You Intel guys don't give up. Is it bothering you this much that Intel just basically released a side-grade for gamers and enthusiasts and AMD is releasing all this stuff for gamers and enthusiasts?
The CPU coming out at the rumored $800 is impossible. It's going to OEMs first and a CPU of that price would inflate the cost of the final system to ridiculous levels.
Get over it guys, AMD still cares about us and Intel doesn't. All Intel has going for enthusiasts right now is bigger bar graphs in old games. If Intel wasn't neglecting and ignoring us I wouldn't be so hostile towards them, but I'm not going to defend a company that's literally killing us off (locking down overclocking, not giving people a reason to upgrade, ignoring us, removing features from CPUs in trade for overclocking, inhibiting overclocking and encouraging de-lidding by using TIM instead of solder for IHS) just because they can encode mp3s in LAME 40 seconds faster or they get 140fps in Skyrim instead of 90.I simply don't get how you guys can't comprehend this chip is going to be faster than 4770k in multi-thread. FX 8350 is already close in fairly compile programs (like x264), and it's getting a 20% to 25% clock speed increase stock.
4.9ghz beats stock 4770k in x264 and new FX is going to run at that speed. It's the first time in a long time where AMD is going to have a significant lead over a new Intel product
. And all they did was make some small tweaks to the existing architecture that is much slower. Contrast with Intel who spent billions on R&D for Haswell and they gave us nothing.
The reason why I use x264 is because I know it's a fair benchmark. Considering AMD is going to have their hands all over next gen games, the quality of compiled code optimizations is going to be very far up in the air.
x264 is one program I have not seen ridiculous gains in between Gentoo and Windows, so I feel it's an accurate representation of FX 8350's true performance. Contrast this with benchmarks like LAME where I see a 60% increase in performance in Gentoo.If you can not prove that a benchmark is fair and does not give excessive gains on certain hardware after being recompiled, I have no interest in it.
I have seen performance double by recompiling. I no longer trust benchmarks that I can not recompile.