I'm not aware of that, but from the review it clearly shows it doesn't work on win7 and does on win8... so again turbocore ^^ which is funny cuz... dur thats what they marketed as! Anyway from recent discoveries i'm probably gonna get one.
Its one of the first games to support multi threaded DX11 driver....... Its where BD will shine a little.
that doesn't seem right the x4 980 is 4 core and is faster then i7 2600k 4 core even at core to core clock? If is multi threaded shouldn't the x6 with 6 physical core faster then x4 980? Windows 8 will kinda fix this
Well the 980 is at 3.7ghz and the 1100 is at 3.3ghz, i recon that's most likely causing the difference, it was only 1 fps between those though. It just seems the intel/SB are slightly slower in BF3 (the beta that is, no idea on retail). That's the funny thing, according to the BF3 benchmarks the difference between the FX-8150 and the 2600k is about the same as the difference between the X4 980 and the 2600k (around 3 fps faster).
Because core parking (well, in Windows 7; dunno whether it works the same in Windows 8) is essentially an OS feature, not something inherent to CPU. Windows 7 can utilize core parking on both Intel & AMD multi-core CPUs; it's basically to save power. It'll turn off cores that doesn't have much work going on, and shift the threads (well, this is more of the scheduler's work) to another core that isn't fully loaded. Of course that has to work in tandem with a good scheduler, so that it knows exactly when to switch on/off a core. In Windows 8, BD performs better might be due to the fact some of its cores are being turned off by core parking that has a better scheduler, therefore freeing the FPU on a module entirely for a core. One of BD flaws is that each module share an FPU, and a module contains two cores. When each BD core has less ALU than a PII and two BD cores share an FPU, it just doesn't sit well with me. That's probably why an FX-8150 with 4 of its cores from every module is turned off is able to perform better at full load compared to all 8 cores turned on. CMIIW, though.
...and according to andtech because of windows 7's handling of multicore threading its causing threads to spread across all 8 cores rather than focus on the ones needed which is going to cause both turbocore and scheduler to not allow core parking etc etc. Anyway that's the gist of it. This seems like a case of some one dropped the ball and you can blame them all. Each party seems to have forgotten something or did something wrong. AMD, Asus and M$. Its caused quite the clusterf**k
I'm guessing that Win7's scheduler is just more used to Intel's HT. HT has been around for quite some time, so maybe it's just natural. To be fair, Win7's probably just view BD cores as, well, cores; it doesn't expect that each core shares a resource (FPU) with another. But, meh, who knows. =/ Welcome to industry 'standards'. =b
I feel like AMD should have made some effort to make the processor work at full speed in the current to date operating system? I mean how can you blame Microsoft for not having support for this?
Well, here's the thing, software is suppose to make hardware work. Hardware shouldn't have to conform to software. Microsoft could have well released a kernel patch to better support the AMD FX series CPUs. The guys over at Linux have been working hard to include better FX series CPU support, I can't see why Microsoft can't do the same. deltatux
I still don't think there'll be much difference tbh. The guys at linux said there'll only be about 5% boost too.
From AMD's own slides they only show an average 4% increase in Windows 8 scheduler and the Linux guys are talking about a max of 10% but that it breaks compatibility.
You can't really use projections of Linux performance to guestimate what, if any, boost Bulldozer could get in Windows. The underlying code is vastly different.
Unless M$ wants to sell windows 8 for some reason XD. Also we could give M$ some slack and say whatever the fix is, it turned out more involving than they expected and is what caused lengthy delays and AMD was forced to release if they hoped to turnover during the holiday season and hope that M$ can release a fix soon. But I won't defend M$. I think its quite BS that they don't have proper support by now. Also Asus just pisses me off with their bios some times, this is just another one of those times. I have also seen reviews with the Asrock showing roughly the same results as the Crosshair V... Asrock is built by asus- see a connection yet? Anyway I have decided to see what the final verdict w/ bf3 will be in a few weeks and then i will make my decision about the FX.
Microsoft can't just update how thread management works in a few months, which is all the time that AMD gave them I'm sure (if they went to MS at all asking to work with them on a fix). Considering how small the improvement on the Windows 8 developer preview is it sounds like an excuse anyway. Considering that Windows has to have a minimum amount of bugs (and especially no serious ones), implementation and testing of a kernel patch to change that sort of thing could take a year or more to implement. It's probably why it's being reserved as a feature for Windows 8.
Except the tree rotted in this case after the apple fell ASRock products are far superior than ASUS products Especially since ASRock support rivals EVGA support
Its there a way to Swap the positions of cores so windows can recognize them in a different way? Right now Im prety sure its from 0 to 7, making 0 2 4 and 6 the real cores and 1 3 5 and 7 the scondary cores, so in my opinion if theres a way to make window asign 0 2 4 6 for games and other apps that doesnt use heavy MT, it would improve performance, right? There alot of confussion regarding BD because in reality its just a Quad core with SMT, And it should be handled as one, If windows and Apps could run on just 1 Thread per Module Im sure the performance on non masive MT software would be much better and use the secondary cores only if the app uses more than 4 threads.