TI E2E Community (Beta)

Tools in 2020: Development tools are key to SoC implementation

Christy Brunton — Thu, 12 Nov 2009 02:40:00 GMT

Reid Tatge, TI Fellow

Customers need chips, tools and software that match the specific needs of their application. In addition, they need everything to be simple to design into a product, easy to program efficiently, ultra-low power and ultra-low cost, available early in the end-product’s preferred life-cycle and broadly supported by third parties. In 2020, I expect customer requirements to stay pretty much the same with the main difference in the underlying technology and how we develop it will be vastly different and far more complex than what we provide now.

From a hardware perspective, these future SoCs will bring multiple DSP and GPP cores together with custom hardware accelerators into a heterogeneous architecture loosely coupled with an asynchronous interconnect. In addition, these devices will have a non-uniform memory architecture and be designed for ultra-low power consumption.

Designing such complex architectures from scratch will no longer be feasible, both from a cost and time-to-market standpoint. Instead, devices will be designed using an iterative approach utilizing reconfigurable system modeling tools, such as G3-type compilation tools that support a rapid design methodology. Specifically, the topology of the SoC will be tunable according to the particular application domain at the individual processor node and memory subsystem level. The chief advantage of this approach is that it leads to completed SoC designs in months, not years.

The next challenge for SoC designers will be making them easy to program so developers can view the system as a loosely-coupled network of processors and be able to access the variety of available processing capabilities without having to involve themselves in all the low-level details that arise in multiprocessor architectures. Additionally, developers will need to be able to program in an HLL (high-level language) while achieving high-performance efficiency. Development tools for these SoCs will support program partitioning, system visualization, multi-core compilation and pre-hardware simulation as well as provide a reliable OS designed to manage the unique characteristics of multi-core architectures.

While it is difficult to anticipate exactly how these SoC devices and development tools will manifest, I can eliminate a number of “promising” possibilities:

• Large, monolithic, mega-CPUs: These fantastically complex architectures take years to define and tune, and then, at least another year to design. Their development environments are closed not for any proprietary reasons but because only the architects can program them.

• Symmetric networks of commodity GPPs: While an outside contender, these architectures solve difficult problems with more of the same. Eventually they collapse under the weight of messaging, data stitching and other forms of overhead.

• Any architecture dependent upon “Magic” tools: It would certainly solve many problems if development tools could generate great code for any arbitrary CPU as well as automatically partition inherently serial programs onto a network of processors. Such a panacea will not be available by 2020.

About the author

A TI Fellow, Reid Tatge leads the development of TI’s compiler technology infrastructure and products in the software development groups. He also works closely with TI and customers to define new DSP architectures.

Bluetooth low energy RC car

Karl — Wed, 11 Nov 2009 18:08:00 GMT

Hello again,

I promised you a video of the Bluetooth low energy remote control car a couple of weeks back, finally here it is: http://e2e.ti.com/media/p/48155.aspx. This is the same one we showed in Munich.

Sorry it took so long!

We won't have the car at next week's event in Beijing unfortunately, due to the complications of shipping stuff like this internationally. However, for those who can come, we'll still have a lot of interesting information. Please stop by and say hello in person if you are there!

Best regards,

Karl

A first taste of Éclair on Droid

Eric Thomas — Wed, 28 Oct 2009 19:56:00 GMT

As my colleague Brian Carlson touched on in his blog post, Motorola's Droid launch marks a huge milestone for the mobile industry. As the first device based on Android 2.0 (Éclair), Droid is a testament to how far the mobile market has progressed with open source solutions.

I've been using Motorola's Droid for a few weeks now and am impressed with the advancments in the mobile user experience. These enhancements - such as the social networking and sync integration - are due in large part to the new features available with the Android 2.0 release. More than any specific feature, however, the radical improvement in overall responsivness and performance of the platform has changed the way I use Android, and the OMAP^TM 3 processor benefit to Android is salient.

As a Marketplace addict, I am routinely keep between 10-15 service applications running to sync with facebook, twitter, and multiiple gmail and IM accounts to ensure that I am the first to know if one of my friends is compelled to share a random, but apparantly pressing, thought. This level of multitasking rendered previous Android phones nearly unusable. If you disagree, just try adding corporate mail via activesync into the mix. I'm running this same profile of applications and more on Droid without sacrificing core performance or limiting the range I roam from my wall charger.

Supporting capabilities like advanced media processing and next generation 3D graphics, the OMAP 3 processor is optimized to address Android 2.0's unique features and enhance its capabilities. The two work together as a tight-knit team to support the features that consumers crave most on mobile devices: High quality video from YouTube, multitasking social networking integration, a more intuitive and attractive user interface, enhanced quality imaging...the list goes on.

We're excited to be a part of Android's continued advancements in the market, and to lend our support to continued open source milestones. In your dreams of electronic sheep, what will Android do next with more processing power?

A big Texas “congrats” to Motorola on their Droid Launch!

Brian Carlson — Wed, 28 Oct 2009 16:12:00 GMT

Our TI team would like to congratulate our friends at Motorola on today's official Droid launch. We're thrilled to be a part of this game-changing device, which BoyGeniusReport called "the most advanced Android device on the market as far as specifications go." Packed with top-notch technologies, Droid promises a new world of speed, performance and connections for the world's mobile consumers. We're especially proud of the TI chips helping Droid fulfill these promises:

OMAP3430 processor produces intense performance: As Engadget's Chris Ziegler highlighted, "Android 2.0 [Éclair] clips along at a nice pace thanks to an OMAP 3 core." And, Android 2.0 is not the only Droid component that benefits from our proven OMAP^TM architecture. The included OMAP3430 processor runs Droid's applications at lightening speed, supports multitasking capabilities and preserves battery power. Built with solid a solid ARM® Cortex^TM-A8 foundation, OMAP 3 processors stand out from the crowd of ARM-based processors by providing an optimal balance of CPU and multimedia performance along with extended battery life enabled by our SmartReflex^TM power and performance management technologies. These benefits shine on the Droid.
WiLink^TM 6.0 solution connects the dots: OMAP processing power pairs with connectivity support from our WiLink 6.0 solution to bring the Droid's interactive features to life. A single-chip mobile WLAN, Bluetooth® and FM solution, the WiLink 6.0 product makes quick connections, rich data transfers and other connected applications a snap to use on the Droid. What does this mean for consumers? Speedy new ways to stay in touch, update social media sites, share content and more. The WiLink solution's highly-sensitive coexist capabilities ensure robust connections that are best suited for each situation too, whether via Bluetooth or WLAN. Talk about smart!

