Once the discovery app determines the hardware available, it downloads different optimized blocks for the different hardware elements, creating the optimal software.
Wow, if this can be really done, that is awesome!
I remember back in the old days when PC video game developers wrote "to the metal," which made it difficult to transport games from one graphics chip platform to another. In those days, literally, everybody talked about the need for Open API for graphics. (Well, I am talking about the days before OpenGL became virtually accpeted by everyone...)
But in the world of mobile, as you pointed out, with all the fragmentation going on with Android, and different CPUs and not to mention multi-cores, it appears that things have gotten even more complicated.
Absolutely Junko - that's what is really important about what Imagination is saying - there's lots of research opportunities in this area to make it work, but the organisations that can would have a signficant advantage.
I am wondering how do FinFeT influence SOC designs, does the properties of the transistor vary so much that it alters the way SoC design has to be approached, do the fundamental design techniques of SoCs remain intact.
the challenge with FinFET is how to make all the non logic transistors. RF, analog, mixid signal, high voltage etc. FinFET has problems.
Intel for its new low power Quark (x86 CPU) has a corporate mandate to develop all the SOC building blocks on a planar transistor due to the high level of integration. Also the planar is lower cost and intel has been loosing recent bay trail sockets to media tech and Qualcomm on price. (both much much lower ).
Does anyone think that the non logic portions of a chip, that is the RF, analog and other building blocks will be developed with FinFet and not with the normal planar transistors, I see already analog has appraoched 45 nm designs, how many technology generations is analog lagging with respect to the digital counterpart. How is the FinFet influence so far on the SoC designs all around.