if (!.Call(`_rxode2_isIntel`)) {
  ## Example 3.1 from
  ## "Solving Differential Equations in R" by Soetaert et al (2012)
  ## https://cran.r-project.org/web/packages/diffEq/vignettes/ODEinR.pdf Example #1
  skip_if_not(file.exists(test_path("test-example-3-1.qs")))
  df.test <- qs::qread(test_path("test-example-3-1.qs"))

  ms <- c("liblsoda", "lsoda", "dop853")
  ode <- rxode2(model = "d/dt(y) = r * y * (1.0 - y/K);")
  .rxWithOptions(list(digits = 6), {
    for (meth in ms) {
      # context(sprintf("Example 3-1 (%s)", meth))

      ## create event table with times at which we observe the system
      et <- eventTable(time.units = NA)
      et$add.sampling(seq(from = 0, to = 20, by = 0.2))

      ## same model, different initial values (2 and 12)
      out1 <- ode$solve(params = c(r = 1, K = 10), events = et, inits = c(y = 2))
      out2 <- ode$solve(params = c(r = 1, K = 10), events = et, inits = c(y = 12))

      ## matplot(x = out1[,1], y = cbind(out1[,2], out2[,2]), type = "l",
      ##    main = "logistic growth", xlab="time", ylab="")

      ## Now use a non-stiff solver
      out2.ns <-
        ode$solve(params = c(r = 1, K = 10), events = et, inits = c(y = 12), method = "dop853")

      df <- round(cbind(out1, out2, out2.ns), 3)

      test_that("Runs example 3.1 correctly", {
        expect_equal(df, round(df.test, 3))
      })

      head(cbind(out1, out2, out2.ns), n = 15)
    }
  })
}