Sean/Jeff:

Thank you. I read the articles AFTER I posed. :unsure:

I ran some tests and I'm amazed by DelimitedSplit8K's speed. I ran tests using both methods on the following string -- just random jibberish:

'1,2,4,5,6,7,8,99,10,2,4,5,6,7,8,99,10,2,4,5,6,7,8,99,10,2,4,5,6,7,8,99,10,2,4,5,6,7,8,99,10'

XML : 55 seconds

DS8K : 43 seconds (28% faster!)

Great work!

I guess the good news here is that if I need to screaming performance, use DS8K. 😀

Good news for my project is that the DBAs understand the XML version; but I'm going to mention DS8K for a project that really needs to scream.

Thank you for all your great work and time! (I enjoy seeing the objective analysis you two have put together on this topic. I just wish Microsoft would address this issue. :-D)

(ah man, i really thought i was submitting a good nugget! i felt like i was hording it for years lol)