Not odd. The 'magic' is in the fact that Valery put an evaluation (=) after the range definition of i. That changes the range to a vector. This is a special feature of Prime.
With that evaluation your sheet shows:
The summation with a single i at its bottom, requires i to be a range; vector is not allowed.
I assure you that, as a Brit, I consumed devilish humour along with my first teething rusks. And being ex-military means that I have to consciously rein it in. 😈
However, dealing with the last bit first, sorting. AFAIA, Mathcad 11 and Mathcad Prime ignore the imaginary component of a complex number. I dug out an old M11 mergesort worksheet that allowed me to play with different sorting options. Naturally, it is broken under Mathcad Prime - I suspect a Boolean shortcut difference is the cause, but I can't see it. Instead, I've used an alternative approach to provide various options for sorting complex numbers.
(You can tell how old the original worksheet is from the reference to Axum S-Plus! I haven't posted the worksheet, because it's a bit of a slapdash thing even by my standards, with expressions dumped where I could find a space)
I plugged my all-singing, all-dancing vec into max and min, and that seems to work for some cases of lack of interest. However, there are couple of more that I need to address, and my current solutions are, at best, 'clumsy', if I'm being charitable.
Stuart
PS. If anyone spots any discrepancies or errors, please post your comments in a Spoiler section and set a 24-hour time lock on it. My poor brain can't cope with any more thinking today.