Overall, the Droid is an impressive milestone that shows how industry leaders are not just dreaming up great new ideas - they're delivering them for consumers to enjoy first hand.

What makes you excited about this reality? Let us know what you think by posting a comment here, or connecting with us on Twitter: @TXInstruments.

ARM Techcon3 - Showcasing ARM and TI Success and Providing a Glimpse into the Mobile Future

Brian Carlson — Sat, 24 Oct 2009 14:16:00 GMT

I always enjoy attending ARM’s annual technology conference (now called Techcon³) to meet with the community of key technology providers that drive today’s hottest new products and will help enable our mobile future.

The “cubed” in Techcon³ refers to focus in three key areas including “Energy Efficiency”, “Internet Everywhere" and “MCU and Tools”. Some of the key topics that really stood out during this conference were addressing the drive for higher, energy-efficient performance, emerging internet and web browsing technologies, the tremendous value and success of Cortex-M3 microcontrollers, innovations in user interfaces and more sophisticated development tools.

Glimpses of the mobile future included a drive to 28/32 nm process technologies, multi-core ARM / SMP, high-performance graphics, a future revolution in mobile memory technologies that bring tremendous bandwidth at manageable power levels, momentum of hypervisor solutions and the enabling of lower-tier handsets with the new Cortex-A5 core.

TI had a strong presence at this conference, announcing over 30 new ARM devices with the expansion of our Stellaris product line and introducing our new Sitara product line , presenting papers and product updates, participating in panels and showcasing the some of the latest products based on TI devices. Many of the ARM partners also showcased their technologies on TI platforms such as the very popular OMAP Beagleboard and EVMs – so TI products were very prominent reflecting our strong ecosystem support.

Did you know that TI is the leading ARM silicon partner, delivering over 5 billion chips ARM-based products out of a total 15 billion ARM shipments? This amazing fact helps drive home the power of the TI-ARM relationship and the connected community that helps us offer the broadest range of ARM-based products from $1 to over 1 GHz, along with an impressive list of TI first-to-market successes with ARM technologies.

If you didn’t attend the show, I recommend you check it out and consider for next year. It really provides a good perspective on where the industry is going with focus on the most pertinent topics that will help drive our mobile future. You also get to see and play with a broad range of the latest

ARM-based gadgets which is always fun!

Bluetooth low energy conference in Munich

Karl — Fri, 23 Oct 2009 14:32:00 GMT

Hello again,

Sorry for not posting to the blog for so long, but we are very busy here in BLE-land these days...

I attended the Bluetooth low energy conference in Munich on Tuesday, and wanted to share my impressions of this event. Attendance was good, somewhere between 120-150 people. Even more important, the questions asked by the audience over the course of the day were of high quality and indicated that the participants were both knowledgable and interested in what BLE has to offer. The conference program included both some presentation on BLE's features and capabilities as well as some interesting presentation from companies about how BLE was useful in their applications (sports equipment, automotive and others).

At the end of the day, there was some time allocated for so-called "elevator pitches" from vendors. Brian and I did a brief overview of TI's BLE offering, highlighting a complete BLE solution, high integration and good co-existence as key reasons why to choose TI. In addition to this, we had some fun showing off our BLE remote-controlled car. One of our engineers took the prototype of our $99 kit and used this to modify a remote controlled car so that it is controlled using Bluetooth low energy. In the car, he modified the USB dongle to control the car's servos using the PWM outputs on the CC2540. The sensor board is used as a remote control by using data from the on-board accelerometer to control the car by tilting the board forwards and backwards to accelerate and brake, and sideways to steer. Quite intuitive, and you get the hang of it quite quickly. The remote control is powered from a CR2032 coin cell, and to our knowledge, this is the first time anybody has shown a demo in public of BLE running on a coin cell (we did a simpler demo back in April, but due to the Japanese radio laws we couldn't show it live, but had to show a video of it instead).

I'll try to get a video of the car demo posted as soon as possible.

Best regards,

Karl

ARM technologies: Resting at the heart of our favorite devices

Melissa Haddad — Wed, 21 Oct 2009 22:03:00 GMT

As consumers, you and I both realize that high-tech devices – from smartphones to heart monitors – continue to revitalize the way we interact with the world, share information, capture memories and experience life. With each technological milestone, engineers and manufacturers alike continue to push the innovation envelope to keep products unique and differentiated. ARM® technologies spur that innovation, and rest at the heart of so many devices that we rely on each day.

This is the focus of the ARM TechCon 3 event, happing this week in Santa Clara. If you are at the show, swing by Ballroom G tomorrow, October 22, at 2:00 PM Pacific to hear about how ARM technologies help designers achieve the scalability, software availability and product differentiation necessary to “wow” our senses and keep us fully connected. The below-listed panelists – Will, Jean, Gene and Tom – will examine how ARM technologies meet market needs today and provide a foundation for future support with high-performance, low-power solutions that drive various markets forward. They’ll answer questions after the session, and would also love to hear about how ARM technology-based solutions help you win. If you’re not at the show, we still want to hear from you. Comment here or send @TXInstruments a Tweet to let us know how ARM solutions help you succeed!

ARM TechCon 3 panel: “Meeting your goals and solving design challenges with ARM technologies”

Date/Time: October 22, 2009 @ 2:00 PM Pacific
Location: Ballroom G, Santa Clara Convention Center
Panelists:
Will Tu, embedded segment market manager with ARM
Jean Anne Booth, worldwide director, Stellaris® MCU marketing and customer-facing engineering with TI
Gene Frantz, TI futurist and principal fellow
Tom Starnes, chief analyst with Objective Analysis (moderator)

Posted by Missy Haddad, GolinHarris Dallas

Working towards easier, speedier mobile content delivery

Dave Lacinski — Thu, 15 Oct 2009 18:38:00 GMT

Yesterday, the WiFi Alliance introduced WiFi Direct™, the name for its peer-to-peer technology. Though the specification is nearing completion, we can begin to imagine how such an offering will impact consumers’ interactions with each other and with data. Imagine being able to easily share photos of your summer vacation with your friends – all via a direct WiFi connection on your mobile device. The industry is getting closer to making these real-life capabilities.

