C _AMPとは？ わかりやすく解説

#include <amp.h>
using namespace concurrency;

int main()
{
    const int ElemCount = 100;
    int ary[ElemCount];
    array_view<int, 1> av(ElemCount, ary);

    parallel_for_each(extent<1>(ElemCount),
    [=](index<1> idx) restrict(amp) 
    {
        int i = idx[0];
        av[i] = i;
    });
    av.synchronize();
}

#include <cstdio>
#include <amp.h>
#include <amp_graphics.h>

using namespace concurrency;
using namespace concurrency::graphics;

int main()
{
    const int width = 10;
    const int height = 3;
    const int depth = 2;
    const int elemCount = width * height * depth;

    std::vector<float_3> srcHost(elemCount);
    std::vector<float_3> dstHost(elemCount);

    for (int z = 0; z < depth; ++z)
        for (int y = 0; y < height; ++y)
            for (int x = 0; x < width; ++x)
            {
                srcHost[z * height * width + y * width + x] = float_3(float(x), float(y), float(z));
            }

    array_view<float_3, 3> srcView(depth, height, width, srcHost);
    array_view<float_3, 3> dstView(depth, height, width, dstHost);

    parallel_for_each(
        dstView.extent,
        [=](index<3> idx) restrict(amp)
    {
        const auto globalx = idx[2];
        const auto globaly = idx[1];
        const auto globalz = idx[0];
        auto temp = srcView[idx];
        if (globaly == 0)
        {
            temp += 0.5f;
        }
        if (globalz == 0)
        {
            temp *= -1;
        }
        dstView[idx] = temp;
    });

    dstView.synchronize();
    for (int z = 0; z < depth; ++z)
        for (int y = 0; y < height; ++y)
            for (int x = 0; x < width; ++x)
            {
                const auto val = dstHost[z * height * width + y * width + x];
                printf("[%d][%d][%d] = (%.1f, %.1f, %.1f)\n", z, y, x, val.x, val.y, val.z);
            }
}