Personally, when comparing CPU's the only benchmarks I use are actual game play and 3d mark. It has a compare tool on it's website. 3dmark is still technically a synthetic, but it is the closest to real world as synthetic can get in my opinion. It gives you a physics score, which is basically how well the CPU on it's own can compute physics calculations. Apart from gaming, real program benchmarks such as a Photoshop is a good go by.
As for the game benchmarks, I always make sure that the reviewer uses settings and resolutions that a gamer will really use. That type of review will show you real world speed comparisons, rather than some reviewers that will run 1600x900 resolution, no AA, low detail settings on $1200 worth of hardware, which is completely pointless in my opinion.
All that being said, That 4570 will be generally faster overall and in some games will show a healthy increase in performance, and others it will not gain much over the old Phenom. As far as a remarkable increase, probably not in most games. The main limiting factor will still be the GPU, more so than the CPU in almost every gaming scenario. I don't notice much difference between my old 965 rig and a buddy's 3570 rig as the GPU's are fairly evenly matched, but his seems smoother in some games than mine does. If we both were to monitor our games, his would be faster in FPS numbers I'm sure, but again it's hard to tell the difference with the naked eye.