removed modulus from main loop

[peterchenadded]

As discussed, minor improvement to remove expensive modulus operator.

----------------------------------------------------------------------------------- benchmark: 3 tests -----------------------------------------------------------------------------------
Name (time in ms)          Min                 Max                Mean            StdDev              Median               IQR            Outliers       OPS            Rounds  Iterations
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
test_small              5.0094 (1.0)        7.4816 (1.0)        5.3191 (1.0)      0.3314 (1.0)        5.2602 (1.0)      0.1653 (1.0)           8;9  188.0013 (1.0)         191           1
test_smooth_1000       52.8989 (10.56)     59.5790 (7.96)      54.3368 (10.22)    1.7066 (5.15)      53.6480 (10.20)    1.1022 (6.67)          2;2   18.4037 (0.10)         16           1
test_large            112.4638 (22.45)    120.6921 (16.13)    116.3287 (21.87)    2.9104 (8.78)     116.3253 (22.11)    5.0715 (30.69)         4;0    8.5963 (0.05)          9           1
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

----------------------------------------------------------------------------------- benchmark: 3 tests -----------------------------------------------------------------------------------
Name (time in ms)          Min                 Max                Mean            StdDev              Median               IQR            Outliers       OPS            Rounds  Iterations
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
test_small              5.1134 (1.0)        7.5196 (1.0)        5.3400 (1.0)      0.2995 (1.0)        5.2774 (1.0)      0.1251 (1.0)          7;10  187.2676 (1.0)         187           1
test_smooth_1000       53.7028 (10.50)     58.3828 (7.76)      55.1907 (10.34)    1.5705 (5.24)      54.4994 (10.33)    1.1452 (9.15)          4;4   18.1190 (0.10)         19           1
test_large            113.6079 (22.22)    122.6846 (16.32)    116.3847 (21.80)    3.6097 (12.05)    114.6583 (21.73)    4.1340 (33.04)         2;0    8.5922 (0.05)          9           1
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

----------------------------------------------------------------------------------- benchmark: 3 tests -----------------------------------------------------------------------------------
Name (time in ms)          Min                 Max                Mean            StdDev              Median               IQR            Outliers       OPS            Rounds  Iterations
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
test_small              5.0263 (1.0)        7.4791 (1.0)        5.2821 (1.0)      0.2391 (1.0)        5.2491 (1.0)      0.1043 (1.0)         10;12  189.3175 (1.0)         190           1
test_smooth_1000       53.2688 (10.60)     61.0834 (8.17)      54.6131 (10.34)    1.7800 (7.44)      54.0731 (10.30)    1.2326 (11.82)         2;1   18.3106 (0.10)         19           1
test_large            113.3363 (22.55)    115.6210 (15.46)    114.3397 (21.65)    0.7581 (3.17)     114.3845 (21.79)    1.0992 (10.54)         4;0    8.7459 (0.05)          9           1
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

[hjweide]

Thanks for putting this up so quickly! Let me think about this a bit and take a look over the next few days.

[peterchenadded]

Sure, take your time. Probably too small to be worthwhile.

[hjweide]

I actually suspected that the compiler would already do this, but the improvement we observe is non-zero, so I want to think about what's actually happening here.