While WiFi Direct may still be in development, the excitement and promise for this technology continues to grow. TI continues to support the latest mobile wireless LAN specifications in our WiLink™ products and will have our WiLink™ 6.0 certified to the new specification when it is ready in mid 2010. We look forward to seeing the new and unique use cases that will emerge with the availability of this technology.

Modules - Driving Innovations and Mobile Momentum

Brian Carlson — Tue, 13 Oct 2009 23:08:00 GMT

We recently blogged about how Dell enabled their Latitude ON "instant access" feature for laptops by leveraging the Texas Instruments OMAP3430 system-on-a-chip. The Dell approach was based on a "system on a system" modular approach to provide the ultra-low power, high-performance of the OMAP3430 mobile processor. You can see the Dell ON module featuring the TI OMAP device at 1:14 minutes into this video.

A modular design approach offers many advantages to product designers such as faster time-to-market, lower cost printed circuit board (PCB), lower cost manufacturing and ease of upgrades for customers. The complexity is managed by the module, and the product designer only has to focus on the external interfaces.

You can leverage OMAP modules from a several vendors to go to market quickly with OMAP 3 technology. These vendors offer a variety of sizes and features, so you can choose one that best fits your application.

One of recent OMAP modules that is notable for mobile designs that are space-constrained is the Logic OMAP35x Torpedo SOM. This dime-sized module could quite possibly be the smallest, most powerful computer in the world!

Links to some of the OMAP 3 modules available from other vendors is below. Various names are used to refer to these products such as "system-on-module" and "computer-on-module". Please comment if there are others, as this list will continue to grow along with the strong industry momentum of OMAP 3 processors.

Modules like these will enable many new, innovative products that will continue the mobile device momentum.

Direct Insight SwiftModule-OM

gumstix Overo Earth

Logic OMAP35x Torpedo SOM

Magniel OMAP-Mod

Variscite VAR-OM35xxSBC

NOTE: The product names used in this post are for identification purposes only. All trademarks and registered trademarks are the property of their respective owners. Any web site you visit by a link from this site is solely the responsibility of the other party providing the site.

Thoughts from CTIA IT: Changing lives and upping the ante

Melissa Haddad — Tue, 13 Oct 2009 15:32:00 GMT

I had the pleasure of traveling to San Diego with TI’s Seshu Madhavapeddy for CTIA Wireless IT and Entertainment last week, and it was quite a trip. We spent all of Thursday at the exhibit hall, exploring company booths, listening to key notes, and speaking with some of our great media and analyst friends – including RCR Wireless, iSuppli, Unstrung, IDC, CCS Insight, PC Mag, and more.

As mentioned in a previous post, Seshu also participated in a smartphone panel at the show (above picture), together with executives from IDC, Qualcomm, HTC, Motorola, Nokia and Samsung. I’m sure you can imagine just by this list that the discussion was extremely dynamic and chalk full of enlightening tid bits. Here are some key points – among others – that really got the panelists and audience going:

There isn’t an “all-in-one” answer for any mobile components – people want devices that fit their personal needs
Lines between smartphones and feature phones are blurred – today’s features will migrate to lower-tier markets, and tomorrow’s will lead the next iteration
A big ecosystem is important, and it’s the rich base of developers that keeps things ticking
Processing power is a key, but not without the “golden rule” – power
Industry needs to continue fueling a transparent Internet experience
“Mobile” is the business of bringing the wisdom of man – from knowledge to content to communication and beyond – right to the palm of your hand
People in emerging markets are experiencing life through a whole new lens, thanks to mobile phones
Going “green” is everyone’s job [Fun fact: Nokia now uses 12,000 less trucks to ship end products because of tighter packaging]

After listening to the executives exchange perspectives on these and other topics, I couldn’t help but reenter the main show hall with a different mindset – one in which I noticed more than ever how truly revolutionary our mobile world is. The show floor was full of professional consumers snacking on mobile content – checking calendars, snapping pictures, breezing through top news stories, typing another email. More importantly though, it was full of the people and companies that make the above-listed trends and topics come to life. As Dan Jones wrote in an Unstrung post, TI’s OMAP processors – combined with the company’s full arsenal of other semi components – play a huge part in this revolution, and “the next generation of chips will up the ante.” What a thrilling adventure to play a part in.

Were you at CTIA this year? If so, post a comment here and let us know what you thought!

Posted by Missy Haddad, GolinHarris

Flash 10 moves to the mobile environment

Amy — Thu, 08 Oct 2009 22:06:00 GMT

This week we've seen a number of unique and exciting things relative to the evolving mobile user experience. Most notable is an announcement from Adobe making Flash 10 available in the mobile environment. This version for flash was initially released for the PC/x86 architecture, but now is available in the mobile environment. With this announcement, Flash 10 capabilities are available on a range of ARM processors, everything from ARM11 and beyond, including the ARM Cortex-A8.

TI has been supporting Flash technology for some time, delivering Flash 9 support on the Nokia N800 series. Earlier this year, TI announced plans to support Flash 10 and we are now delivering on this promise. In addition to enable Flash 10 on the OMAP platform, TI will also enable hardware acceleration of video and graphics.

Adobe demonstrated the first mobile Flash 10 experience on a Palm Pre running on an OMAP 3 processor, showing how the user experience will change when consumers have access to full Flash content. In addition, Adobe also demonstrated Flash 10 on the Nokia N900, which also uses the OMAP 3 processor. While the initial N900 products will support Flash 9.4, Flash 10 support will be available via software update in 2010.

We are very excited to see full Flash support entering the mobile environment. With this capability, the Flash experience on a PC will be identical to that of a mobile device. As an added benefit, these new devices not only get the full Flash experience, but they will also deliver the power advantages for which ARM is known!

