I'm betting that like most AMD fans you have no clear idea whether this would actually benefit you in any meaningful way, you just want more cores.
That's patronizing and ignorant. Lots of real cores (not SMT flimflammery) is great for compiling, for virtualization, and many other loads. Intel likes to optimize for brain damaged single threaded loads, if that's what you've got then go Intel. Better, fix your software. I'm buying AMD the next time around because it more than addresses my needs for significantly less money.
"FMAC" operations are standard floating point operations
No. Multiply-accumulate is a specialized instruction whose main purpose is to preserve intermediate precision to better support solving ill-conditioned sets of equations and the like. It is not necessarily more efficient than a standard two instruction multiply add. Suggestion: consult Wikipedia before trotting out your ignorance.
Meaningless buzzwords can frabjulate the frobozz nicely!
Meaningless only to a dog watching television, such as you. To spell it out: intelligent task scheduling can pair integer-intensive tasks with MMX tasks on the same core. The Linux kernel already includes many similar, core-aware scheduling optimizations.
What Apple did was finally admit to itself that it was incapable of developing a reliable multitasking OS in-house. Taking the plunge and adopting open source Unix is arguably the least evil thing Apple ever did.
Unfortunatly, IOS && OSX of today are greatly modified versions of BSD, with a different kernel and apis. Macs were a great unix desktop ten years ago, now they just kind of blow. Linux is the only way to go these days for Unix && desktop.
Strange what you say there, considering that MacOS X is actually POSIX certified, and Linux for obvious reasons isn't. If you want Linux, use Linux. If you want Unix, Linux isn't Unix. Never was. Never will be.
Apple did not make the cash pile they are sitting on because of the patents.
Perhaps that has something to do with the fact that Apple lost more than $100 billion of market value in the last seven weeks, after becoming a patent troll.
apple doesn't have much FRAND patents but they have a lot of non-FRAND patents that others want.
That is not clearly true at all. It is more accurate to say, Apple says it has a lot of patents that others want. Case in point: the bouncing scroll patent. Invalidated.
My feeling: Apple has a bunch of junk patents but is skilled at gaming the courts.
Apple was a thug long before Steve Jobs declared thermonuclear war, but Jobs was a master at concealing that. Tim Cook isn't. This is one of several reasons that it is just wonderful for the rest of us that Tim Cook now runs Apple.
should influencing the future of technology be Slashdot's main mission?
Maintaining as high as possible a concentration of clueful posters and interested readers should be Slashdot's main mission. Then influencing the future will take care of itself.
If you buy a 12 or 16-core CPU, it's not because you want your Facebook page to load faster. It's because you have some serious parallel workload to process, likely involving a lot of calculations.
What kind of calculations? For many floating point intensive applications, such as gaming or video encoding, the GPU will be doing the vast majority of your floating point calculations, freeing up the CPU cores for fast integer logic. Many parallel applications, like compiling, do little to no floating point.
6200 series have shared FPU (floating point unit). Which means that there are less FPUs that there are processing cores. To multiply two floating point numbers cores are waiting in a queue until FPU is free to use, this happens when all cores are calculating at the same time. If you are doing intensive calculations this is going to be slower than if you used 6100 series. 6100 series have dedicated FPU for each core.
The transistors saved by sharing the FPU are returned to you in integer cores, arguably better use of the transistor budget. Here's a more nuanced view:
Steamroller apparently only shares the MMX/FMAC unit, while standard floating point operations are per-core. Intelligent scheduling could optimize this arrangement nicely.
On the whole it looks to me like I'll be AMD again for my next box. Lots of cores for little $$$, yummy.
There aren't that many places to buy high quality LCD displays in volume and Apple just burnt its bridges with Samsung. I say, with Sharp teetering (no doubt in part because of Apple stinginess) LG is in an excellent position to raise prices.
And by the way, if your report is correct then it is Windows that is broken. NTP works the same way for Windows and Linux. On the other hand never mind, I think you're just pulling facts out of your behind anyway.
All of the clocks on our domain are sync'ed apparently that's a "feature" of Active Directory. Linux doesn't apparently respect that so when our file server somehow got a couple minutes out of sync with our AD, AD decided that it would just lock out access.
You seem to lack any notion of ntpdate and the fact that it is standard on all Linux server installs. Is that hair I see growing on your feet?
Slashdot Mirror
Ah, as pointed out below, AMD's MMX is already per-core, while the relatively rare 256 AVX path is shared.
I'm betting that like most AMD fans you have no clear idea whether this would actually benefit you in any meaningful way, you just want more cores.
That's patronizing and ignorant. Lots of real cores (not SMT flimflammery) is great for compiling, for virtualization, and many other loads. Intel likes to optimize for brain damaged single threaded loads, if that's what you've got then go Intel. Better, fix your software. I'm buying AMD the next time around because it more than addresses my needs for significantly less money.
"FMAC" operations are standard floating point operations
No. Multiply-accumulate is a specialized instruction whose main purpose is to preserve intermediate precision to better support solving ill-conditioned sets of equations and the like. It is not necessarily more efficient than a standard two instruction multiply add. Suggestion: consult Wikipedia before trotting out your ignorance.
Meaningless buzzwords can frabjulate the frobozz nicely!
Meaningless only to a dog watching television, such as you. To spell it out: intelligent task scheduling can pair integer-intensive tasks with MMX tasks on the same core. The Linux kernel already includes many similar, core-aware scheduling optimizations.
What Apple did was finally admit to itself that it was incapable of developing a reliable multitasking OS in-house. Taking the plunge and adopting open source Unix is arguably the least evil thing Apple ever did.
