I'm curious as to why you say this. Of course, I am making the assumption that the mechanism used to identify which rows should be updated is essentially static, possibly identifying a single row via a primary key, or a set of rows via a piece of a logical primary key. Nothing in the original poster's request spoke to the issue of actually finding the data to be updated, just how to update it once it had been found.

Do execution plans varying significantly for on UPDATE depending on which columns are to be updated? My understanding has always been that execution plans deal with locating the rows the statement is working on, not so much what happens to those rows once they are located.