Making your phone do more

Amy — Tue, 06 Oct 2009 17:57:00 GMT

A good friend of mine used to work for a company who manufactures Hefty products. One night she asked, 'what would make our trash bags better?' While the conversation ended with each bag coming with a man who would take out the trash - a not so serious recommendation - the question was a serious one. What else do you want your product to do? At TI we are focused on bringing more capabilities to devices - whether its multimedia, connectivity, more performance or lower power - we have the solutions that are making our handsets even more of a life line than they were years ago.

We continue to pack these devices with consumer electronics quality features. Reiterating this trend, today we announced two new products to our OMAP-DM coprocessor product line, which now supports up to 20MP imaging and 720p camcorder capabilities. The imaging and video features are getting close to those that are available in standalone products today, and we believe this convergence will only increase in the future.

Seshu Madhavapeddy will comment on this trend and others during a CTIA IT & Entertainment panel later this week. The "Smartphone - The Impact of a Fully Functional Device" panel will be held on Thursday, October 8 at 1:00 PM Pacific and will include representatives from In-Stat, HTC, Nokia and Motorola. Seshu will discuss the importance of a system solution as handset manufacturers strive to do more with their mobile devices. Swing by if you're at the show - we'd love to see you.

Will my handset ever take out my trash? Probably not; but it does allow me to get trash pick-up times online.

A new Window of opportunity: Congrats, Microsoft!

Techsavvy — Tue, 06 Oct 2009 17:41:00 GMT

We at TI would like to congratulate Microsoft on the introduction of the new devices based on Windows Mobile 6.5 operating system (OS). We are excited by the capabilities of Windows Mobile 6.5 and how they will advance the user experience. With this latest version, consumers will have a dashboard-like experience with quick access to favorite applications, and an improved user interface for a better Smartphone experience. If you combine Windows Mobile 6.5 with our OMAP platform, the possibilities for sweet new application experiences only increase as OMAP capabilities are fully utilized on the latest OS iteration. TI is fully committed to supporting the Windows Mobile roadmap, including Windows Mobile 6.5, and we look forward to the new and innovative devices that will be built on this platform. Congratulations again!

Easily share TI stories and information with others

Jason Strimas — Wed, 30 Sep 2009 21:02:00 GMT

We've made it easier for you to share what you find on TI.com with your friends and colleagues. If you find something interesting and you want to share it then use the "Share" button in the top right corner of the page. Either rollover or click the button you'll be able to choose from any one of the options listed. Maybe you want to email the page, bookmark it in Delicious, add it to your Facebook page, or embed it in your blog.

Here are some examples of where you'll find the share button.

TI E2E Community

The share functionality supports embedded content. That means that you can share a video and it will play on the page that you add it to. The embed functionality only works with certain sites that support it such as Blogger, Wordpress and TypePad.

Check out all the places you can share TI.com pages and information with others. Why not give it a try and see how it works?

Heralding the arrival of Dell Latitude ON

Ramesh Iyer — Tue, 29 Sep 2009 18:07:00 GMT

Dell's Latitude ON is finally here ! At the heart of it is Texas Instruments' OMAP3 running Linux in tandem with a Intel processor. Blog and YouTube video from Dell at:

http://en.community.dell.com/blogs/direct2dell/archive/2009/09/28/latitude-on-arrives.aspx

Before you say "huh?" to this unique and revolutionary architecture, consider the problems compounding all traditional x86-based laptops/notebooks/netbooks:

1. Very long boot-up time (measured in minutes, not seconds)

2. Battery performance of 2-3 hours that is heavily dependent on usage. If you are like me, you should be lugging an extra battery pack but it also wreaks havoc on your arms

3. Infrastructure and administration costs associated with upkeep of corporate laptops. More on this topic in a separate blog on cloud computing and thin-client machines.

4. Losing all connectivity when the lid is shut

The ON architecture, available as a add-on, to Dell's Latitude E series and Latitude Z series product lines, resolves the aforementioned thorny issues with great gusto! Using TI's OMAP3 processor and Linux software, the platform is capable of:

1. Booting up in < 2 seconds = Instant ON

2. Manage all user data viz. email, calendar, contacts, browser, access to the Citrix client (in true cloud computing fashion)

3. Remains connected even with the lid shut without significant impact to battery performance = Always ON

4. Seamless synchronization of email through MS EXchange or POP

5. Delivering a true all-day battery experience at 17 hours on a 6-cell battery

It is very easy to market new products to the average consumer. But, it is a very difficult ordeal to market it to enterprise customers because they must be approved by corporate IT. Dell has, once again, demonstrated its leadership in the corporate enterprise segment with a platform solution that is also future proof as the industry moves towards true cloud computing.

Thin-client machines are here for real. OMAP delivers the best user experience, bringing together the best of two worlds: Blackberry-type functionality in a notebook form-factor product.

There's more innovation to come - stay tuned !

Intel Developer Forum 2009 - Day 1

Ramesh Iyer — Wed, 23 Sep 2009 04:37:00 GMT

My first reaction as I stepped out of the 1^st day of Intel's Developer Forum - no earth shattering news yet. But, nobody organizes developer forums like Intel, at least amongst their semiconductor peers. Expansive show floor layouts, innovative themes, conveniently scattered whiteboards throughout the floor for the geek in all of us to portray our expectation of a favorite gadget or of future technologies, and most importantly, well orchestrated keynotes. Your marketing machinery continues to impress us.

As we all expected, Moblin 2.1 for smartphones and MIDs is upon us. The MID and netbook categories are still alive despite all the press that MIDs have not become a commercial success and netbooks are nothing but lower cost laptops with minimal differentiation. But, we believe that Intel is really gunning for the smartphone segment - Moblin 2.1 clearly focused on social networking and intuitive user interface (e.g. concept of personalized zones, notifications panel, etc), a theme that was echoed repeatedly from Paul Otellini's keynote through the rest of the tracks. Still no answers on what this really does to the software development cycle but Moblin 2.1 is looking fancier and more rich compared to its predecessors.

A lot of the sessions that I attended focused more on their strategy for the server segment, a lucrative area for Intel, and much easier to retain leadership than it is to penetrate the CE and smartphone space where they don't have a huge presence. Of notable mention in the server space are two high-end micro-architectures, Nehalem and Sandy Bridge. Westmere is a member of the Nehalem family, which includes trusted execution technology and new instructions for AES acceleration. Security, under the guise of cloud computing, is finally emerging into its own. Sophisticated power management and remote management are two additional core competencies that were discussed in cloud computing environments.

