Graphic quality in the RStudio panel

If you’ve ever plotted a line chart in R and, when you saw the plot in the RStudio panel, thought “wow, what terrible quality!”, you’re not alone. But don’t worry, the solution is quite simple!

First, let’s plot a chart using RStudio’s default settings:

knitr::include_graphics("img/Rplot.png")

Figure 1: Plot without anti-aliasing.

Wow, imagine using something like that on a poster! Let’s try again, now with Cairo as the graphics device and using anti-aliasing:

# Adding anti-aliasing
trace(grDevices::png, quote({
  if (missing(type) && missing(antialias)) {
    type = "cairo-png"
    antialias = "subpixel"
  }
}), print = FALSE)

Tracing function "png" in package "grDevices"

[1] "png"

# Plotting again
plot(mtcars$mpg, type = "l")

Figure 2: Plot with anti-aliasing.

Holy anti-aliasing, right? (Zoom in on both to really see the difference)

Well, do I need to do this every time I plot a chart or start an R session? Not at all, just add these lines to your .Rprofile. I always recommend using {usethis} to edit configuration files, as you might get lost among the possible paths R will check. Using {usethis} ensures you’re creating or editing the correct file.

usethis::edit_r_profile() will open a window with the file for editing. Then, paste the call above (removing the plot line, of course), make sure your .Rprofile ends with a blank line (because R ignores the last line), save, and restart your session. Done! Now anti-aliasing will always be applied to your R plots, regardless of the package used, whether base or {ggplot}, for example.