結果:
This just came out. @Michelle Hirsch spoke to Jousef Murad and answer his questions about the big change in the desktop in R2025a and explained what was going on behind the scene. Enjoy!
The Big MATLAB Update: Dark Mode, Cloud & the Future of Engineering - Michelle Hirsch
Large Language Models (LLMs) with MATLAB was updated again today to support the newly released OpenAI models GPT-5, GPT-5 mini, GPT-5 nano, GPT-5 chat, o3, and o4-mini. When you create an openAIChat object, set the ModelName name-value argument to "gpt-5", "gpt-5-mini", "gpt-5-nano", "gpt-5-chat-latest", "o4-mini", or "o3".
This is version 4.4.0 of this free MATLAB add-on that lets you interact with LLMs on MATLAB. The release notes are at Release v4.4.0: Support for GPT-5, o3, o4-mini · matlab-deep-learning/llms-with-matlab
Share your ideas, suggestions, and wishlists for improving MathWorks products. What would make the software absolutely perfect for you? Discuss your idea(s) with other community users.
Guidelines & Tips
We encourage all ideas, big or small! To help everyone understand and discuss your suggestion, please include as much detail as possible in your post:
- Product or Feature: Clearly state which product (e.g., MATLAB, Simulink, a toolbox, etc.) or specific feature your idea relates to.
- The Problem or Opportunity: Briefly describe what challenge you’re facing or what opportunity you see for improvement.
- Your Idea: Explain your suggestion in detail. What would you like to see added, changed, or improved? How would it help you and other users?
- Examples or Use Cases (optional): If possible, include an example, scenario, or workflow to illustrate your idea.
- Related Posts (optional): If you’ve seen similar ideas or discussions, feel free to link to them for context.
Ready to share your idea?
Click here and then "Start a Discussion”, and let the community know how MATLAB could be even better for you!
Thank you for your contributions and for helping make MATLAB Central a vibrant place for sharing and improving ideas.
Independent researcher: Nguyễn Khánh Tùng
ORCID: 0009-0002-9877-4137
Email: traiphieu.com@gmail.com
Website: https://traiphieu.com
The NKTg Law (Law of Variable Inertia) not only holds value in physics but also opens up wide possibilities for applications in programming and simulation. The remarkable point here is that the same law, the same formula, can be implemented across a wide range of different programming languages.
In the content below, you will find a collection of 150 code snippets, each corresponding to one of the world’s leading programming languages:
Python, C++, Java, C, C#, JavaScript, TypeScript, PHP, Ruby, Swift, Go, Rust, Kotlin, Dart, Scala, R, MATLAB, Julia, Haskell, Perl, Shell, SQL, Visual Basic, Assembly, Ada, Fortran, Prolog, Scheme, Lisp, Scratch, Smalltalk, Pascal, Groovy, PowerShell, Apex, ABAP, ActionScript, Algol, Alice, AmbientTalk, AngelScript, APL, Arc, Arduino, ASP.NET, AssemblyScript, ATS, AWK, Ballerina, BASIC, VHDL, Verilog, Assembly, AutoHotkey, AutoLISP, AWK, Bash, bc, Boo, Clojure, COBOL, Common Lisp, Crystal, D, Delphi/Object Pascal, Dylan, Eiffel, Elixir, Elm, Emacs Lisp, Erlang, F#, Factor, Falcon, Fantom, Felix, Forth, Fortress, Frink, Gambas, GAMS, GAP, Genie, GLSL, Hack, Haxe, HDL, HLSL, Hope, HTML, HyperTalk, Icon, IDL, Inform, Io, Ioke, J, J#, JScript, JavaFX Script, Io, Ioke, J, J#, JScript, Julia, Kotlin, LabVIEW, Ladder Logic, Lasso, Lava, Lisp, LiveCode, Logo, Lua, M4, Magik, Maple, Mathematica, MATLAB, Mercury, Modula-2, Modula-3, MoonScript, Nemerle, NetLogo, Nim, Nix, Objective-C, Objective-J, OCaml, OpenCL, OpenEdge ABL, Oz, PL/I, PL/SQL, PostScript, Promela, Pure, Q#, Racket, RAPID, REBOL, Red, Rexx, Ring, Solidity, SPARK, SPSS, Squirre
All the code snippets illustrate how to calculate the fundamental quantities of The NKTg Law on Varying Inertia:
The movement tendency of an object in space depends on the relationship between its position, velocity, and mass.
