> It's really just easiest if you take nothing for granted, and even assume that clock-cycles aren't of a fixed duration but of a given quanta of work. Thanks

> It's really just easiest if you take nothing for granted, and even assume that clock-cycles aren't of a fixed duration but of a given quanta of work. Thanks to spread-spectrum clocking, this really ought to be the default assumption for most computers. It's a small variance, but it's there.