Russian Startup Develops x86 Emulator for ARM, Could Be Great for Windows RT Devices
Emulators are a fantastic tool to get antique Atari ST, Amiga and C64 games to run on your PC. In the future, there may be an emulator that is solving ARM headaches.
It is entirely unclear how consumers will react to the availability of Windows RT notebooks: systems that will be running on ARM architecture rather than Intel or AMD x86, such as the Surface RT. Even if it is Windows that will be showing up on that screen, you won't be able to run the typical x86-designed application on these systems, which could be a significant hurdle for the adoption of ARM in this space.
A Russian startup said it may have a solution for this problem. Elbrus Technologies reportedly developed an x86 emulator that runs on ARM systems and delivers 40 percent native ARM performance for x86 applications. That may not be enough to run Photoshop, but Elbrus noted that it could hit 80 percent by the end of 2014.
Realistically, 40 percent or even 80 percent is just a band-aid, workaround and just a short term solution for those who continue to rely on Windows applications on an ARM device. For ARM, this is not quite the desired outcome of its march into the x86 arena: In the end, it will have to convince software makers to build fast and efficient applications for ARM, and not x86 applications that run with the help of an emulator on its notebooks.

Not I, said the Apple, I'm interested in keeping my platform 100% closed.
Not I, said the Microsoft, I'm working on forcing all my customers to a closed platform.
Android will most certainly get this, though- that much is certain, but Android is not where this is needed.
BTW, now that Microsoft seems to want to become the next Apple, I was thinking of conjunctive names for them; how about Microple or Micrapple?
They have it already: it is Micro and it is Soft, just like Bill Gates' thingy!
Microdrone? Micrap? Microverpriced?
I would be cool to have a win 98 and xp emulation on a phone all the clasic at your hand.
I'll wait ans see if we go anywhere new this time.
I can see Apple buying this company (just like the PPC emulator Rosetta) and using it for a next-generation 1lb macbook air. A new SoC using the ARM Cortex A15 cores (like the Samsung Exynos 5x00) will easily be competitive with Intel's Atom for performance and use less power.
"We've been down this road before with Intel's Uranium processor and DEC's Alpha processor (and others)."
Apple's transition from PPC to X86 Rosetta seemed to be successful...
It's no whare near as energy efficient as arm, but intel will get there.
The real reason for this is keep choice - with AMD not competing with INTEL in the desktop segment it would be good for the giant to have an equal opponent. If they will be left alone (all thing seems to be heading that way) they we will pay for it - and progress will be halted (slowed down).
FPGAs are great for running arbitrary logic in a shallow and wide fashion. They're awesome for re-programmable function generators, oscilloscopes, bitcoin mining, etc... but they're absolutely horrible for modern general purpose processors which are deep and narrow by design.
I have several Altera dev boards sitting on my desk and even the most powerful of them can at best run ARM Cortex A9 with most of them running M1 or even less.
Soft instantiating instruction set processors on FPGAs (using the FPGAs logic elements) is so inefficient that most manufacturers are now including hard-instantiated ARM processors (a fully fabricated ARM processor) which is directly connected to the FPGAs routing fabric.
That was a little different---the x86 chips available at the time were a good deal faster natively than the then top of the line Mac PPC chips, thus the performance hit from Rosetta was not as bad as it would it have been if the x86 chips were slower natively than the PPC chips, which is the situation ARM chips are currently. In another words, 40% of ARM native is not 40% of x86 speed...its much slower than that.
is Intel's advanced power gating, but this only helps lower the power usage when the chip is idle! At full load, less transistors = less power used! With too much Intel inside, there will be too much money out of everyone's pockets!
Unless there are serious architectural deficiencies, more transistors = more work done at full load. The fabrication process doesn't have as much of an impact as one might think. Intel employs a lot of unique and interesting techniques to maximize instruction throughput, at the cost of power but not power efficiency or die area. Lower die area does not necessarily imply lower power consumption because logic networks that aren't being switched consume very little power.
Some architectures are inherently biased towards floating point instructions while others are inherently biased towards general purpose integer instructions. Intel owns the general purpose integer field, IBM owns floating point, while AMD owns the floating point / mm2.
http://www.youtube.com/watch?v=H9NeRQjGwfs
I understand x86 can't work on RT (maybe it can now). But .NET should run fine.
So, if someone makes some little Notepad++ in .NET for $5, it's illegal to sell it outside the App store? - even if it should run fine on RT or x86 with .NET installed?