NKTg = f(x, v, m)
In which:
- x is the position or displacement of the object relative to the reference point.
- v is the velocity.
- m is the mass.
The movement tendency of the object is determined by the following basic product quantities:
NKTg₁ = x × p
NKTg₂ = (dm/dt) × p
In which:
- p is the linear momentum, calculated by p = m × v.
- dm/dt is the rate of mass change over time.
- NKTg₁ is the quantity representing the product of position and momentum.
- NKTg₂ is the quantity representing the product of mass variation and momentum.
- The unit of measurement is NKTm, representing a unit of varying inertia.
The sign and value of the two quantities NKTg₁ and NKTg₂ determine the movement tendency:
- If NKTg₁ is positive, the object tends to move away from the stable state.
- If NKTg₁ is negative, the object tends to move toward the stable state.
- If NKTg₂ is positive, the mass variation has a supporting effect on the movement.
- If NKTg₂ is negative, the mass variation has a resisting effect on the movement.
The stable state in this law is understood as the state in which the position (x), velocity (v), and mass (m) of the object interact with each other to maintain the movement structure, helping the object avoid losing control and preserving its inherent movement pattern.
# Python:
versatile, easy to learn, strong for AI and data science
x, v, m, dm_dt = 2.0, 3.0, 5.0, 0.1
p = m * v
NKTg1 = x * p
NKTg2 = dm_dt * p
print(f"p={p}, NKTg1={NKTg1}, NKTg2={NKTg2}")
Java
// Java: enterprise applications, Android
public class NKTgLaw {
public static void main(String[] args) {
double x=2, v=3, m=5, dm_dt=0.1;
double p = m*v, NKTg1 = x*p, NKTg2 = dm_dt*p;
System.out.printf(
"p=%.2f NKTg1=%.2f NKTg2=%.2f%n", p, NKTg1, NKTg2);
}
}
Implementing the same law across 150 programming ecosystems demonstrates its universality and flexibility, while also confirming that any language—whether general-purpose and popular, or specialized and classical—can apply the NKTg Law to simulate, analyze, and handle practical problems.
Full list of 150 programming languages (complete) — due to post size limits I placed the complete list on an external page for easy viewing and download:
You can refer to the following four related articles to gain a deeper understanding of the NKTg Law and its applications
Variable Inertia From Abstract Concept to the Era of Measuring the ...
Inertia: A Millennia-Long Journey Inertia, the ability to maintain an object’s motion, has existed since ancient times. Aristotle viewed motion as an inherent property; an object stops only when t...
These got released last week and the process for using them on your local machine with MATLAB is very similar to how you use the local deepseek models as I demonstrated in my February blog post How to run local DeepSeek models and use them with MATLAB » The MATLAB Blog - MATLAB & Simulink
You need Ollama and the LLMs with MATLAB package installed (Details on how to do this in the blog post above). Then you run the following in your operating systems' command line
ollama pull gpt-oss:20b
Over to MATLAB and set up a chat session
>> chat = ollamaChat("gpt-oss:20b")
chat =
ollamaChat with properties:
ModelName: "gpt-oss:20b"
Endpoint: "127.0.0.1:11434"
TopK: Inf
MinP: 0
TailFreeSamplingZ: 1
Temperature: 1
TopP: 1
StopSequences: [0×0 string]
TimeOut: 120
SystemPrompt: []
ResponseFormat: "text"
FunctionNames: []
txt = generate(chat,"Who are you?")
txt =
"I’m ChatGPT – a conversational AI developed by OpenAI. My core is the GPT‑4 language model, which has been trained on a massive mix of text from books, websites, articles and other sources to understand and generate human‑like language. I don’t have feelings, consciousness, or a personal identity; I’m a tool that can help answer questions, brainstorm ideas, explain concepts, draft text, and more. My goal is to understand the context you give me and respond in a helpful, accurate and safe way. If there’s something specific you’d like to know or do, just let me know!"
This is the smaller of the two, new open models and it is bringing my aging desktop to its knees. My GPU is too small to do the work so I think everything is happening on the CPU and its slooooow. Will try on my Mac next
Let me know if you try this out!
