Hi @Anonymous,
A collection by definition exists in memory on the local machine. Once a collection is created, performance should be almost instantaneous so there won't be much of a difference depending on the number of columns. That being said, there are sometimes other reasons for limiting the number of columns in a collection. This is done by shaping the data before creating the collection using the AddColumns, DropColumns, RenameColumns and ShowColumns functions. These are typically used inside of the Collect or ClearCollect() function. If you are working with collections, these functions are very important to know about. An example doing what you are requesting would be
ClearCollect(colCustomers, ShowColumns(Customers, "LastName", "Email")
to only show a few columns from what could be a list with a lot of columns. Please note that the column names must be enclosed in double quotes. Also please be aware that these functions are not delegatable, so if your list is >2000 items, you will run into problems with them when you create the collection unless you filter the list down to <2000 items.
The inverse function DropColumns() is used to eliminate columns from the data before making the collection. None of these functions actually changes the underlying data in the list in any way.
For a complete reference see function-table-shaping.
