Benchmark: running sub, sum, each for at least 6 CPU seconds...
       sub:  7 wallclock secs ( 6.00 usr +  0.00 sys =  6.00 CPU) @ 15883.17/s (n=95299)
       sum:  6 wallclock secs ( 6.02 usr +  0.00 sys =  6.02 CPU) @ 40993.85/s (n=246783)