Today kicked off the start CTIA 2011, and Anand is already over in Orlando, Florida covering the event. I'm missing out this time, instead working on a couple of important reviews coming up, including the Galaxy S 4G and HTC Thunderbolt which you should stay tuned for.

Nexus S 4G

However, there's been some interesting news out of the conference from Google and Sprint. The first bit of news is the Nexus S 4G for Sprint, which superficially is the same as the Nexus S that we reviewed back in December. What's different is inclusion of support for Sprint WiMAX 4G and CDMA2000-EVDO Rev.A instead of GSM/UMTS like in its younger cousin. 

Samsung and Google haven't disclosed what WiMAX baseband is going into the Nexus S 4G, but it seems possible that Samsung would choose its own CMC730 SM single chip solution for WiMAX and likely something from Qualcomm for CDMA2000 support for Sprint voice and data. I hate speculation, but I'd be surprised if Samsung didn't at least go with its own WiMAX chipset. It'll be interesting to see how very similar hardware performs with different baseboards in our battery life tests. For comparison, the HTC EVO 4G and EVO Shift 4G both use use a Sequans SQN1210. Update: The Samsung Epic 4G uses the CMC730S for WiMAX and a Qualcomm QSC6085 CDMA2000 baseband, so CMC730 + QSC6085 therefore seems like a likely combination for the Samsung Nexus S 4G. 

The rest of the specs remain exactly the same between the Nexus S and Sprint-specific Nexus S 4G. Same 5 MP rear camera and VGA front camera, lack of 720P video capture, 1 GHz Hummingbird SoC with PowerVR SGX540, 16 GB iNAND internal flash, 802.11n/b/g, NFC, GPS, and 1500 mAH battery. 

It's odd that there are no logos on the back of Sprint's Nexus S 4G press shots

Both retain the same mass - 129 grams, and outline size. The original Nexus S specs give thickness as 10.88 mm, or 0.428 inches. Interestingly enough, the Nexus S 4G specs given by Sprint list thickness as 0.44 inches, or 11.17 mm. Either there's some rounding error going on from the conversion, or the Nexus S 4G is ever so slightly thicker, which seems possible given requisite MIMO for WiMAX and the number of extra antennas that entails. Both of the EVOs and the Thunderbolt are relatively beefy phones thanks in no small part to MIMO. 

The software experience on the Nexus S 4G will be pure Google. That means no Sprint NASCAR or Sprint TV applications preloaded. No doubt the marketplace will have a tab where you can grab all of that if you so choose, however.

We'll review the Nexus S 4G in due time when it launches. For now, Sprint has a Nexus S 4G landing page with information at sprint.com/nexus. 

Google Voice on Sprint

The other big announcement is that Sprint is partnering up with Google to enable native Google Voice support for all of its customers. The feature isn't live yet, but is coming soon, and you can sign up here to be notified if you're an existing Sprint customer. What's different here is that unlike every other carrier plus Google Voice combo, no number port or call abstraction layer is required. 

If you already have a Google Voice number, you can use it natively on your phone for calls and texts. There's no abstraction or forwarding going on like Google Voice normally adds. Likewise, if you don't have a Google Voice number, you can transparently use your existing Sprint number with Google Voice - this isn't a number port at all. Sprint is essentially dynamically routing calls appropriately using some routing table magic. Everything gets routed over - voicemails, texts, and calls. Even better, no application is required to use Google Voice, so even featurephones and dumbphones will work natively, you just won't get access to immediate transcription and other features available only through web or the Google Voice app. You can even switch back and forth between using your number with Google Voice and normal PSTN routing. 

Perhaps best of all is that texts can be sent through Google Voice using the Google Voice for Android application for free, without a messaging plan. That works so long as you don't send any SMSes through the native messaging app, just Google Voice. Sprint has that and a few other interesting tidbits in its FAQs. 

