You might remember just a couple of months ago Intel made an announcement that it would allow certain custom Atom-based designs to be built at TSMC – a drastic move on its own right to allow another fab facility to build Intel design processors. Today’s news is even more earth shattering in some ways – Intel is basically saying that they will start opening up the “on-chip bus” and open the “shared, modular 32nm library” that could allow third parties to build custom ASICs with Intel x86 cores on them and then fabricate them on the Intel 32nm production lines.
Image from Semiaccurate.com
The reasoning behind this drastic move for Intel is pretty obvious – if Intel wants to move into the world of cell phones and other super-low-voltage markets, the need for custom silicon for each product is more and more apparent. Intel for years has been pushing that it can use existing or only slightly modified core architectures using x86 ISAs for products like MIDs, phones, etc. But ARM’s dominance in that field has never really been matched and it looks like Intel is finally admitting that the task-specific architecture model has benefits.
Charlie poses a great question in his article:
If we look at products like NVIDIA’s Tegra, it’s strength lies in the power it can provide to a product at minimal battery life while implementing typically x86-based functions like HD video and now Flash acceleration on an ARM-based design. Intel definitely sees these products as competition and if we know anything about Intel it’s that the company can dominate in just about any field it really sets its mind to.
If you are looking for more discussion on this topic, you should stop by PC Perspective around 10pm ET for our live podcast recording (will be posted on the front page as news) as it will undoubtedly be on the table for debate. Or you can subscribe to the podcast now and get the audio sent your way in the morning.
Charlie poses a great question in his article:
So, with a new, more open and friendly Intel, the question is, does
Intel drive power down to phone use levels before ARM gets the drivers
and Firefox plugins that x86 has?
If we look at products like NVIDIA’s Tegra, it’s strength lies in the power it can provide to a product at minimal battery life while implementing typically x86-based functions like HD video and now Flash acceleration on an ARM-based design. Intel definitely sees these products as competition and if we know anything about Intel it’s that the company can dominate in just about any field it really sets its mind to.
If you are looking for more discussion on this topic, you should stop by PC Perspective around 10pm ET for our live podcast recording (will be posted on the front page as news) as it will undoubtedly be on the table for debate. Or you can subscribe to the podcast now and get the audio sent your way in the morning.