Brian
It looks like you have a winner! I ran a test of DelimitedSplit8K using your method against 3 other methods including the Itzik Ben-Gan method. Your version consistently outperformed the Itzik method--not by much, but clearly a slight edge based on the numbers on my machine. The test results might be different on different servers of course.
The test results are attached.
Good work!
Steven Willis