http://www.tomshardware.com/reviews/core-i3-gaming,2588-9.html
http://www.tomshardware.com/reviews/athlon-ii-x3-440-gaming-performance,2619-5.html
look at the far cry 2 and crysis performance, at the first review and then at the second. Both benches conducted with a hd5850 and the phII 720 ~ athlon II x3, same goes for the i7 870 and the 920 The first review clearly shows that a phII x3 @ 2.8ghz (and an athlon x3 3.0ghz, for that matter) produces the same performance as the i7, and sometimes even surpasses it by 1 fps.
Wait, what happens in the 2nd review?
The first review came out by my personal favorite, Thomas Soderstorm, who likes to show things the way they are, even if it means writing a whole article with monotonic graphs all the way.
The second review came right after the first one, clearly manipulating graphics setting - Why would
anyone who buys these kind of rigs play far cry 2 at high and not ulrta.. and don't even get me started about the almost complete lack of AA / AF in tomshardware reviews. Why you ask? The less graphical settings, the more cpu difference is shown. This is a joke of course since even a ~300$ gpu, which is a sum most people won't be willing to pay, will bottleneck even the "lowly" athlon II x3 cpu @ maximum settings - again, show me 1 person in this planet who wouldn't want to crank max settings on these kind of rigs.
Further reading on the 2nd article shows the minimum fps, which doesn't really say anything unless you're really into paying ~180$ just for a 10 fps increase on world in conflict (and a slight decrease in stalker)
which is a single-player RTS.. Doh..
But why stop there? Now lets see how these cpus fare against each other with hd5870 crossfire configuration. These cards will have no problem playing every single game available at 2560x1600, and even in an eyefinity setup which, by connecting 3 full-hd monitors, can produce more pixels than the 2560x1600 at a lower price, and even get 6 of these monit... Well you know that already! but price is probably not an issue for you if you get 2 5870.. No? Guess not! They only show resolutions up to 2560x1600,
again at high settings only which is of course misleading because when they write "high settings" you naturally assume that these are the highest settings possible. Guess not.
(Well this is getting long but I never had the energy to write all this and I started already so... Ok moving on
)
Last but not least (finally my boss is getting angry at me
), the article shows the athlon II with dual 5870 and the i7 with a single 5850. The athlon already wins at 2560x1600, but that's where they stop. High settings of course, why cranking it all up?
Finally, my point is - Toms, you could say that the i7-930 is justified if a guy wants to buy a hefty gpu config,
but not based on this article like you're just doing. I'm not calling bias or anything, it's just too bad that you didn't show us this data that you showed in the aforementioned artiel
and the same data only with the highest graphical settings possible (including AA / AF maxed).