(Page 2 of 2)
Intel is working on another SoC radio that operates in the 60-gigahertz range--a range for which no standards yet exist. At this frequency, the device could send and receive wireless data rates of up to three gigabits per second, Krishnamurthy says. This would make the platform ideal for streaming high-quality video and instantly syncing a collection of computers, mobile phones, and other gadgets.
Another paper that will be presented at the conference details an energy-saving graphics processor targeted at mobile devices. It processes multiple pixels at once--something that only power-hungry desktop chips are capable of today. Krishnamurthy says, "We want to bring this capability to mobile devices and make them in our next gen of SoC." He adds that traditional techniques don't work well at low voltages, but Intel engineers tweaked the design to drastically reduce the voltage required.
About an on-chip digital temperature sensor, Krishnamurthy says, "We'd like to be able to sense what's going on at various spots on the die." This is important because certain parts of the chip heat up faster than others do. A future chip could be smart enough to off-load some tasks to different parts of the chip so that it doesn't suffer heat damage.
Tom Halfhill, an analyst at research firm In-stat, says it is not surprising that Intel wants to enter the market for smaller mobile computers with its own SoC architectures, given the growth that it has seen in recent years. "You're seeing a second generation of personal computers: smart phones, netbooks, and these mobile Internet devices--personal computers you carry on your person," he says. "That's where the growth market will be in the future, and that's where Intel hasn't traditionally had good solutions."
Halfill says that Intel will need to catch up quickly with ARM, which supplies chips to the majority of the mobile market. "Intel is late to the game in some cases," he says. "ARM was the first for low-power, embedded chips."
But Krishnamurthy is optimistic about Intel's chances of taking the lead. "We see no show stoppers as we proceed with our goals," he says.
Researchers have revealed details of China's latest homegrown microprocessor.
With 45 million transistors and energy-saving features, the Atom processor could usher in a whole new era of mobile computing.
Its silicon laser could mean a solution to one of the great challenges facing the semiconductor industry: how to move data fast enough to keep up with tomorrow's ultrafast computers.
I entirely agree with you on this one. Companies such as Intel are in an extremely volatile industry. Infact all companies in the technology industry are at the risk of losing money, if they don't keep up with new developments. The companies must also keep an eye on their sales price. Afterall, this is a competitive industry.
There are too many articles in Technology Review which present large corporations like Intel as innovators. This is like giving millions of dollars of free advertising to Intel Corporation.
Other companies developed high-speed, low-power processors years ago; the ARM RISC processors, among others. Intel's processors dominate the market because of the Wintel monopoly, not because of any technical merit.
Manufacturing in the United States is in trouble. That's bad news not just for the country's economy but for the future of innovation.
This document is part of the “How-To Guide for Most Common Measurements” centralized resource portal. This tutorial provides a detailed guide for measurement and device considerations to take temperature measurements using thermocouples. Get an introduction to thermocouples, which are inexpensive sensing devices widely used with PC-based data acquisition systems. Also review some specific thermocouple examples and learn how thermocouples work and ways to integrate them into a data acquisition measurement system.
View full PDF >
Listen to story > 
Our list of the 50 most innovative companies, including the following:
On Twitter
Become a Fan on Facebook
Subscribe to the Feed
Mapou
357 Comments
Intel's Coming Disappointment
Nice article. It seems that Intel really wants to introduce a line of GPGPU-type mobile processors. It is taking essentially the same approach to SOC processor design that everyone else (ARM, Nvidia, etc.) has taken or is thinking of taking. By this, I mean that Intel's engineers have convinced themselves that the future of mobile computing is tied to heterogeneous multicore processors. This is a grave mistake, in my opinion, because these hybrid monsters are a major pain in the arse to program.
Everyone can sense that the industry is ripe for a major paradigm shift. This means that the old players should tread carefully. Very big money is at stake. A single astute competitor that comes out with a low-power and easy-to-program homogeneous multicore processor is all it will take to blow everyone else out of the water, Intel included.
Any universal multicore processor, i.e., one that is based on a universal programming model and that can be programmed for different tasks (graphics, signal processing, general computing, etc.) without jumping through hoops, will transform the market overnight and force existing processor technology to go the way of the buggy whip and cloth diapers. When that happens, there shall be much weeping and gnashing of teeth among those who failed to heed the writing on the wall.
How to Solve the Parallel Programming Crisis
Reply