Unfortunatly, IOS && OSX of today are greatly modified versions of BSD, with a different kernel and apis. Macs were a great unix desktop ten years ago, now they just kind of blow. Linux is the only way to go these days for Unix && desktop.
Strange what you say there, considering that MacOS X is actually POSIX certified, and Linux for obvious reasons isn't. If you want Linux, use Linux. If you want Unix, Linux isn't Unix. Never was. Never will be.
A grain of truth always makes for better fud and rubbish, hmm? Vendors of Unix-like systems such as Linux and FreeBSD do not typically certify their distributions, as the cost of certification and the rapidly changing nature of such distributions make the process too expensive to sustain.[18] So... FreeBSD, commonly considered "real Unix" (directly descends from the Bell Labs code base) is not Posix-certified? What does that tell you about Open Groups pricey rubber stamp? Here is the truth: FreeBSD and Linux conform well to Posix, both take binary compatibility and source compatibility very seriously, and together lead the way in evolution of Unix. While not carrying the rubber stamp. That is life in the real world.
Seems to me, you pulled the "going rate" for a FRAND patent out of your butt.
For once, we have a judge who just wants to do her job.
Lucy Koh is another judge who apparently just wants to do her job, but as an Apple employee.
Ooh, some Apple spinmod did not like that post, oh dear.
Motorola wanted more then 10x the going rate for the FRAND patents.
Citation please.
Apple did not make the cash pile they are sitting on because of the patents.
Perhaps that has something to do with the fact that Apple lost more than $100 billion of market value in the last seven weeks, after becoming a patent troll.
For once, we have a judge who just wants to do her job.
Lucy Koh is another judge who apparently just wants to do her job, but as an Apple employee.
apple doesn't have much FRAND patents but they have a lot of non-FRAND patents that others want.
That is not clearly true at all. It is more accurate to say, Apple says it has a lot of patents that others want. Case in point: the bouncing scroll patent. Invalidated.
My feeling: Apple has a bunch of junk patents but is skilled at gaming the courts.
Regarldess of the merit of your speculation on the importance of rounded corners relative to the functionality of a phone, it would seem that the legal world is now coming around to the belated realization that rounded corners are not actually something you can patent.
Apple was a thug long before Steve Jobs declared thermonuclear war, but Jobs was a master at concealing that. Tim Cook isn't. This is one of several reasons that it is just wonderful for the rest of us that Tim Cook now runs Apple.
should influencing the future of technology be Slashdot's main mission?
Maintaining as high as possible a concentration of clueful posters and interested readers should be Slashdot's main mission. Then influencing the future will take care of itself.
If you buy a 12 or 16-core CPU, it's not because you want your Facebook page to load faster. It's because you have some serious parallel workload to process, likely involving a lot of calculations.
What kind of calculations? For many floating point intensive applications, such as gaming or video encoding, the GPU will be doing the vast majority of your floating point calculations, freeing up the CPU cores for fast integer logic. Many parallel applications, like compiling, do little to no floating point.
6200 series have shared FPU (floating point unit). Which means that there are less FPUs that there are processing cores. To multiply two floating point numbers cores are waiting in a queue until FPU is free to use, this happens when all cores are calculating at the same time. If you are doing intensive calculations this is going to be slower than if you used 6100 series. 6100 series have dedicated FPU for each core.
The transistors saved by sharing the FPU are returned to you in integer cores, arguably better use of the transistor budget. Here's a more nuanced view:
Looking at the results for the two core, four thread i3-3220, the picture is more interesting, and dare I say surprising. At one extreme, the A10 is 12% worse than the i3 in an Excel test. At the other end, it outperforms the i3 by around 42% in TrueCrypt AES Encryption. Overall, the A10 shows a 4.4% performance boost over Intel's i3-3220, a figure that is close enough to call the two equal on a diverse CPU workload. This surprised me due to the fact the i3 showed very good single-threaded performance and the A10/A8 only have 2 FP units.
how about 6300 series, is there a dedicated FPU?
Steamroller apparently only shares the MMX/FMAC unit, while standard floating point operations are per-core. Intelligent scheduling could optimize this arrangement nicely.
On the whole it looks to me like I'll be AMD again for my next box. Lots of cores for little $$$, yummy.
Knife to know you, Sharp, it's been a slice.
So what happens next quarter when Apple has had yet another 27-40% growth spurt, fueled by a large number of new products recently released?
What happens if Apple has a shrink spurt instead?
I doubt that would happen, historically the Japanese and Koreans haven't exactly been bff....
Oh how true, whereas America and Japan have always been best buddies, I totally get your point.
perception is what matters in stock valuations, especially a stock as widely held by "enthusiast investors" as AAPL
Are those the same "enthusiast investors" who drove AAPL down $120 in the seven weeks since the iPhone 5 introduction?
There aren't that many places to buy high quality LCD displays in volume and Apple just burnt its bridges with Samsung. I say, with Sharp teetering (no doubt in part because of Apple stinginess) LG is in an excellent position to raise prices.
And by the way, if your report is correct then it is Windows that is broken. NTP works the same way for Windows and Linux. On the other hand never mind, I think you're just pulling facts out of your behind anyway.
All of the clocks on our domain are sync'ed apparently that's a "feature" of Active Directory. Linux doesn't apparently respect that so when our file server somehow got a couple minutes out of sync with our AD, AD decided that it would just lock out access.
You seem to lack any notion of ntpdate and the fact that it is standard on all Linux server installs. Is that hair I see growing on your feet?
They could specify the exact wording, location and size of the ads.
They better not forget to specify the color scheme too, because Apple used light grey on white last time.