Intel is taking a leadership role in pushing semiconductor process technologies, with plans to sample devices in 22nm in 2011 followed by 15nm in 2013. Their 45nm Lincroft platform (part of Moorestown) is being positioned for handheld products (e.g. smartphones) - not a surprise considering that the smartphone segment continues its relentless growth despite a weak world economy. What did surprise some of us was their first public acknowledgement of PoP (Package-on-Package) technology (where the processor and the memory are stacked on top of each other) - PoP's are slightly more expensive than non-PoP and only the silicon vendors like Texas Instruments have invested heavily in PoP. These are early signs from Intel that they are leaving no stone unturned to get a footing into the lucrative smartphone/embedded processing space.

Security, which until now has been relegated to curious academicians and cryptographers, is emerging into the limelight as an important technology piece for secure cloud computing based products, both on the server and client sides. Intel has embedded native instructions in Westmere to accelerate the performance of AES and key scheduling. In addition, their instruction set accelerates full disk encryption and application level encryption. There is no doubt that saving every MHz makes perfect sense on the server (which handles thousands of web transactions). TI's OMAP already accelerates AES despite the fact that AES was originally conceived to be a "software friendly" algorithm.

We could see Intel's design team make a clear effort to distinguish and differentiate between the user interfaces in Moblin for netbooks versus MID's. There was a clear acknowledgement that the netbook browser would never work on a MID. Although MID's are still being positioned as pocketable devices, the perfect form-factor continues to evade the best OEM. But, Intel is clearly positioning itself as a one-stop system integrator, with Moblin + customizable user-interface + customizable applications. That is the loudest message to the ARM camp - simplify or unify or beware of the Intel juggernaut. As I have written in other blogs, software complexity (in smartphones and other mobile products) cannot continue to increase because it only prolongs the development cycle.

My last session of the day was a panel discussion on Moblin's ecosystem partners, most of who are ecosystem partners of TI. Canonical, Novell, Asianux, Linpus, Wind River and Monta Vista. I expected a lot of controversial banter (even if orchestrated) but walked out without any new information except for Chris Kenyon's revelation of an announcement from Canonical around Ubuntu Remix on 9.23.09. I tried my best to stoke the embers by asking the panel about the impact of Chrome-OS. Novell's response left a lot to be desired - we won't know until it is launched and some vague speculation around Google's strategy that was positioned as inferior to Moblin. I presume we would refer to it as dancing around the topic; some might call it denial.

Let me summarize what I believe is Intel's strategy to grow their market share and to diversify beyond the PC space:

1. Cloud computing, especially on the server side

2. Push the envelope of semiconductor process, and continue to drive power consumption down. As someone quipped, they do have a social responsibility too. And, the biggest ROI (in terms of power savings) is in servers.

3. Software as a strategy for smartphones, MID's and netbooks. Moblin 2.1 has arrived and it is looking good. Integrate the telephony piece and smartphone OEM's have a compelling reason to experiment.

But, their Achilles Heel continues to be power consumption for handheld/mobile/portable client products. That is where the ARM camp and TI have a stranglehold. For now, that is.

ARMed and ready for the mobile future

Amy — Tue, 22 Sep 2009 20:52:00 GMT

We've been talking for some time about the mobile Internet and the ability to deliver full browsing and web 2.0 applications on a device with all day battery life. Thanks to TI's OMAP processors, this is becoming a reality.

Over the past few months we've seen new devices, features and services hit the market. Each new advancement generates excitement for what consumers will hold in their hands tomorrow, and how mobile technologies will continue to change our lives. Through all the excitement, competition remains fierce as companies introduce strategies to innovate and differentiate in the market place.

TI is at the forefront of the mobile revolution. We are continuing to invest in the OMAP platform to maintain a leading market position. Built with a solid ARM foundation, OMAP processors provide the optimal mix of speed and power-sensitivity that drives the market's cutting-edge products. With the introduction of the OMAP3630 processor, we've moved to the 45nm process node which will give our customers increased design efficiencies and higher performance. Additionally, we are committed to our next generation OMAP 4 platform delivering even higher quality multimedia features such as 1080p video record and playback, 20 megapixel (MP) imaging and approximately a week of audio play time. The new platform will also provide significant improvements in performance and play time compared to today's most popular Smartphones, with 10x faster Web page loading times, more than 7x higher computing performance, 6x higher video resolution, 10x better graphics performance and 6x longer audio play time.

Additionally, our flexible, open platform supports for all major mobile operating systems including Symbian, Windows Mobile and Linux (with various distributions supported such as Android, LiMo and Maemo). Matched with our open ecosystem, resources and support model, this OS-agnostic approach enables developers to deliver breakthrough mobile applications and new user experiences.

At the end of the day, we believe Smartphones are becoming smarter and doing more without compromising battery life. Performance and power are two vital components to mobile success. As mentioned in a previous Mobile Momentum post, it's important to consider apples-to-apples characteristics when you study these comparisons. By focusing on the gigahertz message, for example, some tend to forget about the importance of power management. What's the point of having a Ferrari that goes from 0 to 60 in two seconds if it only gets 2 miles per gallon? At TI, we don't sacrifice one for another: speed and low power come hand in hand.

We've stood at the forefront of wireless innovation for more than a decade. From apps processors to wireless connectivity, analog and power management components, TI offers the largest scope of technologies and expertise that fit our customers' unique mobile needs. With all of this in our pocket, we continue to prove our leadership and expand into new markets that address mobile communications, consumer and computing needs. We're already enjoying the fruits of this labor with the launch of OMAP-3 based devices like the Palm Pre, Samsung i8910 and ARCHOS 5 Internet Tablet, and couldn't be more confident about the road ahead.

Intel on Linux, Netbooks and rivalry with ARM

Ramesh Iyer — Fri, 18 Sep 2009 16:17:00 GMT

Imad Sousou's (Director of Open source Technology Center at Intel) interview with David Meyer (ZDNet UK) caught our attention to justify this blog response.

