• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Windows 8 and Bulldozer with turbo disabled

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

DarthGrantius

Registered
Joined
Sep 25, 2012
Location
New Zealand
Hi all,

On Windows 7 I liked how it used the first core of each module of my FX-8120 during 4 threaded gaming as per maximum performance.

Now in Windows 8, with the scheduler upgrades it seems like it uses cores 0-3 during gaming while 4-7 remain parked causing the first 2 modules to be loaded up.

My question is, as I have APM and turbo turned off and it overclocked to ~4ghz, how do I get Windows to spread the games to core 0,2,4,6 for more performance? I seem to have lost a few FPS since I upgraded even with same or newer drivers.

Sorry if I'm confusing, it just seems like the improvements they made are for turbo enabled chips at stock settings to ramp up the first 2 modules. Or has Windows 8 changed so that what it reports as core's 0, 1 ,2 and 3 are actually 0, 2, 4, 6?
 
Turbo will not work with cpu speed clocked manually to 4.0Ghz all the time. TurboCore would only move cpu speed up to 4.1Ghz on an FX-8120 cpu.

AMD tried to say that loading the first two modules would do the deed, but I have seen where that will only work for certain types of apps. So I am not really surprised you see a little less performance.
 
Turbo will not work with cpu speed clocked manually to 4.0Ghz all the time. TurboCore would only move cpu speed up to 4.1Ghz on an FX-8120 cpu.

AMD tried to say that loading the first two modules would do the deed, but I have seen where that will only work for certain types of apps. So I am not really surprised you see a little less performance.

Yeah I disabled it to get a much lower voltage and to stop TDP throttling.

I can see why single threaded operations can benefit from sky high turbo core clocks, but for multithreading from what I've read you see a much bigger performance increase by giving each thread a module as opposed to shared threads at a higher clock speed - FPU sharing perhaps?
 
Back