statement definition
  • the with_optimizer_type operator is deprecated and replaced by with_shortestpath_algorithm

Preferences

The description of all preferences can be found at this page. A number of new preferences have been added to cover existing or new aspects of the platform. They are summarised below.

New preferences

Interface tab

  • Startup Remember Gama windows sizes
  • Startup Several prompts related to the use of workspaces
  • Startup Setup a model to run at start

Editors tab

  • Edition More options (3) for automatic typing
  • Edition Turns experiment buttons into a drop down list
  • Syntax Coloring according to Gama theme (light|dark)

Execution tab

  • (New) Parameters Customize parameter view
  • Parallelism Use all available threads in batch mode

Display tab

  • Chart preferences Choose resolution of charts
  • (Removed) Advanced
  • OpenGL Limit the number of frames
  • OpenGL Sensitivity of keyboard/mouse/trackpad
  • OpenGL Ambiant light intensity
  • OpenGL Default camera orientation

Data and Operator

  • Random Number Generator Display RNG in parameter view
  • (Removed) Optimization Many options have been removed to enforce reproducibility

(New) Experimental

This tab holds experimental preferences that should be use with care

Setting and sharing preferences

Gama 1.9.1 brings new options for setting preferences and sharing them among models.

Passing preferences to GAMA at startup

Modellers running the headless or gui versions of GAMA can now pass preferences to the executable using arguments (either in the headless script or in the Gama.ini file). The syntax is -Dpref_name=value (for instance -Dpref_display_synchronized=true to synchronise displays, including snapshots of headless GAMA, with the simulation).

Global or workspace scopes

The default behaviour of GAMA makes sharing preferences between workspaces and models easy, since they are global to the user account. In some instances, however, it can be necessary to restrict them to a local scope (i.e. a workspace). In that case, launching GAMA with the -Duse_global_preference_store=false will make it save its preferences in the current workspace and not globally anymore.

Bug fixes

You can also check the complete list of the closed issues on the github repository. Keep in mind that this list is incomplete as a lot of problems where solved without being linked to any issue on github (via the mailing list or internally for example).

Added models

The library of models has undergone some changes. Besides making sure all the models compile and run fine under the new version of GAMA, it also brings some new models, which are listed below:

Usage of the pedestrian skill

New graph capabilities

Utilities

Elements of GAML syntax

New batch capabilities

Toy models

Declaration and usage of field

New user interaction modalities

New camera and light definitions

Physics engine demonstrations

New driving skill

New network capabilities

Usage of the image type

New mathematical tests

Home

  1. What's new (Changelog)

Platform

  1. Installation and Launching
    1. Installation
    2. Launching GAMA
    3. Updating GAMA
    4. Installing Plugins
  2. Workspace, Projects and Models
    1. Navigating in the Workspace
    2. Changing Workspace
    3. Importing Models
  3. Editing Models
    1. GAML Editor (Generalities)
    2. GAML Editor Tools
    3. Validation of Models
  4. Running Experiments
    1. Launching Experiments
    2. Experiments User interface
    3. Controls of experiments
    4. Parameters view
    5. Inspectors and monitors
    6. Displays
    7. Batch Specific UI
    8. Errors View
  5. Running Headless
    1. Headless Batch
    2. Headless Server
    3. Headless Legacy
  6. Preferences
  7. Troubleshooting

Learn GAML step by step

  1. Introduction
    1. Start with GAML
    2. Organization of a Model
    3. Basic programming concepts in GAML
  2. Manipulate basic Species
  3. Global Species
    1. Regular Species
    2. Defining Actions and Behaviors
    3. Interaction between Agents
    4. Attaching Skills
    5. Inheritance
  4. Defining Advanced Species
    1. Grid Species
    2. Graph Species
    3. Mirror Species
    4. Multi-Level Architecture
  5. Defining GUI Experiment
    1. Defining Parameters
    2. Defining Displays Generalities
    3. Defining 3D Displays
    4. Defining Charts
    5. Defining Monitors and Inspectors
    6. Defining Export files
    7. Defining User Interaction
  6. Exploring Models
    1. Run Several Simulations
    2. Batch Experiments
    3. Exploration Methods
  7. Optimizing Model Section
    1. Runtime Concepts
    2. Optimizing Models
  8. Multi-Paradigm Modeling
    1. Control Architecture
    2. Defining Differential Equations

Recipes

  1. Manipulate OSM Data
  2. Diffusion
  3. Using Database
  4. Using FIPA ACL
  5. Using BDI with BEN
  6. Using Driving Skill
  7. Manipulate dates
  8. Manipulate lights
  9. Using comodel
  10. Save and restore Simulations
  11. Using network
  12. Headless mode
  13. Using Headless
  14. Writing Unit Tests
  15. Ensure model's reproducibility
  16. Going further with extensions
    1. Calling R
    2. Using Graphical Editor
    3. Using Git from GAMA

GAML References

  1. Built-in Species
  2. Built-in Skills
  3. Built-in Architecture
  4. Statements
  5. Data Type
  6. File Type
  7. Expressions
    1. Literals
    2. Units and Constants
    3. Pseudo Variables
    4. Variables And Attributes
    5. Operators [A-A]
    6. Operators [B-C]
    7. Operators [D-H]
    8. Operators [I-M]
    9. Operators [N-R]
    10. Operators [S-Z]
  8. Exhaustive list of GAMA Keywords

Developing GAMA

  1. Installing the GIT version
  2. Developing Extensions
    1. Developing Plugins
    2. Developing Skills
    3. Developing Statements
    4. Developing Operators
    5. Developing Types
    6. Developing Species
    7. Developing Control Architectures
    8. Index of annotations
  3. Introduction to GAMA Java API
    1. Architecture of GAMA
    2. IScope
  4. Using GAMA flags
  5. Creating a release of GAMA
  6. Documentation generation

Tutorials

  1. Predator Prey
  2. Road Traffic
  3. 3D Tutorial
  4. Incremental Model
  5. Luneray's flu
  6. BDI Agents

Community

  1. Team
  2. Projects using GAMA
  3. Scientific References
  4. Training Sessions

Versions of GAMA

Resources

  1. Videos
  2. Conferences
  3. Code Examples
  4. Pedagogical materials
Clone this wiki locally