[Imad's comments in italics]

There are certain operating-system vendors who take Moblin completely as is and use it, and add customization and provide support, and there are those who take various technologies from Moblin and incorporate them into their own operating systems--although, when people do that, they tend to focus on the user experience.

Version one of Moblin was more of an enabling project. It's basically just a set of technologies that Linux distributors and people who use Linux can use for the optimization aspect on our (Intel) platform. Moblin v2.0 is when we started really doing an integrated OS with its own reference user experience that OEMs and vendors can take and customize

There are a lot of things that we've done with Moblin that are fairly cutting-edge, such as social-networking integration. We created the infrastructure that lets you very easily integrate streams from all the social networks with simple (application programming interfaces) and then be able to allow the user, in an integrated way, to update their status and so on. Moblin also boots in five seconds. It's things like this--we look at things from a user experience perspective, then we build all the plumbing, such as the kernel and adding libraries and 3D infrastructure.

I don't think it's up to us to change how (customers view Netbooks). In a lot of ways it's the other way around. We will support and do what our customers and end users want to do rather than trying to force some usage model on the end users

Imad focused on a few key issues that are critical to highlight:

1. Delivering unique user experience

2. Delivering choice and flexibility to customers

3. Making it easier and faster to build the final product

For all it's glory of being open source and free, there is a lot of work that has to be put into Linux distributions to make them commercially deployable. Smartphone development programs have to borrow a leaf from the PC world - 16-18 months of development time is too long especially when the useful life of a smartphone (in countries like Japan and Korea) is 6-8 months. On the positive side, it ensures that consumers remain hungry to gobble up new products as soon as they are launched. I am sure there is a perfectly valid argument/justification behind the 16-18 months of development cycle (e.g. modem telephony certification, etc) - but, if you think outside the box - there's got to be a way to reduce this. There's got to be a faster way for semiconductor suppliers to get a faster return on the mega millions spent on complex SoC's. Never been done before should not equate to can-never-be-done. Android broke this theory - but phones with Android are still spending 12+ months in development. We expect a similar phenomena from Chrome-OS, and only time will tell if netbooks will launch any sooner. We would welcome your suggestions on your ideas to reduce development cycle time in smartphones.

OK, on to other pieces of Imad's interview:

The Nokia N900 is fundamentally an MID--a device that can do PC functionality and that includes a phone. End users want devices that can bring a full Internet experience, viewing the Web in a real way with Flash completely running, with all the normal plug-ins and content you would normally see. By definition, that's fundamentally what a MID is.

In my opinion, I don't see the MID as a new category. It's more an evolution of smartphones. There is some confusion (surrounding) the name. It's good to have terminology for something like an MID--it distinguishes it in what you are really getting. With the term "Netbooks," people said it was just a small notebook, and we said fine, it's just a small notebook, so let's call it something that means that.

It's very hard to comment, because I haven't seen any of (the resulting devices). For the past year and a half, I've been hearing about the ARM Netbook or smartbook, but I haven't seen anything to see if it works.

From my personal experience, over the past several years I have been seeing people trying to put a PC OS on ARM, but you simply don't get the experience. It's the phone-browsing experience on a larger screen.

Unless the ARM ecosystem is able to fix the software problem of being able to provide a real Internet experience, real compatibility and those things, it will be difficult because the Internet is designed fundamentally for the PC.

It's (the web) designed for the PC in general and, unless your platform is designed to be PC-compatible, you will always--if you are able to show 90 percent of the Internet but you cannot show ESPN and MTV and whatever your top 10 Web sites are, which are generally media-rich, that's what people use those devices for. These are connected devices for Internet--the point of being connected is to use the Internet

Hmmmm.....plenty of contradictions to vector into a diatribe?

I am so glad that Intel (or Imad alone?) acknowledged that terminologies (like MID) can be very confusing. But, Intel also has publicly gone on record claiming ownership for this category. Any way you slice it, MID's inherit a lot of their usage models and features from smartphones. Which is why ARM-based MID's (like those from Texas Instruments' OMAP) have been so successful commercially while we have not yet seen consumer acceptance of Atom-based MID's although there was a lot of press around Intel's MID products. What went wrong? Or, what did the ARM camp do right?

First, TI (I can only talk about TI for obvious reasons) focused less on product names and focused more on delivering the best silicon and the best software that truly delivers a differentiated user experience. Archos' media tablet and Nokia's N900 are the first in a long series of salvos.

Second, we focused on ensuring that we can deliver the best browsing experience. Nokia, since the days of N800 and N810 (both of which used TI's OMAP2), and recently with N900 (which uses OMAP3), has clearly shown that no-compromise browsing experience can be delivered on an ARM architecture (like OMAP) if you optimized Adobe's (full) Flash player to leverage the integrated multimedia and graphics engine (inside OMAP). Therefore, Intel's premise of "the web was designed for x86" is redundant - more on this later in the blog.

Third, users demand an all-day battery experience. Intel (if you are really listening to your customers) should know that their customers hate the notion of 2-3 hour battery life. The usage model for portable appliances (like MIDs) will be identical to smartphones - charge it and forget it for atleast 1 day.

Fourth, and this is a marketing observation: If consumers wanted to buy PC's/notebooks, they will flock to name-brand PC customers with an Intel-branding on it. Yes, you are the undisputed kings in the PC world. The same consumer will flock to traditional smartphone OEMs for (a non-x86) smartphone and to CE giants for (non-x86) portable appliances. To ignore this simple "state of the mind" is futile (but easy).

Moving on to other interesting observations:

I am sure that the community will disagree with Intel's (very old and very antiquated) notion that the web was designed for the PC. Browsers may have been implemented on a PC; but, so are complex SoC's and DSP algorithms. Surely, you are not going to claim that these were also designed for the PC? A fundamental misconception is the availability of Adobe Flash (which is really what Imad is alluding to) for ARM based processors. Rest assured that the ARM community is changing that (as we speak) - you should really stop hiding behind this veil and focus more on adoption of your processors in the space beyond PCs/notebooks/netbooks. Embedded processing has its own challenges and nobody questions INtel's ability to get there - the real question is : Will you ever be able to catch up with ARM SoC's (like OMAP) which have a head-start by 2-3 years and who will not stop to innovate.

