#include <cstdio>
#include <vector>
#include <numeric>
#include <algorithm>
typedef long long i64;
const int N = 100000 + 10;
const i64 INF = 1LL << 60;
int n, k, t, x[N];
auto calc(std::vector<i64> &cur) {
const int tot = cur.size();
std::vector<std::pair<i64, i64>> res;
for (int i = 0; i < tot;) {
i64 sum = 0, min = 0;
int j = i;
while (j < tot) {
min = std::min(min, sum += cur[j++]);
if (sum >= 0) break;
}
if (sum >= 0) res.emplace_back(min, sum);
if (j == tot) cur.erase(cur.begin(), cur.begin() + (sum < 0 ? i : j));
i = j;
}
return res;
}
bool check(std::vector<std::pair<i64, i64>> &u, std::vector<std::pair<i64, i64>> &v, i64 cur) {
std::reverse(u.begin(), u.end());
std::reverse(v.begin(), v.end());
while (!(u.empty() && v.empty())) {
if (!u.empty() && cur + u.back().first >= 0) {
cur += u.back().second;
u.pop_back();
} else if (!v.empty() && cur + v.back().first >= 0) {
cur += v.back().second;
v.pop_back();
} else {
return false;
}
}
return true;
}
bool solve(int s) {
std::vector<i64> a, b;
for (int i = k; i > 1; --i) a.push_back(2LL * t * s - (x[i] - x[i - 1]));
for (int i = k; i < n; ++i) b.push_back(2LL * t * s - (x[i + 1] - x[i]));
i64 temp = std::accumulate(a.begin(), a.end(), std::accumulate(b.begin(), b.end(), 0LL));
if (temp < 0) return false;
auto u = calc(a), v = calc(b);
if (!check(u, v, 0)) return false;
std::reverse(a.begin(), a.end());
std::reverse(b.begin(), b.end());
for (auto &i : a) i = -i;
for (auto &i : b) i = -i;
u = calc(a), v = calc(b);
return check(u, v, temp);
}
int main() {
scanf("%d%d%d", &n, &k, &t);
for (int i = 1; i <= n; ++i) scanf("%d", &x[i]);
int l = 0, r = x[n];
while (l < r) {
int mid = (l + r) >> 1;
if (solve(mid)) r = mid; else l = mid + 1;
}
printf("%d\n", l);
return 0;
}
