[OT] Checked opCast

[Ola Fosheim Grøstad via Digitalmars-d]

On Sunday, 3 July 2016 at 00:57:04 UTC, Timon Gehr wrote:
> How do you decide what 'redundant work' is? Is this combination 
> of branches and type casts really particularly cheap to execute?

He just provided an example, so preserving the  essence of the 
logic for all kinds of U and T is important. The sizeof test is 
flawed in general (it does not reflect the maximum for all types 
on all hardware).

The question was more whether "break" usually is more clear and 
performant for this type of code. And the answer is no...