I had expected NVidia to do something.
AMD and Fusion. Intel with AVX(Larabee, et al.) and integrated video. NVidia had to either develop its own processor, buy a design/company, or fight a battle in the future it would likely lose … not due to the quality of the competitors or their parts, but simply because the deck was stacked against it.
Their direction is interesting. Going ARM and a fusion like thing as a CPU + GPU (though I doubt they will call it an APU … they are all about the APU … where A==G).
Whats interesting is that this is a RISC design, decidedly non-x86. This is a gamble. Or a gambit. not sure which. I want to think about it more.
Past history (never a great indicator of future predictive power) does suggest (if you violate the above canon and in fact use it for prediction) that this direction will not succeed.
Of course, then you have the fact that this processor could potentially run iOS, and Android, and say Ubuntu or some other Linux distro, and … and windows too.
Maybe I’ll write a longer piece for another venue, analyzing this.
Anyone still running Windows NT ia64? Or on an Alpha platform?
Binary compatibility has been a major issue for some time. I am not sure if or how they will address it. Would be curious to find out more.
We’ve heard the RISC vs CISC battle cry before, and ya know … CISC won.
Then again, we also heard the SMP vs DMP (shared memory vs distributed memory), and if you asked me 5 years ago, I would have said distributed memory has won. But nothing is forever.
The only thing you get by not changing, is stagnation. NVidia’s point here is that processor architecture has stagnated to a degree. Its x86 + 64 bits.
Now if they buy Tilera too … nah … lets not go there.
I want to think about this more.