 ./testingmlp

Initializing array made of 33554432 64-bit words (256.00 MiB).
Legend:
  BandW: Implied bandwidth (assuming 64-byte cache line) in MB/s
  % Eff: Effectiness of this lane count compared to the prior, as a % of ideal
  Speedup: Speedup factor for this many lanes versus one lane
Applying Sattolo's algorithm... chain total: 33554432
Time to sum up the array (linear scan) 0.002 s (x 8 = 0.018 s), bandwidth = 112775.3 MB/s
Size: 33554432 (262144.00 KiB, 256.00 MiB)
---------------------------------------------------------------------
- # of lanes --- time (s) ---- BandW -- ns/hit -- % Eff -- Speedup --
---------------------------------------------------------------------
           1     3.164235        647      94.3       0%        1.0
           2     1.564678       1309      46.6     101%        2.0
           3     1.044693       1960      31.1     100%        3.0
           4     0.785171       2608      23.4      99%        4.0
           5     0.631248       3244      18.8      98%        5.0
           6     0.527432       3883      15.7      99%        6.0
           7     0.454349       4508      13.5      97%        7.0
           8     0.399504       5126      11.9      97%        7.9
           9     0.356725       5741      10.6      96%        8.9
          10     0.323821       6324       9.7      92%        9.8
          11     0.295799       6924       8.8      95%       10.7
          12     0.272921       7504       8.1      93%       11.6
          13     0.252469       8112       7.5      97%       12.5
          14     0.236324       8666       7.0      90%       13.4
          15     0.222983       9185       6.6      85%       14.2
          16     0.207731       9859       6.2     109%       15.2
          17     0.196219      10437       5.8      94%       16.1
          18     0.186190      11000       5.5      92%       17.0
          19     0.176983      11572       5.3      94%       17.9
          20     0.169033      12116       5.0      90%       18.7
          21     0.161752      12661       4.8      90%       19.6
          22     0.154898      13222       4.6      93%       20.4
          23     0.148677      13775       4.4      92%       21.3
          24     0.142893      14332       4.3      93%       22.1
          25     0.137773      14865       4.1      90%       23.0
          26     0.133031      15395       4.0      89%       23.8
          27     0.128765      15905       3.8      87%       24.6
          28     0.124731      16419       3.7      88%       25.4
          29     0.134415      15236       4.0    -225%       23.5
          30     0.134676      15207       4.0      -6%       23.5
          31     0.137940      14847       4.1     -75%       22.9
          32     0.128903      15888       3.8     210%       24.5
          33     0.124372      16467       3.7     116%       25.4
          34     0.125965      16259       3.8     -44%       25.1
          35     0.121501      16856       3.6     124%       26.0
          36     0.116982      17507       3.5     134%       27.0
          37     0.117322      17456       3.5     -11%       27.0
          38     0.115186      17780       3.4      69%       27.5
          39     0.110525      18530       3.3     158%       28.6
          40     0.110889      18469       3.3     -13%       28.5
Maybe you have about 28 parallel paths?