作ったコードは公開して使ってもらいましょう！ということでその方法をブログで紹介します。
GitHub や File Exchange で公開しているコードがあれば、ぜひこのスレで教えてください！
ブログで紹介している大まかな３ステップをここにまとめます。
1. GitHub でコードを公開・開発する
- GitHub 上でのリポジトリ公開はコミュニティ形成にもつながります。
- R2025a 以降は MATLAB の Markdown サポートも強化されており、README.md を充実させると理解や導入が促進されます。
2. File Exchange に展開（GitHub と連携して自動同期）
- File Exchangeで公開することで MATLAB 内から検索・インストールが可能になります。
- GitHub と File Exchange の連携設定により、GitHub の更新を自動的に File Exchange に反映させることも可能です。
3. 「Open in MATLAB Online」ボタンやリンクを追加
- GitHub リポジトリに「Open in MATLAB Online」リンクやボタンを埋め込むことで、ブラウザ上でコードを試せます。
Long before I joined MathWorks, I was a member of the academic Research Software Engineering (RSE) community where part of my mission was to introduce basic software engineering concepts to the research community. Things like version control, testing and even simply writing code instead of using only pointy-clicky GUIs before copying and pasting the results plot into a word document. I've seen things..........*shudders*
The RSE movement is still going very strong and I am elated that MathWorks is increasingly interacting with it. One example of such interaction is a video tutorial contributed by my colleauge @Mihaela Jarema to a comminity seminar series called 'A summer of Testing' It's linked to below
The video assumes you've never run a test before and gently guides you through the principles. Along the way you'll learn about some of MATLAB's superb testing capabilities. Things like
- Unit testing Framework
- Test Browser App
- Code Coverage
- Test Fixtures (Setup and teardown)
- Test driven devellopment
- Function argument validation
- CI/CD using GitHub actions
Go check out out.
Resharing a fun short video explaining what MATLAB is. :)
I just wanted here to share a link to some .gif animations I created over the years with Matlab :-)
I think gif animations are great supports for scientific diffusion.
Hello to all!
I would like to share with the Matlab and Simulink community this video about Neural Networks in Simulink.
This is a series of videos that use a multilayer perceptron implemented in Simulink as a case study. Why Simulink? Because it's a visual and intuitive modeling tool, you can see the forward propagation of this network and better understand the flow. The objective of this series is to show the implementation using Simulink for both simulation and Arduino, as well as its training using Matlab and Matlab with Deep Learning Toolbox, and a video of training with Python.
The video is in Spanish, but the Simulink model is available in English for the entire community; subtitles are also available.
The files are located in the first comment of each video. We hope you find it interesting and enjoyable. Best regards!
Here I share the link to the first video.
In many parts of Africa, particularly in technical universities and engineering institutes, physical laboratories are scarce or poorly equipped. This reality deeply limits the hands-on experience students deserve, especially in fields like control systems, signal processing, power electronics, and fluid mechanics.
But MATLAB and Simulink can fill part of this gap.
As an educator and researcher, I’ve made it my mission to promote MATLAB as a didactic simulation environment that brings real-world experimentation into the virtual space—affordable, accessible, and scalable. Whether simulating dynamic systems, visualizing electromagnetic fields, or tuning PID controllers interactively, students can develop strong intuition without needing costly hardware.
🔧 I’ve used MATLAB to teach:
- Power systems and control theory without needing real generators or oscilloscopes,
- Hydrology and environmental modeling without field sensors,
- Robotics and AI concepts even where no robot is available.
🌍 This is more than a tool for me. It’s a bridge between educational ambition and limited infrastructure.
I dream of creating MATLAB-based virtual laboratories across African institutions. And I know I’m not alone.
Is anyone else here working on similar goals in under-resourced regions? Let’s connect and make it real.
— Patrick K.N.
As someone who grew up programming in C#, I often find myself wishing for a tighter, more native integration between MATLAB and C#.
There’s so much I dream of doing—leveraging the power of Simulink models or MATLAB’s advanced numerical libraries inside my .NET desktop or web applications. Of course, I know there are some workarounds: COM automation, MATLAB Engine API for .NET, or using MATLAB Compiler SDK… but let’s be honest: it’s not quite as seamless as I’d hope.
I imagine a world where:
- I could directly call MATLAB functions from C# as if they were .NET assemblies, without middleware.
- Simulink blocks could generate portable C# code (not just C/C++).
- MATLAB UI components could be embedded in WPF/WinForms apps natively.
Until then... we make do with what we have. But the vision remains.
