Awesome to have the heavy hitters on the case! Jeff, where you say:
You're not actually supposed to use any data from that. You're only supposed to use the presence of rows instead of writing a loop
I understand that, but if you run that it returns 5 columns of zeros. All called [n]. And I can't work out why that's necessary. Surely one column of zeros would be enough.
Edit: I think I get it now! The cross joins mean that 10^5 rows are generated, and you're saying it's irrelevant that 5 meaningless columns just happen to be generated. Doing an extra CROSS JOIN would presumably result in 6 meaningless columns and a million rows. Does that sound right?
Edit 2:
a readless Tally Table
So that means a tally table that doesn't have to be read from the disk, and is presumably therefore faster than the permanent tally table I usually use. How am I going so far?
...One of the symptoms of an approaching nervous breakdown is the belief that ones work is terribly important.... Bertrand Russell