#include <bits/stdc++.h>

int ri() {
	int n;
	scanf("%d", &n);
	return n;
}

std::vector<int> a, x, y;
std::vector<int64_t> dp;

int64_t cost(int i, int j) {
	int64_t x_dist = std::abs(a[j] - x[i]);
	int64_t y_dist = y[i];
	return x_dist + y_dist;
}

void monotone_minima(int x1, int x2, int y1, int y2) {
	int m = y1 + (y2 - y1) / 2;
	assert(x1 < m);
	int64_t min = 1000000000000000000;
	int ind;
	for (int i = x1; i < x2 && i <= m; i++) {
		int64_t c = (i ? dp[i - 1] : 0) + cost(i, m);
		if (min > c) min = c, ind = i;
	}
	dp[m] = std::min(dp[m], min);
	if (y1 < m) monotone_minima(x1, ind + 1, y1, m);
	if (m + 1 < y2) monotone_minima(ind, x2, m + 1, y2);
}

void solve(int l, int r) {
	if (r - l == 1) {
		dp[l] = std::min(dp[l], (l ? dp[l - 1] : 0) + cost(l, l));
	} else {
		int m = l + (r - l) / 2;
		solve(l, m);
		monotone_minima(l, m, m, r);
		solve(m, r);
	}
}

int main() {
	int n = ri();
	a.resize(n);
	x.resize(n);
	y.resize(n);
	for (auto &i : a) i = ri();
	for (auto &i : x) i = ri();
	for (auto &i : y) i = ri();
	dp.resize(n, 1000000000000000000);
	solve(0, n);
	std::cout << dp[n - 1] << std::endl;
	return 0;
}