What's the difference of tiledlayout and subplot?

The function subplot can already make tiled layout. Why tiledlayout is introduced in 2019b?

[dpb]

Same reason Proctor & Gamble introduces new, improved versions of Tide detergent.

Answer 1

[Adam Danz]

tiledlayout has additional features not supported in subplot. Changing subplot would cause backward compatibility issues.

• Sean de Wolski's September 2019 blog post reviews some limitations to subplot and some new features available in tiledlayout.
• MathWorks posted a thread in reddit highlighting a tiledlayout feature that lets you add tiles without defining a layout matrix using the flow feature.
• A recent community highlight shows some tiledlayout spacing options that are not supported with subplot.
• Global legend, global titles, and global axis labels that span the entire grid of axes are easy to do with tiledlayout.

Demo

Two blocks below produce nearly the same figure using tiledlayout and subplot while using as few lines as possible without sacrificing best-practices.

A subplot/tile will be added for each column of x but the number of columns of data are unknown!

Produce data

rng(210606) % for reproducibility of this dataset
n = randi([4,9]); % number of subplots/tiles needed
x = randn(200,n) .* (1./10.^[0:n-1]);

tiledlayout

fig = figure(); 
tlo = tiledlayout(fig, 'flow');
arrayfun(@(col)histogram(nexttile(tlo),x(:,col)),1:n); % *see below
title(tlo,'Global title')
ylabel(tlo, 'Global ylabel')
xlabel(tlo, 'Global xlabel')

The arrayfun line is the same as this loop:

for i = 1:n

ax = nexttile(tlo);

histogram(ax, x(:,i));

end

subplot

The label placement took a long time to figure out relative to the tiledlayout example.

fig = figure(); 
nRows = ceil(sqrt(n));  % compute number of subplot rows
nCols = ceil(n/nRows);   % compute number of subplot columns
arrayfun(@(col)histogram(subplot(nRows,nCols,col),x(:,col)),1:n); % *see below
sgtitle('Global title')
annotation(fig,'textarrow',[.15 .15], [.52 .52], ... % <-- estimated margin
    'String','Global ylabel',...
    'TextRotation',90,...
    'FontSize',12,...
    'HorizontalAlignment','Center',...
    'VerticalAlignment','Middle',...
    'HeadStyle','none',...
    'LineStyle','none',...
    'TextMargin', .1);
annotation(fig,'textarrow',[.6 .6], [.03 .03], ... % <-- estimated margin
    'String','Global xlabel',...
    'FontSize',12,...
    'HorizontalAlignment','Center',...
    'VerticalAlignment','Middle',...
    'HeadStyle','none',...
    'LineStyle','none',...
    'TextMargin', .1);

The arrayfun line is the same as this loop:

for i = 1:n

ax = subplot(nRows,nCols,i);

histogram(ax, x(:,i));

end