#include <bits/stdc++.h>

using namespace std;

const int64_t INF = INT64_C(1) << 60;

int main() {
  ios::sync_with_stdio(false);
  cin.tie(0);
  int N;
  cin >> N;
  vector<int64_t> A(N), B(N);
  for (int i = 0; i < N; i++) {
    cin >> A[i] >> B[i];
  }
  vector<int64_t> X = {-INF, INF}, Y = {-INF, INF};
  int a = N / 2, b = N - a;
  for (int bit = 0; bit < (1 << a); bit++) {
    int64_t x = 0, y = 0;
    for (int i = 0; i < a; i++) {
      if ((bit >> i) & 1) {
        x += A[i];
      } else {
        y += B[i];
      }
    }
    if (x - min(x, y) == 0) Y.emplace_back(y - min(x, y));
    if (y - min(x, y) == 0) X.emplace_back(x - min(x, y));
  }
  sort(X.begin(), X.end());
  sort(Y.begin(), Y.end());
  int64_t res = INF;
  for (int bit = 0; bit < (1 << b); bit++) {
    int64_t x = 0, y = 0;
    for (int i = 0; i < b; i++) {
      if ((bit >> i) & 1) {
        x += A[i + a];
      } else {
        y += B[i + a];
      }
    }
    if (x - min(x, y) == 0) {
      int i = upper_bound(X.begin(), X.end(), y - min(x, y)) - X.begin();
      res = min(res, abs(X[i] - (y - min(x, y))));
      res = min(res, abs(X[i - 1] - (y - min(x, y))));
    }
    if (y - min(x, y) == 0) {
      int i = upper_bound(Y.begin(), Y.end(), x - min(x, y)) - Y.begin();
      res = min(res, abs(Y[i] - (x - min(x, y))));
      res = min(res, abs(Y[i - 1] - (x - min(x, y))));
    }
  }
  cout << res << '\n';
  return 0;
}