Nvidia Waiting In the Wings In FTC-Intel Dispute 143
The NY Times has a Bits Blog piece speculating on some of the fallout if the FTC prevails in its anti-competition lawsuit against Intel. The Times picks out two among the 26 remedies proposed by the regulator, and concludes that they add up to Nvidia being able to license x86 technology. This could open up 3-way competition in the market for combined CPU-graphics chips. There is a good deal of circumstantial evidence pointing to the possibility that Nvidia has been working on x86 technology since 2007, including the presence on its employment rolls of more than 70 former Transmeta workers.
Re:Ugg... (Score:5, Informative)
IBM gave up?
16-core 4GHz processor modules would like to have a word with you.
http://en.wikipedia.org/wiki/POWER7 [wikipedia.org]
Re:Ugg... (Score:4, Informative)
VIA is still at it, they're just attacking the Atom end of the market, now. This is where they were before Atom came along, but they have been developing newer processors.
Re:Ugg... (Score:3, Informative)
It seems you forgot ARM processors...this tiny, insignificant part of the market which, by now, perhaps ships more CPUs annually than Intel has ever produced.
Re:Ugg... (Score:3, Informative)
Re:Is x86 shit? (Score:5, Informative)
There have been quite a few different architectures, all supported by Microsoft and Windows.
http://en.wikipedia.org/wiki/PowerPC [wikipedia.org]
http://en.wikipedia.org/wiki/IA64 [wikipedia.org]
Even though Microsoft abandon PowerPC long ago (XBox excluded), they still support IA64 to this day.
The biggest problem hasn't really been vendor support, but compatibility. PowerPC held Apple back for the longest time because users had no good solutions for running x86 Windows apps when needed, whereas now they have WINE and native booting. IA64, while having some x86 compatibility, does not have clear enough benefits for consumers, and generally runs existing apps slower.
Ironically enough, AMD pretty much killed IA64 and gave x86 a longer life when they came out with x86-64, thus cutting off Intel's attempts to replace x86. Smart business decision for AMD, but it hampered attempts to replace x86.
Re:Ugg... (Score:3, Informative)
Re:Is x86 shit? (Score:1, Informative)
While there's no absolute line between what counts as RISC and what counts as CISC, all x86s released today are more RISCy than a direct to silicon version of their entire instruction set. They really couldn't do it natively these days.
As for power savings running in the native RISC code or via a decoder, well the decoder is a very small percentage of a typical CPUs real-estate and it does more than just translate anyway (it's what allows for superscalar execution, etc. -even low power embedded CPUs have an instruction decoder of some sort these days).
There is a way in which the x86 does lose out to other architectures in terms of power efficiency though. It simply doesn't have the mechanisms to scale its speed up and down as required. eg. ARM has PLL registers and run levels to set its clock rate at run time and a way for interrupts to quickly turn the CPU on, execute and turn it back off again. The x86 will never have those features (kind of hard to put something like that in without breaking backwards compatibility).
Re:Competition is a Good Thing (Score:3, Informative)
Power is dead? Tell that to Nintendo, Microsoft and Sony, all of whom are using PPC chips of various kinds in their current generation consoles.
Cell is basically a PPC core with a bunch of specialist number-crunching coprocessors attached. And its by no means dead unless you consider the fact that a Cell CPU is found in every one of the 27 million and counting PS3 systems out there as being dead.
I will grant that PPC is dead as a desktop CPU with x86 being the only viable solution at this point for mainstream general purpose computers.