It'll be interesting whether the perceptible extra latency that usually accompanies Google Voice calls completely disappears as a result of native carrier routing. This is huge news for Sprint, as it will soon be offering the most native implementation of Google Voice to date. 

When 3DMark11 launched earlier this year, it seemed like the time was right for an update to PCMark Vantage as well. While we don't have an exact release date or pricing, Futuremark has now confirmed their intention to update the product, but instead of PCMark11 we're getting PCMark 7. The release does say "coming soon", which usually means Futuremark is less than a month or two away from the official launch (but don't hold us to that).

Over the years, we've complained more than a few times about Futuremark's 3DMark scores; It's a long story, but I've posted a comment below for the interested. PCMark on the other hand has been more useful at giving us an easy to run application suite. (If you haven't tried it, SYSmark requires some serious voodoo to get it to run on most modern systems—many of the patches for Vista will break the benchmark, and let's not even get into Windows 7.) The new version looks to provide an updated performance analysis for modern Windows 7 PCs, which may mean Vista is no longer supported. [Update: It should run on Windows Vista, but "missing API support" means the full suite won't run properly, so the final score will be affected.]

We've been a bit critical of PCMark Vantage with regards to inflated SSD results, so it will be interesting to see if PCMark 7 provides a more realistic picture of what an SSD can do for the overall system performance. Yes, we know they're faster, but the problem with Vantage is that even relatively mediocre SSDs did an excellent job at inflating benchmark scores. The PCMark 7 press release states, "Hardware innovations like solid state drives (SSDs), and new form factors such as netbooks and tablets are greatly increasing the range of PC performance available to businesses and home users," so we should see some new metrics addressing such changes at the very least.

System requirements list a DX9 video card as the minimum, but DX10 and DX11 are necessary for some additional tests (i.e. tests that aren't part of the main PCMark score). There are seven test suites with 20 different workloads designed to give various views of system performance. The seven test suites are PCMark, Lightweight, Entertainment, Creativity, Productivity, Computation, and Storage. The Lightweight suite is the one truly new entrant, and Computation could be interesting if there's any support for GPGPU type workloads; the other suites tend to correlate with one or more of the suites in Vantage (Memories, TV and Movies, Music, Communications, Productivity, and HDD—thankfully, the Gaming Suite that tended to be more of a storage benchmark is now MIA).

The full (and somewhat short) press release is below.

New Benchmark Includes 7 Suites for Complete PC Performance Analysis

HELSINKI, FINLAND – MARCH 21, 2011 – Futuremark®, the developer of the world's most popular benchmarking software, announced PCMark® 7 today, a new benchmark offering complete PC performance analysis for Windows 7. PCMark 7 includes 7 separate test suites combining more than 20 individual workloads covering storage, computation, image and video manipulation, web browsing and gaming. A release date has not been announced though the official website states that PCMark 7 is coming soon.

"Hardware innovations like solid state drives (SSDs), and new form factors such as netbooks and tablets are greatly increasing the range of PC performance available to businesses and home users," said Jani Joki, Director of PC Products and Services at Futuremark. "With so much choice available, PCMark 7 is an essential and easy to use tool to test and compare PC performance accurately and reliably across a wide range of usage scenarios."

PCMark 7 provides a set of 7 suites for measuring different aspects of PC performance with a high degree of accuracy. Overall system performance is measured by the PCMark Suite. The Lightweight Suite measures the capabilities of entry level systems and mobility platforms unable to run the full PCMark suite. Common use performance is measured by the Entertainment, Creativity and Productivity scenario suites. Component performance is measured by the Computation and Storage hardware suites. The Storage suite is ideal for testing SSDs and external hard drives in addition to the system drive.

For more information please visit http://www.pcmark.com/ or follow Futuremark on Facebook.

Last year we ran a little series called Ask the Experts where you all wrote in your virtualization related questions and we got them answered by experts at Intel, VMWare as well as our own head of IT/Datacenter - Johan de Gelas.

