// Code written by Andy Huchala based on a joint program written with Jonathan Lee. // This implementation uses long long so I think it needs to run on a 64 bit machine. // This program computes dimensions for which there exist 8 or more non-isomorphic irreducible E6 representations. // See OEIS A343266. // Compile the program with the following command: g++ e6_new.cpp -o e6_new.exe -w // This implementation takes advantage of the dualities present in the E6 dynkin diagram. // Invariant under swapping both dual pairs at same time: x0<-> x4, x1<-> x3 in the ordering imported from mathematica. #include #include #include #include #include #include // Requires this file in the same directory, available at https://github.com/sercantutar/infint/blob/master/InfInt.h #include "InfInt.h" class Pair { public: // Stores coordinates in 96 bits: 64 + 32 bit std::pair coordinates; long double log; Pair(const std::pair & coordinates, const long double & log) { this->coordinates = coordinates; this->log = log; } }; bool operator<(const Pair & p1, const Pair & p2) { return p1.log > p2.log; } int main() { // checking up to 2^16 should get you at least 65 digits, i.e. [2^16-1,0,0,0,0,0] has dim 432199258280946504763829243161882977114556067368648124986539450368 const unsigned short int UPPER_BOUND_FACTOR = -1; long double* cachedLogs = (long double*) malloc(sizeof (long double) * UPPER_BOUND_FACTOR); const long long i1 = 1LL; const long long i16 = 16LL; const long long i32 = 32LL; const long long i48 = 48LL; const int i1_32bit = 1; const int i16_32bit = 16; const long i1shift16 = (1LL << 16LL) - 1LL; const int i1shift16_32bit = (1 << 16) - 1; const int cums[4] = {0, 16, 32, 48}; const InfInt denom = "23361421521715200000"; InfInt dim1 = "1"; InfInt dim2 = "0"; for (long i = 1; i < UPPER_BOUND_FACTOR; i++) { cachedLogs[i] = std::log ( i); } std::pair newPair = {0LL, 0}; std::priority_queue fringe; std::set> seen; const long long numToCompute = 20000; int numSeen = 1; fringe.push(Pair(newPair, 99.6740257330631)); long double lastLog = 0.0; long long newCoord; long double newIrrep; long long curCoord1; int curCoord2; unsigned short int x0; unsigned short int x1; unsigned short int x2; unsigned short int x3; unsigned short int x4; unsigned short int x5; std::cout<< "Initialization complete\n"; std::ofstream myfile ("e6_rep.txt"); if (myfile.is_open()) { int triple = 0; int quad = 0; int quint = 0; int hext = 0; int sept = 0; int oct = 0; int count_double = 0; while (numSeen < numToCompute) { Pair cur = fringe.top(); fringe.pop(); seen.erase(cur.coordinates); curCoord1 = cur.coordinates.first; curCoord2 = cur.coordinates.second; x0 = (unsigned short int) (curCoord1 & i1shift16) + 1; x1 = (unsigned short int) ((curCoord1 >> i16) & i1shift16) + 1; x2 = (unsigned short int) ((curCoord1 >> i32) & i1shift16) + 1; x3 = (unsigned short int) ((curCoord1 >> i48) & i1shift16) + 1; x4 = (unsigned short int) (curCoord2 & i1shift16_32bit) + 1; x5 = (unsigned short int) ((curCoord2 >> i16_32bit) & i1shift16_32bit) + 1; dim1 = "1"; dim1 *= (x0); dim1 *= (x1); dim1 *= (x2); dim1 *= (x3); dim1 *= (x4); dim1 *= (x5); dim1 *= (x0+x1); dim1 *= (x1+x2); dim1 *= (x2+x3); dim1 *= (x2+x5); dim1 *= (x3+x4); dim1 *= (x0+x1+x2); dim1 *= (x1+x2+x3); dim1 *= (x1+x2+x5); dim1 *= (x2+x3+x4); dim1 *= (x2+x3+x5); dim1 *= (x0+x1+x2+x3); dim1 *= (x0+x1+x2+x5); dim1 *= (x1+x2+x3+x4); dim1 *= (x1+x2+x3+x5); dim1 *= (x1+2*x2+x3+x5); dim1 *= (x2+x3+x4+x5); dim1 *= (x0+x1+x2+x3+x4); dim1 *= (x0+x1+x2+x3+x5); dim1 *= (x0+x1+2*x2+x3+x5); dim1 *= (x0+2*x1+2*x2+x3+x5); dim1 *= (x1+x2+x3+x4+x5); dim1 *= (x1+2*x2+x3+x4+x5); dim1 *= (x1+2*x2+2*x3+x4+x5); dim1 *= (x0+x1+x2+x3+x4+x5); dim1 *= (x0+x1+2*x2+x3+x4+x5); dim1 *= (x0+2*x1+2*x2+x3+x4+x5); dim1 *= (x0+x1+2*x2+2*x3+x4+x5); dim1 *= (x0+2*x1+2*x2+2*x3+x4+x5); dim1 *= (x0+2*x1+3*x2+2*x3+x4+x5); dim1 *= (x0+2*x1+3*x2+2*x3+x4+2*x5); dim1 /= denom; count_double = ((x4 > x0) && (x3 >= x1))||((x4 >= x0) && (x3 > x1)); if (dim1 == dim2) { if (triple) { if (quad) { if (quint) { if (hext) { if (sept) { if ( oct || count_double) { myfile <> i16) & i1shift16) + 1; x2 = (unsigned short int) ((newCoord >> i32) & i1shift16) + 1; x3 = (unsigned short int) ((newCoord >> i48) & i1shift16) + 1; x4 = (unsigned short int) (curCoord2 & i1shift16_32bit) + 1; x5 = (unsigned short int) ((curCoord2 >> i16_32bit) & i1shift16_32bit) + 1; if (((x4 < x0) && (x3 <= x1))||((x4 <= x0) && (x3 < x1))) { continue; } // std::cout<< std::to_string(x0)+ " " +std::to_string(x1)+ " " +std::to_string(x2)+ " " +std::to_string(x3)+ " " +std::to_string(x4)+ " " +std::to_string(x5)+ " " +std::to_string(x6)+"\n"; newIrrep = 0; newIrrep += cachedLogs[x0]; newIrrep += cachedLogs[x1]; newIrrep += cachedLogs[x2]; newIrrep += cachedLogs[x3]; newIrrep += cachedLogs[x4]; newIrrep += cachedLogs[x5]; newIrrep += cachedLogs[x0+x1]; newIrrep += cachedLogs[x1+x2]; newIrrep += cachedLogs[x2+x3]; newIrrep += cachedLogs[x2+x5]; newIrrep += cachedLogs[x3+x4]; newIrrep += cachedLogs[x0+x1+x2]; newIrrep += cachedLogs[x1+x2+x3]; newIrrep += cachedLogs[x1+x2+x5]; newIrrep += cachedLogs[x2+x3+x4]; newIrrep += cachedLogs[x2+x3+x5]; newIrrep += cachedLogs[x0+x1+x2+x3]; newIrrep += cachedLogs[x0+x1+x2+x5]; newIrrep += cachedLogs[x1+x2+x3+x4]; newIrrep += cachedLogs[x1+x2+x3+x5]; newIrrep += cachedLogs[x1+2*x2+x3+x5]; newIrrep += cachedLogs[x2+x3+x4+x5]; newIrrep += cachedLogs[x0+x1+x2+x3+x4]; newIrrep += cachedLogs[x0+x1+x2+x3+x5]; newIrrep += cachedLogs[x0+x1+2*x2+x3+x5]; newIrrep += cachedLogs[x0+2*x1+2*x2+x3+x5]; newIrrep += cachedLogs[x1+x2+x3+x4+x5]; newIrrep += cachedLogs[x1+2*x2+x3+x4+x5]; newIrrep += cachedLogs[x1+2*x2+2*x3+x4+x5]; newIrrep += cachedLogs[x0+x1+x2+x3+x4+x5]; newIrrep += cachedLogs[x0+x1+2*x2+x3+x4+x5]; newIrrep += cachedLogs[x0+2*x1+2*x2+x3+x4+x5]; newIrrep += cachedLogs[x0+x1+2*x2+2*x3+x4+x5]; newIrrep += cachedLogs[x0+2*x1+2*x2+2*x3+x4+x5]; newIrrep += cachedLogs[x0+2*x1+3*x2+2*x3+x4+x5]; newIrrep += cachedLogs[x0+2*x1+3*x2+2*x3+x4+2*x5]; fringe.push(Pair(newPair, newIrrep)); seen.insert(newPair); } for (int i = 0; i < 2; i++) { newCoord = curCoord2 + (i1_32bit << cums[i]); newPair = {curCoord1,newCoord}; if (seen.count(newPair)) { continue; } x0 = (unsigned short int) (curCoord1 & i1shift16) + 1; x1 = (unsigned short int) ((curCoord1 >> i16) & i1shift16) + 1; x2 = (unsigned short int) ((curCoord1 >> i32) & i1shift16) + 1; x3 = (unsigned short int) ((curCoord1 >> i48) & i1shift16) + 1; x4 = (unsigned short int) (newCoord & i1shift16_32bit) + 1; x5 = (unsigned short int) ((newCoord >> i16_32bit) & i1shift16_32bit) + 1; // don't add dual pairs if (((x4 < x0) && (x3 <= x1))||((x4 <= x0) && (x3 < x1))) { continue; } // std::cout<< std::to_string(x0)+ " " +std::to_string(x1)+ " " +std::to_string(x2)+ " " +std::to_string(x3)+ " " +std::to_string(x4)+ " " +std::to_string(x5)+ " " +std::to_string(x6)+"\n"; newIrrep = 0; newIrrep += cachedLogs[x0]; newIrrep += cachedLogs[x1]; newIrrep += cachedLogs[x2]; newIrrep += cachedLogs[x3]; newIrrep += cachedLogs[x4]; newIrrep += cachedLogs[x5]; newIrrep += cachedLogs[x0+x1]; newIrrep += cachedLogs[x1+x2]; newIrrep += cachedLogs[x2+x3]; newIrrep += cachedLogs[x2+x5]; newIrrep += cachedLogs[x3+x4]; newIrrep += cachedLogs[x0+x1+x2]; newIrrep += cachedLogs[x1+x2+x3]; newIrrep += cachedLogs[x1+x2+x5]; newIrrep += cachedLogs[x2+x3+x4]; newIrrep += cachedLogs[x2+x3+x5]; newIrrep += cachedLogs[x0+x1+x2+x3]; newIrrep += cachedLogs[x0+x1+x2+x5]; newIrrep += cachedLogs[x1+x2+x3+x4]; newIrrep += cachedLogs[x1+x2+x3+x5]; newIrrep += cachedLogs[x1+2*x2+x3+x5]; newIrrep += cachedLogs[x2+x3+x4+x5]; newIrrep += cachedLogs[x0+x1+x2+x3+x4]; newIrrep += cachedLogs[x0+x1+x2+x3+x5]; newIrrep += cachedLogs[x0+x1+2*x2+x3+x5]; newIrrep += cachedLogs[x0+2*x1+2*x2+x3+x5]; newIrrep += cachedLogs[x1+x2+x3+x4+x5]; newIrrep += cachedLogs[x1+2*x2+x3+x4+x5]; newIrrep += cachedLogs[x1+2*x2+2*x3+x4+x5]; newIrrep += cachedLogs[x0+x1+x2+x3+x4+x5]; newIrrep += cachedLogs[x0+x1+2*x2+x3+x4+x5]; newIrrep += cachedLogs[x0+2*x1+2*x2+x3+x4+x5]; newIrrep += cachedLogs[x0+x1+2*x2+2*x3+x4+x5]; newIrrep += cachedLogs[x0+2*x1+2*x2+2*x3+x4+x5]; newIrrep += cachedLogs[x0+2*x1+3*x2+2*x3+x4+x5]; newIrrep += cachedLogs[x0+2*x1+3*x2+2*x3+x4+2*x5]; fringe.push(Pair(newPair, newIrrep)); seen.insert(newPair); } } myfile.close(); } free(cachedLogs); return 0; }