# Talk:Eban numbers

From Rosetta Code

## Optimizations?[edit]

From the task description:

Only numbers less than one sextillion10will be considered in/for this task.^{21}

This will allow optimizations to be used.

I am mildly curious as to what optimizations this refers to. Maybe I am missing something, but I am failing to see how limiting to one sextillion offers any way to optimize. Not so big a deal for the Perl 6 example as it counts the *-ban numbers up to one sextillion 3 times in less than a second (on my system). I just wonder what I am overlooking. --Thundergnat (talk) 21:06, 22 March 2019 (UTC)

- See the
**REXX**solution (I think it has the best comments for eliminating numbers that have an**e**in them). At this time, I believe there are four other computer programming language solutions that use (more or less) the same algorithm.

- The algorithm roughly is:
- separate the (decimal) number into
*periods*(into groups of three digits starting from the right). - for the
*units*position, eliminate one, three, five, seven, eight, and nine. - for the
*tens*position, eliminate ten, eleven, twelve, teens, twenty, seventy, eighty, and ninety. - for the
*hundreds*position, eliminate all numbers greater than zero and aren't a blank, as*any*hundrd has an__e__**e**.

- separate the (decimal) number into

- This algorithm will work up to the sextillions (and beyond the septillions, it's hit and miss, so to speak). There are some numbers that have multiple spellings, so I included that sextillion limit to bypass these minefields. -- Gerard Schildberger (talk) 23:31, 22 March 2019 (UTC)

- Also of note, the algorithm mentioned (above) should have the numbers pluralized, but then
**twenty**would become**twenties**. and I thought the 2^{nd}spelling would maybe confuse some people. -- Gerard Schildberger (talk) 23:31, 22 March 2019 (UTC)

- Ah. Makes sense. I went a different way with Perl 6 but I can see your point. Thanks. --Thundergnat (talk) 23:56, 22 March 2019 (UTC)