source code
/* The Computer Language Benchmarks Game
https://salsa.debian.org/benchmarksgame-team/benchmarksgame/
contributed by Brad Chamberlain
derived from the Swift version by Ralph Ganszky
and the Chapel version by Nelson et al.
*/
use DynamicIters, IO, Math;
config const n = 200, // image size in pixels (n x n)
maxIter = 50, // max # of iterations per pixel
limit = 4.0, // per-pixel convergence limit
chunkSize = 1; // dynamic iterator's chunk size
param bitsPerElt = 8; // # of bits to store per array element
type eltType = uint(bitsPerElt); // element type used to store the image
proc main() {
const xsize = divCeilPos(n, bitsPerElt), // the compacted x dimension
imgSpace = {0..#n, 0..#xsize}; // the compacted image size
var image: [imgSpace] eltType, // the compacted image
xval, yval: [0..#n] real; // pre-computed (x,y) values
// precompute (x, y) values from the complex plane
const inv = 2.0 / n;
forall i in 0..#n {
xval[i] = inv*i - 1.5;
yval[i] = inv*i - 1.0;
}
// compute the image
forall (y, xelt) in dynamic(imgSpace, chunkSize) {
const xbase = xelt*bitsPerElt,
cr = (xval[xbase+0], xval[xbase+1], xval[xbase+2], xval[xbase+3],
xval[xbase+4], xval[xbase+5], xval[xbase+6], xval[xbase+7]),
ci = yval[y];
var Zr, Zi, Tr, Ti: bitsPerElt*real;
for 1..maxIter { // for the max # of iterations
Zi = 2.0*Zr*Zi + ci; // update Z and T
Zr = Tr - Ti + cr;
Tr = Zr*Zr;
Ti = Zi*Zi;
if (Tr + Ti > limit) then // break out if we haven't converged
break;
}
// store 'bitsPerElt' pixels compactly into the final image
var pixval: eltType;
for param i in 0..<bitsPerElt do
if (Tr(i) + Ti(i) <= limit) then // if 'C' is within the limit,
pixval |= 0x1 << (bitsPerElt-i-1); // turn the corresponding pixel on
image[y, xelt] = pixval;
}
// Write the file header and the image array.
stdout.writef("P4\n");
stdout.writef("%i %i\n", n, n);
stdout.writeBinary(image);
}
//
// Helper function to compare an 8-tuple and a singleton
//
inline operator >(xs, y) {
for x in xs do
if x <= y then
return false;
return true;
}
notes, command-line, and program output
NOTES:
64-bit Ubuntu quad core
chpl version 2.2.0
built with LLVM version 18.1.3
Copyright 2020-2024
Hewlett Packard
Enterprise Development LP
Copyright 2004-2019 Cray Inc.
Sat, 05 Oct 2024 22:55:41 GMT
MAKE:
mv mandelbrot.chapel-3.chapel mandelbrot.chapel-3.chpl
/opt/src/chapel-2.2.0/bin/linux64-x86_64/chpl --fast mandelbrot.chapel-3.chpl -o mandelbrot.chapel-3.chapel_run
rm mandelbrot.chapel-3.chpl
19.91s to complete and log all make actions
COMMAND LINE:
./mandelbrot.chapel-3.chapel_run --n=16000
(BINARY) PROGRAM OUTPUT NOT SHOWN