import numpy as np


def solve(area):
    if area.shape[0] <= 1:
        return 0
    count = solve_max(area)
    count += solve_min(area) - 1
    max_ind = np.argmax(area)
    min_ind = np.argmin(area)
    left_ind = min_ind if min_ind < max_ind else max_ind
    right_ind = min_ind if min_ind > max_ind else max_ind
    count += solve(area[0:left_ind])
    count += solve(area[left_ind+1: right_ind])
    count += solve(area[right_ind+1:area.shape[0]])
    return count


N = int(input())
input_np = np.array(input().split()).astype(np.int32)

print(solve(input_np))