Yes, all good advice.

Our testing was the only activity on the server. As the process of changing alignment is destructive the only files left on the disk were the test files we created in SQLIO and each time we ran tests the same size file was used so the comparison should be representative.

From my testing so far, I don't see it is worth the effort in making the same changes on the production systems, however, I am open to retesting using other quantifiable metrics that may prove more consistently that it is worth it.