Anyone else here trying to bridge MATLAB and C# in their workflow? I’d love to hear your experiences or ideas!
— Patrick K.N.
I found some beautiful computational art made by a developer called @yuruyurau who used a language called Processing. Unfortunately, I know very little about this language so I asked Claude to convert it to MATLAB for me.
Give it a try yourself and show me what you come up with.
Details here: Pair programming with Claude to produce computational art in MATLAB » The MATLAB Blog - MATLAB & Simulink
I have started a blog series on the history of image display in MATLAB. If this topic interests you, and if there is something in particular you would like me to address in the series, let me know.
t = turtle(); % Start a turtle
t.forward(100); % Move forward by 100
t.backward(100); % Move backward by 100
t.left(90); % Turn left by 90 degrees
t.right(90); % Tur right by 90 degrees
t.goto(100, 100); % Move to (100, 100)
t.turnto(90); % Turn to 90 degrees, i.e. north
t.speed(1000); % Set turtle speed as 1000 (default: 500)
t.pen_up(); % Pen up. Turtle leaves no trace.
t.pen_down(); % Pen down. Turtle leaves a trace again.
t.color('b'); % Change line color to 'b'
t.begin_fill(FaceColor, EdgeColor, FaceAlpha); % Start filling
t.end_fill(); % End filling
t.change_icon('person.png'); % Change the icon to 'person.png'
t.clear(); % Clear the Axes
classdef turtle < handle
properties (GetAccess = public, SetAccess = private)
x = 0
y = 0
q = 0
end
properties (SetAccess = public)
speed (1, 1) double = 500
end
properties (GetAccess = private)
speed_reg = 100
n_steps = 20
ax
l
ht
im
is_pen_up = false
is_filling = false
fill_color
fill_alpha
end
methods
function obj = turtle()
figure(Name='MATurtle', NumberTitle='off')
obj.ax = axes(box="on");
hold on,
obj.ht = hgtransform();
icon = flipud(imread('turtle.png'));
obj.im = imagesc(obj.ht, icon, ...
XData=[-30, 30], YData=[-30, 30], ...
AlphaData=(255 - double(rgb2gray(icon)))/255);
obj.l = plot(obj.x, obj.y, 'k');
obj.ax.XLim = [-500, 500];
obj.ax.YLim = [-500, 500];
obj.ax.DataAspectRatio = [1, 1, 1];
obj.ax.Toolbar.Visible = 'off';
disableDefaultInteractivity(obj.ax);
end
function home(obj)
obj.x = 0;
obj.y = 0;
obj.ht.Matrix = eye(4);
end
function forward(obj, dist)
obj.step(dist);
end
function backward(obj, dist)
obj.step(-dist)
end
function step(obj, delta)
if numel(delta) == 1
delta = delta*[cosd(obj.q), sind(obj.q)];
end
if obj.is_filling
obj.fill(delta);
else
obj.move(delta);
end
end
function goto(obj, x, y)
dx = x - obj.x;
dy = y - obj.y;
obj.turnto(rad2deg(atan2(dy, dx)));
obj.step([dx, dy]);
end
function left(obj, q)
obj.turn(q);
end
function right(obj, q)
obj.turn(-q);
end
function turnto(obj, q)
obj.turn(obj.wrap_angle(q - obj.q, -180));
end
function pen_up(obj)
if obj.is_filling
warning('not available while filling')
return
end
obj.is_pen_up = true;
end
function pen_down(obj, go)
if obj.is_pen_up
if nargin == 1
obj.l(end+1) = plot(obj.x, obj.y, Color=obj.l(end).Color);
else
obj.l(end+1) = go;
end
uistack(obj.ht, 'top')
end
obj.is_pen_up = false;
end
function color(obj, line_color)
if obj.is_filling
warning('not available while filling')
return
end
obj.pen_up();
obj.pen_down(plot(obj.x, obj.y, Color=line_color));
end
function begin_fill(obj, FaceColor, EdgeColor, FaceAlpha)
arguments
obj
FaceColor = [.6, .9, .6];
EdgeColor = [0 0.4470 0.7410];
FaceAlpha = 1;
end
if obj.is_filling
warning('already filling')
return
end
obj.fill_color = FaceColor;
obj.fill_alpha = FaceAlpha;
obj.pen_up();
obj.pen_down(patch(obj.x, obj.y, [1, 1, 1], ...
