Platform: NVIDIA CUDA
  Device: GeForce GT 750M
    Driver version  : 335.23 (Win32)
    Compute units   : 2
    Clock frequency : 1085 MHz

    Global memory bandwidth (GBPS)
      float   : 24.51
      float2  : 25.27
      float4  : 25.68
      float8  : 13.01
      float16 : 12.34

    Single-precision compute (GFLOPS)
      float   : 580.07
      float2  : 765.75
      float4  : 739.55
      float8  : 757.51
      float16 : 740.08

    Double-precision compute (GFLOPS)
      double   : 37.16
      double2  : 37.11
      double4  : 37.05
      double8  : 36.90
      double16 : 36.61

    Transfer bandwidth (GBPS)
      enqueueWriteBuffer         : 2.52
      enqueueReadBuffer          : 2.39
      enqueueMapBuffer(for read) : 2.48
        memcpy from mapped ptr   : 2.65
      enqueueUnmap(after write)  : 2.40
        memcpy to mapped ptr     : 2.79

    Kernel launch latency : 100.45 us