In closing: Smartbooks, netbooks - too many names, too much confusion. I think that Intel and the ARM community must rally around delivering the best processing experience at the lowest power consumption at the most affordable price at the fastest RoI (I am sure that this will bring a smile to our shareholders).

I have a simple theory - break the $100 retail price for "mobile computing" and you will spark plenty of innovation amongst hardware OEM's to experiment with different form-factors of delivering computing experience. Cloud computing and thin-clients are a step in that direction. The key is not more MHz or more GHz; the real solution might be "just enough GHz" - and that my friends will not be 2+ GHz but barely north of 1+ GHz. For all the supporters of software programmability and the flexibility to adapt to a changing world, there is plenty of innovation in the semiconductor world, so common-place algorithms can be embedded in silicon gates leaving the processor free for other useful tasks.

A big congrats and “high 5” for ARCHOS!

Amy — Tue, 15 Sep 2009 16:22:00 GMT

As you can tell by now, here at TI we're pretty excited about the promise of Android community, as well as the great traction that our OMAP 3 platform is receiving. Today we get to celebrate milestones with both as ARCHOS today unveiled their Android-based ARCHOS 5 Internet tablet, which takes the personal multimedia experience to a whole new level. The ARCHOS 5 is built with a solid collection of TI technologies - including an OMAP^TM 3 processor, NaviLink^TM solution, WiLink^TM solution and analog components - that together enable awesome device capabilities, like:

Uncompromised, full web browsing experience that leverages the ARM® Cortex^TM-A8 core, with performance up to 800Mhz
720p (HD) video playback all formats up to H.264 HP
A compelling, intuitive user interface
Stunning performance at low, smartphone-like power budgets
"Always-on" connectivity in all user environments for seamless data connections (WLAN/BT/FM and GPS)

Besides sporting these eye-catching features, the ARCHOS 5 is another proof point as to why we stand committed to Android. Our engineers work diligently with the Android platform during OMAP 3 processor development and validation, which helps customers bring cool new devices to consumers' hands quickly in an open environment. In fact, the ARCHOS team noted in a Feb. 9, 2009 press release that they selected TI's products not only for their low-power and high performance, but to leverage the company's experience in developing compelling, highly-competitive Android-based products: "The level of software investment made by TI on the Android platform allows ARCHOS to quickly and cost effectively deliver an Android-based device that supports advanced features such as the newly integrated voice capabilities."

ARCHOS team: We're thrilled for you, and can't wait to get our hands on our very own ARCHOS 5! Congrats.

Android is in the air at GigaOm's Mobilize Event

Amy — Thu, 10 Sep 2009 19:22:00 GMT

We at TI would like to congratulate Motorola on the successful launch of their first Android-based solution. The promise of Android is strong, and with the availability of new devices in the market, consumers are starting to touch, see and feel the true power of an open community. We will soon have access to some of the world's most innovative applications at our fingertips as devices from Motorola and others fuel our desires for cutting-edge, intuitive features and applications. We look forward to Motorola's continued commitment to the Android market with even more devices, and to strengthened kernel releases that will keep emerging from the Android community. Opportunities for innovation are limitless as both organizations continue to flourish. Congratulations again!

The GPS Wave

Dave Lacinski — Wed, 02 Sep 2009 22:53:00 GMT

WOW -- GPS seems to be taking the mobile market by storm these days. At TI we are seeing strong penetration into handsets and other portable devices. I find the user cross section especially interesting -- how many of us know someone who would have never looked at a map, now totally enamored with the fact that their handheld device can figure out their location and tell them how to get where the want to be!

Market analysts like iSuppli recently published that virtually every smartphone will have GPS capability by 2011, with more than 305 Mu in 2014. ABI reported new exciting applications for GPS in sports and fitness that will fuel additional growth. Couple those with the fact we are still waiting for the tsunami of various location based services to break and create opportunities to monetize the location technology.The market possibilities are too numerous to list, but they are creating an exciting time in the industry.

In 2008, our NaviLink™ solution was the most popular standalone handset GPS chip, grabbing nearly 40% market share (Source: A guide to mobile connectivity chips report, Linley Group, February 2009). We are continuing to refine this product line with new enhancements that make implementation even easier for the OEM to create a premier user experience and lower the cost through our integration strategies with market leading TI mobile WLAN and Bluetooth solutions. Integrated solutions take the best of breed for a single function silicon device into a powerful combination of technologies that can coexist in a very small footprint.

So is it really just about the phone? I would say that there are many tastes in the marketplace -- some users will have a preference for the “all in one” smartphone, but others will prefer to have specialized devices. Only time will tell how this all plays out.

What do you think about the dedicated navigation device versus the smartphone debate?

------- Dave Lacinski

From 2009 to 2020: A history of developments in programmability

Christy Brunton — Wed, 02 Sep 2009 19:18:00 GMT

Alan Gatherer, CTO Communications Infrastructure Group

Predicting the future is primarily an act of the imagination. However, Digital Signal Processors are showing some strong trends and I think it is possible to predict what will happen in the next few years as we move towards the next order of magnitude increase in computational efficiency.

Here are my thoughts on the next 12 years.

• 2009: Multi-core is here. With the increase in SoC architectures, single-core CPU devices have become more the exception than the rule.

• 2012: Network-on-Chip (NoC) arrives. A NoC is a high-performance device, which is really a grouping of processing islands connected by packet-based, point-to-point asynchronous communication highways.

• 2010–2015: Component-based software. The number of cores on a device is still fairly modest, and individual software components are developed for a single computational cluster by “component developers” and then “assembled” onto a multi-core system. Development tools for this methodology improve steadily as virtualization of hardware through middleware is driven by efforts such as the SCA (Software Communications Architecture) for SDR (software-defined radio). Auto generation of glue code between components becomes the norm.

• 2015–2020: Single Program Multiple Data (SPMD). The component-based approach begins to fail as the number of cores reaches 32. Turning to techniques used in high-performance computing (HPC), the embedded software community develops the SPMD approach where a program can be compiled to run over multiple cores. While initially requiring explicit description of the communication flow, pragmas are now employed to enable the parallel nature of algorithms to be exploited by a variety of multi-core devices.

