I've used both setups and I think that either TR is flawed, or their results are limited to skyrim. It is no secret that nvidia performs substantially better in skyrim, especially with SLI - crossfire just can't compare. Conversely, crossfire does much better in other games. I don't think you can make a blanket statement that one is completely better than the other, because some games are better on one and some games aren't...it goes back and forth. For instance, I highly suspect if you do similar testing with metro 2033, you would find that crossfire performs very much substantially better than any SLI setup. Again....depends on the game IMO. AMD gaming evolved games sometimes stink with SLI, and TWIMTBP games sometimes stink with crossfire. Is this surprising? Not really.
So in a nutshell, I think it just depends on which games are being tested. My opinion is that one isn't smoother than another as i've used both. Its whatever though - if another website has more data with more games that would be great as well; perhaps SKY can do such testing. He did uncover the uneven boost speeds several months ago, which was a great find - and a good reason why I made the switch to MSI GTX 680. They're the best :biggrin:. Anyway, such testing for stutter would be great as long as it is factual with proper testing methods, and not subjective.