I would take these results with caution. First, the 23% difference was only observed in one game, the others were about 10%. Second, it is not clear what clocks the cards ran at. It is entirely possible that the Titan was more held back than the 290X due to the Titan having only 80C of temperature to work with - something you can easily change like you can set the uber mode on the 290X btw.
If you look at more detailed benchmarks, you will find that both cards perform roughly the same per clock at 4k. Sometimes one is in front, sometimes the other:
http://www.pcgameshardware.de/Geforc...est-1093796/5/
Bookmarks