EdgeColor=EdgeColor, FaceAlpha=0));
obj.is_filling = true;
end
function end_fill(obj)
if ~obj.is_filling
warning('not filling now')
return
end
obj.l(end).FaceColor = obj.fill_color;
obj.l(end).FaceAlpha = obj.fill_alpha;
obj.is_filling = false;
end
function change_icon(obj, filename)
icon = flipud(imread(filename));
obj.im.CData = icon;
obj.im.AlphaData = (255 - double(rgb2gray(icon)))/255;
end
function clear(obj)
obj.x = 0;
obj.y = 0;
delete(obj.ax.Children(2:end));
obj.l = plot(0, 0, 'k');
obj.ht.Matrix = eye(4);
end
end
methods (Access = private)
function animated_step(obj, delta, q, initFcn, updateFcn)
arguments
obj
delta
q
initFcn = @() []
updateFcn = @(~, ~) []
end
dx = delta(1)/obj.n_steps;
dy = delta(2)/obj.n_steps;
dq = q/obj.n_steps;
pause_duration = norm(delta)/obj.speed/obj.speed_reg;
initFcn();
for i = 1:obj.n_steps
updateFcn(dx, dy);
obj.ht.Matrix = makehgtform(...
translate=[obj.x + dx*i, obj.y + dy*i, 0], ...
zrotate=deg2rad(obj.q + dq*i));
pause(pause_duration)
drawnow limitrate
end
obj.x = obj.x + delta(1);
obj.y = obj.y + delta(2);
end
function obj = turn(obj, q)
obj.animated_step([0, 0], q);
obj.q = obj.wrap_angle(obj.q + q, 0);
end
function move(obj, delta)
initFcn = @() [];
updateFcn = @(dx, dy) [];
if ~obj.is_pen_up
initFcn = @() initializeLine();
updateFcn = @(dx, dy) obj.update_end_point(obj.l(end), dx, dy);
end
function initializeLine()
obj.l(end).XData(end+1) = obj.l(end).XData(end);
obj.l(end).YData(end+1) = obj.l(end).YData(end);
end
obj.animated_step(delta, 0, initFcn, updateFcn);
end
function obj = fill(obj, delta)
initFcn = @() initializePatch();
updateFcn = @(dx, dy) obj.update_end_point(obj.l(end), dx, dy);
function initializePatch()
obj.l(end).Vertices(end+1, :) = obj.l(end).Vertices(end, :);
obj.l(end).Faces = 1:size(obj.l(end).Vertices, 1);
end
obj.animated_step(delta, 0, initFcn, updateFcn);
end
end
methods (Static, Access = private)
function update_end_point(l, dx, dy)
l.XData(end) = l.XData(end) + dx;
l.YData(end) = l.YData(end) + dy;
end
function q = wrap_angle(q, min_angle)
q = mod(q - min_angle, 360) + min_angle;
end
end
end
Nice to have - function output argument provide code assist when said function is called
This is a feature which doesn't apear to currently exist, but I think alot of matlab users would like, particularly ones who write alot of custom classes.
Imagine i have a custom class with some properties:
classdef CustomClass < handle
properties
name (1,1) string = "default name"
varOne (1,1) double = 0
end
methods
function obj = CustomClass(name,varOne)
obj.name = name;
obj.VarOne = varOne;
end
end
end
Then imagine I have a function which returns one of these custom class objects:
function [obj] = Calculation(Var1,Var2,name)
arguments (Input)
Var1 (1,1) double
Var2 (1,1) double
end
arguments (Output)
obj (1,1) CustomClass
end
results = Var1 + Var2;
obj = CustomClass(name,result);
end
With this class and this function which returns one of these class objects, I would like the fact that I provided "(1,1) CustomClass" in the output arguemnts block of function "Calculation(Var1,Var2,name)" to trigger code assist automaticaly show me, when writing code that the retuned value from this funciton has properties "name" and "varOne" in the object.
For istance, if I write the following code with this function and the class in the Matlab search path
testObj = Calculation(1,1,"test");
testObj.varOne = 10; %the property "varOne" doesn't apear in code assist when writing this line of code
I would like that the fact function "Calcuation(Var1,Var2,name) has the output arguments block enforcing that this function must return an object of "CustomClass" to make code assist recognise that "testObj" is a "CustomClass" object, just as if testObj was an input argument to another function which had an input argument requiring that "testObj" was a "CustomClass" object.
