#include <bits/stdc++.h>
using namespace std;
void just_do_it();
int main() {
#ifdef KAMIRULEZ
freopen("kamirulez.inp", "r", stdin);
freopen("kamirulez.out", "w", stdout);
#endif
ios_base::sync_with_stdio(0);
cin.tie(0);
just_do_it();
return 0;
}
const int maxK = 22;
namespace NTT {
const long long mod = (119 << 23) + 1;
long long bin_pow(long long x, long long y) {
long long res = 1;
while (y) {
if (y & 1) {
res = res * x % mod;
}
y >>= 1;
x = x * x % mod;
}
return res;
}
long long mod_inv(long long x) {
return bin_pow(x, mod - 2);
}
int maxL;
int maxN;
long long root;
void fft(long long A[], bool inv) {
for (int i = 0; i < maxN; i++) {
int rev = 0;
for (int j = 0; j < maxL; j++) {
rev |= ((i >> j) & 1) << (maxL - 1 - j);
}
if (i < rev) {
swap(A[i], A[rev]);
}
}
for (int len = 2; len <= maxN; len <<= 1) {
long long mul = (inv ? mod_inv(root) : root);
for (int tmp = len; tmp < maxN; tmp <<= 1) {
mul = mul * mul % mod;
}
for (int i = 0; i < maxN; i += len) {
long long cur = 1;
for (int j = 0; j < (len >> 1); j++) {
long long X = A[i + j];
long long Y = A[i + j + (len >> 1)];
A[i + j] = (X + cur * Y) % mod;
A[i + j + (len >> 1)] = (X + (mod - cur) * Y) % mod;
cur = cur * mul % mod;
}
}
}
if (inv) {
long long mul = mod_inv(maxN);
for (int i = 0; i < maxN; i++) {
A[i] = A[i] * mul % mod;
}
}
}
long long A[1 << maxK];
long long B[1 << maxK];
void mul(long long _A[], long long _B[], long long C[], int _maxL) {
maxL = _maxL;
maxN = (1 << maxL);
root = bin_pow(bin_pow(3, 119), 1 << (23 - maxL));
for (int i = 0; i < maxN; i++) {
A[i] = _A[i];
B[i] = _B[i];
}
fft(A, false);
fft(B, false);
for (int i = 0; i < maxN; i++) {
C[i] = A[i] * B[i] % mod;
}
fft(C, true);
}
};
const long long mod = 998244353;
const long long maxN = 1e13 + 20;
const int maxL = 50;
const int block_size = 59049;
const int max_block_cnt = max(block_size, (int)(maxN / block_size)) + 20;
long long pwA_small[block_size];
long long pwA_big[max_block_cnt];
long long ternary[max_block_cnt];
long long pwB[maxL];
long long big[2][1 << maxK];
long long small[1 << maxK];
long long prod[2][1 << maxK];
void just_do_it() {
long long N, A, B, C;
cin >> N >> A >> B >> C;
pwB[0] = 1;
for (int i = 1; i < maxL; i++) {
pwB[i] = pwB[i - 1] * B % mod;
}
pwA_small[0] = 1;
for (int i = 1; i < block_size; i++) {
pwA_small[i] = pwA_small[i - 1] * A % mod;
}
pwA_big[0] = 1;
pwA_big[1] = pwA_small[block_size - 1] * A % mod;
for (int i = 2; i < max_block_cnt; i++) {
pwA_big[i] = pwA_big[i - 1] * pwA_big[1] % mod;
}
ternary[0] = 1;
for (int i = 1; i < max_block_cnt; i++) {
ternary[i] = ternary[i / 3];
for (int j = 0; j < i % 3; j++) {
ternary[i] = ternary[i] * C % mod;
}
}
int block_cnt = (N + 1) / block_size;
int K = 0;
while ((1 << K) < block_size) {
K++;
}
for (int i = 0; i < block_cnt; i++) {
long long pref = 1LL * i * block_size;
int low = (pref & ((1 << K) - 1));
long long high = (pref >> K);
big[0][low] = (big[0][low] + pwA_big[i] * ternary[i] % mod * pwB[__builtin_popcountll(high)]) % mod;
big[1][low] = (big[1][low] + pwA_big[i] * ternary[i] % mod * pwB[__builtin_popcountll(high + 1)]) % mod;
}
for (int i = 0; i < block_size; i++) {
small[i] = (small[i] + pwA_small[i] * ternary[i]) % mod;
}
NTT::mul(big[0], small, prod[0], K + 1);
NTT::mul(big[1], small, prod[1], K + 1);
long long res = 0;
for (int i = 0; i < (1 << K); i++) {
res = (res + pwB[__builtin_popcount(i)] * prod[0][i]) % mod;
res = (res + pwB[__builtin_popcount(i)] * prod[1][i + (1 << K)]) % mod;
}
for (long long i = 1LL * block_cnt * block_size; i <= N; i++) {
res = (res + pwA_big[block_cnt] * pwA_small[i % block_size] % mod * pwB[__builtin_popcountll(i)] % mod * ternary[block_cnt] % mod * ternary[i % block_size]) % mod;
}
res = (res + mod - 1) % mod;
cout << res;
}