Given the growing importance of IT/Datacenter technology we wanted to run another round, this time handled exclusively by Johan. The topics are a little broader this time. If you have any virtualization or cloud computing related questions that you'd like to see Johan answer directly just leave them in a comment here. We'll be picking a couple and will answer them next week in a follow up post.

So have at it! Make the questions good - Johan is always up for a challenge :)

The transition to smaller form factors hasn't been exclusively a client trend over the past several years, we've seen a similar move in servers. The motiviation is very different however. In the client space it's about portability, in the datacenter it's about density. While faster multi-core CPUs have allowed the two-socket 1U server to really take off, they have also paved the way for a new category of density optimized servers: the micro server.

The argument for micro servers is similar to that for ultra low power clients. Only a certain portion of workloads really require high-end multi-socket servers, the rest spend much of their time idle and thus are better addressed by lower power, higher density servers. Johan typically argues that rather than tackling the problem with micro servers it's a better idea to simply increase your consolidation ratio into fewer, larger servers. There are obviously proponents on both sides of the fence but Intel estimates that the total market for micro servers will reach about 10% of its total shipments over the next 4 - 5 years. It's a small enough market for Intel not to be super concerned about but large enough that it needs to be properly addressed.

Today Intel believes that it addresses this market relatively well with the existing Xeon lineup. Below is a table of Sandy Bridge Xeons including a 45W and 20W part, these two being directed primarily at the micro server market:

Drop clock speed (and voltage) low enough and you can hit the lower TDPs necessary to fit into a micro server. Thermal constraints are present since you're often cramming a dozen of these servers into a very small area.

Long term there is a bigger strategy issue that has to be addressed. ARM has been talking about moving up the pyramid and eventually tackling the low end/low power server market with its architectures. While Xeon can scale down, it can't scale down to the single digit TDPs without serious performance consequences. Remember the old rule of thumb: a single microprocessor architecture can only address an order of magnitude of TDPs. Sandy Bridge can handle the 15 - 150W space, but get too much below 15W and it becomes a suboptimal choice for power/performance.

The solution? Introduce a server CPU based on Intel's Atom architecture. And this is the bigger part of the announcement today. Starting in 2012 Intel will have an Atom based low power server CPU with sub-10W TDPs designed for this market. Make no mistake, this move is designed to combat what ARM is planning. And unlike the ultra mobile space, Intel has an ISA advantage in the enterprise market. It'll be tougher for ARM to move up than it will be for Intel to move down.

Intel's slide above seems to imply that we'll have ECC support with this server version of Atom in 2012, which is something current Atom based servers lack.

The only real question that remains is what Atom architecture will be used? We'll see an updated 32nm Atom by the end of 2011 but that's still fundamentally using the same Bonnell core that was introduced back in 2008. Intel originally committed to keeping with its in-order architecture for 5 years back in 2008, that would mean that 2012 is ripe for the introduction of an out-of-order Atom. Whether or not that updated core will make it in time for use in Atom servers is still up for debate.

We first met Indilinx in early 2009, with its Barefoot controller at the heart of OCZ's Vertex SSD. Until SandForce showed up in 2010 with the SF-1200/1500 series of controllers, the Indilinx Barefoot was the cheaper alternative to Intel. If it wasn't for Indilinx I'm not sure Intel would've had any real competition until last year. 

The OCZ/Indilinx relationship was always a close one, but it's about to get a lot closer. Today OCZ announced that it has signed a definitive agreement to acquire Indilinx Co., Ltd for ~$32M in OCZ shares. The deal is expected to close within 30 days.

Why would OCZ acquire Indilinx? To be honest, it's the only way to ensure growth in the future. While OCZ is currently the retail/channel face of SandForce, SF also happens to sell to all of OCZ's competitors. OCZ may get early access to SandForce hardware, but so could anyone else who shows up with a big enough check. Ultimately to stick around in this business OCZ needs to be able to build its own controller and that's what the Indilinx acquisition is designed to allow.