Maybe this is a feature that may be added to matlab in future releases? (please and thank you LOL)
This is a feature which doesn't apear to currently exist, but I think alot of matlab users would like, particularly ones who write alot of custom classes.
Imagine i have a custom class with some properties:
classdef CustomClass < handle
properties
name (1,1) string = "default name"
varOne (1,1) double = 0
end
methods
function obj = CustomClass(name,varOne)
obj.name = name;
obj.VarOne = varOne;
end
end
end
Then imagine I have a function which returns one of these custom class objects:
function [obj] = Calculation(Var1,Var2,name)
arguments (Input)
Var1 (1,1) double
Var2 (1,1) double
end
arguments (Output)
obj (1,1) CustomClass
end
results = Var1 + Var2;
obj = CustomClass(name,result);
end
With this class and this function which returns one of these class objects, I would like the fact that I provided "(1,1) CustomClass" in the output arguemnts block of function "Calculation(Var1,Var2,name)" to trigger code assist automaticaly show me, when writing code that the retuned value from this funciton has properties "name" and "varOne" in the object.
For istance, if I write the following code with this function and the class in the Matlab search path
testObj = Calculation(1,1,"test");
testObj.varOne = 10; %the property "varOne" doesn't apear in code assist when writing this line of code
I would like that the fact function "Calcuation(Var1,Var2,name) has the output arguments block enforcing that this function must return an object of "CustomClass" to make code assist recognise that "testObj" is a "CustomClass" object, just as if testObj was an input argument to another function which had an input argument requiring that "testObj" was a "CustomClass" object.
Maybe this is a feature that may be added to matlab in future releases? (please and thank you LOL)
I would like to zoom directly on the selected region when using 
on my image created with image or imagesc. First of all, I would recommend using image or imagesc and not imshow for this case, see comparison here: Differences between imshow() and image()? However when zooming Stretch-to-Fill behavior happens and I don't want that. Try range zoom to image generated by this code:
on my image created with image or imagesc. First of all, I would recommend using image or imagesc and not imshow for this case, see comparison here: Differences between imshow() and image()? However when zooming Stretch-to-Fill behavior happens and I don't want that. Try range zoom to image generated by this code:fig = uifigure;
ax = uiaxes(fig);
im = imread("peppers.png");
h = imagesc(im,"Parent",ax);
axis(ax,'tight', 'off')
I can fix that with manualy setting data aspect ratio:
daspect(ax,[1 1 1])
However, I need this code to run automatically after zooming. So I create zoom object and ActionPostCallback which is called everytime after I zoom, see zoom - ActionPostCallback.
z = zoom(ax);
z.ActionPostCallback = @(fig,ax) daspect(ax.Axes,[1 1 1]);
If you need, you can also create ActionPreCallback which is called everytime before I zoom, see zoom - ActionPreCallback.
z.ActionPreCallback = @(fig,ax) daspect(ax.Axes,'auto');
Code written and run in R2025a.
This week's Graphics and App Building blog article guides chart authors and app builders through the process of designing for a specific theme or creating theme-responsive charts and apps.
- Learn how dark theme may impacts charts and apps
- Discover best practices for theme-adaptive workflows
- Step-by-step examples for both script-based plots and advanced custom charts and UI components
- Discover new tools like ThemeChangedFcn, getTheme, and fliplightness
A Guide to Themes for Chart Creators and App Builders
The long-awaited dark theme is now available in MATLAB R2025a. This article serves as a guide for designing charts and apps in a specific theme or that adapt automatically to either theme. By default, the graphics theme matches the desktop theme, so charts and apps appear in dark or light theme depending on system settings. These preferences can be adjusted
Web サイトの選択
Web サイトを選択すると、翻訳されたコンテンツにアクセスし、地域のイベントやサービスを確認できます。現在の位置情報に基づき、次のサイトの選択を推奨します: United States
また、以下のリストから Web サイトを選択することもできます。
南北アメリカ
- América Latina (Español)
- Canada (English)
- United States (English)
ヨーロッパ
- Belgium (English)
- Denmark (English)
- Deutschland (Deutsch)
- España (Español)
- Finland (English)
- France (Français)
- Ireland (English)
- Italia (Italiano)
- Luxembourg (English)
- Netherlands (English)
- Norway (English)
- Österreich (Deutsch)
- Portugal (English)
- Sweden (English)
- Switzerland
- United Kingdom (English)