• 2015: The Death of the FPGA. An important footnote in the history of programmability is the demise of the FPGA. Small multi-core CPUs consume significantly less power as well as provide a richer set of mapping options for complex algorithms and communication patterns than does the distributed fabric of ALUs and LUTs that make up FPGAs.

• 2020: The CPU disappears. Spreading functionality across multiple CPUs drastically simplifies the silicon overhead on each CPU, and hardware-based OS support manages NoC traffic efficiently. Programmers are unaware of the communication between CPUs and can develop/debug code without having to know which individual execution units are involved. Programming follows more the overall flow of data than its individual parts.

The range of devices available in 2020 will be about the same as it is in 2009. In 2020, embedded DSPs will still be a heterogeneous combination of CPUs and accelerators. Even though programmers are unaware of the individual devices when programming, it will still be true that some devices perform certain tasks much better than others.

Since much of the value of SoCs is placed in the careful choice of peripherals, CPU and DSP manufacturers differentiate themselves by providing the best combination of different IP blocks and how they connect. In the end, the quality of development tools and application software support will determine the first-tier players.

About the author

Alan Gatherer is the CTO for the High Performance Multicore Processors group at TI and is responsible for all strategic development of TI’s digital baseband modems for 3G wireless infrastructure. Since joining TI in 1993, he has worked on various digital modem technologies including cable modem, ADSL and 3G handset and basestation modems. In addition, he holds 60 patents and is author of the book “The Application of Programmable DSPs in Mobile Communications.”

Newest OMAP 3 processor-based device sets even higher standards in MID market

Amy — Fri, 28 Aug 2009 19:31:00 GMT

We've seen some really cool and innovative OMAP 3 processor-based devices enter the market this year. Adding to the list this week, Nokia introduced the N900 based on our OMAP3430 processor. The N900 is already garnering praise. Tim Conneally with Betanews called it a powerful MID option with "the chassis and connectivity of a smartphone, and the guts and operating system of a MID." Chippy with UMPCPortal notes "based on the specs, the form factor and the operating system it's fair to say that Nokia has beaten Intel into the 'MID' market, stolen some credit and is well positioned to bring out more OMAP/Maemo 5-based products into the MID market."

It's exciting to see the innovation that is taking place. Our customers continue to push the envelope with their devices, giving consumers robust solutions that put even more features and functionalities at their fingertips.

The DLP Discovery Kit offers exceptional flexibility in light modulation

Wizard — Fri, 28 Aug 2009 15:47:00 GMT

According to Wikipedia, my favorite source of information, “a spatial light modulator (SLM) is an object that imposes some form of spatially-varying modulation on a beam of light.” This brings to mind my recent trip to Mexico to visit the Mayan city of Chichen Itza. The pyramid in the center of this ancient city was positioned in such a way that the shadows would form the shape of a plumed serpent during the spring and autumn equinox. This is an interesting example of spatial light modulation on a very large scale. The Mayans were able to take visible light and play with its positioning to make an unexpected image appear.

So how does this relate to a DLP-chip? A digital micromirror device (DMD) simply takes light and moves it around to make different images. It is easy to see the effects of the modulated light in a projector, TV, or movie theater. But one distinguishing advantage of the DMD over many other light modulators is its flexibility. Since the micromirrors are made of aluminum, they reflect light well beyond the visible range. Aluminum is highly reflective throughout the visible, near-infrared, and ultraviolet regions of the spectrum. A human eye can detect light from roughly 380 nm to 750 nm. We learned in a previous blog that a mantis shrimp can see a broader spectrum of light with sensitivity into the UV and IR spectrum. But the aluminum surface of a DMD micromirror can go way beyond even what a mantis shrimp can see by efficiently reflecting light from UV (200 nm) all the way into IR (>10,000 nm). Aluminum is one of the most highly reflective materials available. With the exception of a small dip below 90% reflectivity efficiency around 820 nm, aluminum reflects over 92% through most of the UV/visible spectrum and approaches 98% reflectivity in the IR band.

This opens up many opportunities for light modulation that were not available several years ago. Depending on what you use as a “sensor,” the possibilities are endless. A sensor could be a CCD camera or it could be skin absorbing UV light or it could be photo-sensitive material.

UV applications include medical procedures, photolithography, and rapid prototyping. IR applications include scene simulations, patterning of IR sensitive materials, and security/surveillance. And of course there are applications that use a broad spectrum such as hyperspectral imaging and telecommunications. Don’t think of light as just the energy we can see. Think of light in a broader sense from UV to IR. What can we do by redirecting, patterning, shaping, or otherwise modulating a broad spectrum of light? Making a plumed serpent is pretty cool but there are many incredible applications in development right now. We live in exciting times.

New Graphical Parametric Selection Tool available

rogerio — Fri, 28 Aug 2009 15:12:00 GMT

Hello design engineers,

As some people say "Time is money" and to reduce the time you usually spend looking for components, TI is developing a new set of Graphical Parametric Selection Tools (GPS). You may be already familiar with the Analog to Digital Converter (ADC) GPS tool (http://focus.ti.com/en/multimedia/flash/selection_tools/adc/adc.html), and it is now available a new one for Precision Amplifiers (http://focus.ti.com/en/multimedia/flash/selection_tools/precamp/precamp.html).

The experience has shown that by using these tools users can select the right parts in a shorter period of time and with fewer number of "clicks". Also, the Precision Amplifier GPS tool allows you to see the closest option available. Here is an example: your design needs a precision amplifier with GBW of 60MHz and Iq per channel lower than 10mA. By using the new Precision Amplifier GPS tool, you can see easily that OPA37 meets your design requirement.

We added a new functionality to the GPS tools based in the feedback we got from users like you: once you move your mouse over the parameter, the tool will display for few moments an explanation of that particular parameter. Here is an example:

Please, feel free to test our new selection tools and send us your feedback. We take them very seriously and you can help us to make these tools better. There is a little red box at the bottom of the tool that allows you to post your comments, questions and feedback for us.

We also have a survey that you can take, if you are using IE as browser. You can expect more tools coming out soon and we will keep you updated about the new ones.

Thanks for your time (and since "Time is money" I am done for today).

Regards, Rogerio