Quote:
Shortly after AMD's Bulldozer architecture launched, AMD had us anticipating a couple of hotfixes that were supposed to improve FX-8150's performance. But Windows 8 was the ultimate goal. Now that the operating system is out, does it help FX-8150?
Quote:
It was a bummer, then, when AMD told us not to expect much from Windows 8 when it introduced FX-8350. But of course we wanted to go back and run benchmarks to follow up on our Bulldozer-based coverage. Can Microsoft's latest help make up some of the performance we were expecting to see back when FX-8150 launched, or are any possible operating system-related benefits already baked in?
Source.
The short answer is: at a 1% global difference, not really. The results from the different benchmarks are all over the place, sometimes there is almost no difference between the three, sometimes the unpatched Windows 7 wins, sometimes it's the patched Windows 7 that wins and sometimes it's Windows 8 that wins.
Windows 8 wins both audio transcoding tests, the first by 4 and 3 seconds compared to the patched Windows 7 and the unpatched Windows 7, the second by 5 and 6 seconds compared to the patched Windows 7 and the unpatched Windows 7.
However, if, for example, you're using Visual Studio 2010, the FX-8150 takes 20 seconds more to compile Chrome on Windows 8 than on the patched Windows 7 and 22 seconds more than on the unpatched Windows 7.
If you're using Adobe After Effects CS6, the unpatched Windows 7 wins by 6 and 5 seconds compared to the patched Windows 7 and Windows 8 respectively.
The only notable case where Windows 8 wins is in Dirt Showdown, where the unpatched Windows 7 comes second and the patched Windows 7 comes last, but that is with Medium detail settings with no AA. In the Ultra quality settings with 8x MSAA, they all perform the same.
In my opinion the supposed added efficiency of scheduling related threads to the same module which is supposed to be done in Windows 8 is negated by the fact that there are less resources available to each thread in the same module. Likewise, scheduling two related threads to different modules gives them more resources, but then they take longer to communicate with each other, so in many cases it ends up as a toss-up.
Edited by tpi2007 - 10/31/12 at 4:16am


















