That just means there aren’t 4K times as many file read ops. The read() calls (which most of the time just increment the index into the buffer and returns what

That just means there aren’t 4K times as many file read ops. The read() calls (which most of the time just increment the index into the buffer and returns what’s there) are still made. The usual way of doing what he’s trying to do will look exactly the same in C# as in his php version, with explicit buffering.