Indilinx was working on its first 6Gbps controller codenamed Jet Stream, however the design never surfaced. Delays do happen in chip design and I'm guessing Indilinx just couldn't make it happen with Jet Stream. 

What does this mean for SSDs today? Probably very little. OCZ will continue to ship SandForce based drives, however at some point I'm expecting to see an exclusive OCZ branded controller. While this move will likely guarantee OCZ's ability to differentiate amongst its smaller competitors, I'm not entirely convinced  it will be enough to compete with the likes of SandForce. We've even seen Intel struggle with that battle. 

Anything can happen though. Two years ago I didn't expect anyone other than Intel to dominate SSDs. A year ago it was SandForce. By this time next year anything could happen.

The DMA (Digital Media Adapter) aka media streamer market has been dominated by Sigma Designs, Realtek and of late, the Intel CE 4xxx series. The low end market is catered to by the Boxchip and Amlogic based media streamers.

Having got the codec compatibility right (at least on paper), the choice for the media streamer manufacturers was to either go the 3D route or add extra non-core functionality in the next-gen products. 3D is yet to go mainstream despite the best efforts of the top tier manufacturers.

In 2011, you can expect a slew of media streamers running Android. The TViX Xroid A1 has already been announced and demonstrated to be running Android on the SMP8656. Read on for our coverage of the launch of another Android-based media streamer, the Nixeus Fusion XS, which also happens to be Marvell's debut vehicle in this product space.

We try not to bring you too much news about product announcements unless there's something particularly intriguing about them; we get inundated by them and most of the time it's the most generic of refreshes. Happily that's not the case with Lenovo's shiny new ThinkPad X220 notebooks.

Inexplicably Lenovo is opting to label these two very different notebooks under the same X220 header: one is a tablet clocking in at 3.88 pounds with a 4-cell battery; the other is an ultraportable that weighs less than three pounds. Both come with support for either SSDs or mechanical hard disks (with a 4GB SSD option as a special order).

We'll start with the ultraportable X220. Lenovo is shipping it with a 12.5" 1366x768 LED-backlit screen, but you can upgrade to an IPS panel. It maxes out at 8GB of DDR3 and has a strong spread of Sandy Bridge mobile processors to choose from, starting with the Core i3-2310M at 2.1GHz and going all the way up to the i7-2620M at 2.7GHz. Strangely, only the i7-equipped models come with USB 3.0 connectivity. Reviews of the X220 are already popping up on the internet and the IPS screen is proving as impressive as you'd expect, but not nearly as impressive as the battery running time: Lenovo claims up to 15 hours on a 9-cell battery, a hyperbolic figure to be sure but not as crazy as you'd think. NotebookReview's test model came with a 6-cell battery and was pushing nine hours.

You can see and eventually order the ultraportable X220 here, and MSRP is expected to start at a not-unseemly $899.

The other X220 is the tablet model. Again it ships with a 12.5" 1366x768 LED-backlit screen, but in this case the only choice is the finish you want on the IPS panel: Infinity Glass or Corning Gorilla. Yes, the X220 tablet comes with an IPS panel standard, proving that Lenovo understands what ViewSonic couldn't figure out with their tablet: that viewing angles are really important. Unfortunately the X220 tablet is nearly a pound heavier than its ultraportable cousin and doesn't come with an option for USB 3.0 connectivity. Lenovo quotes nine hours of running time with the 8-cell battery.

The X220 Tablet isn't up on Lenovo's site yet, but MSRP is expected to start at $1,199.

Today, Steve Jobs took a sabbatical from his sabbatical to hop up on stage and tell us all about the iPad 2, the next revision of Apple’s wildly popular tablet PC.

The announcement concerned both hardware and software – the iPad 2 is coming to the US on March 11, and with it will come the iOS 4.3 update, iMovie for iPad, and GarageBand for iPad. It will launch at the same capacities and price points as its predecessor, will come in both black and white, and launches internationally on March 25. Read on for details.

AMD just sent over a bunch of shots of an upcoming product that we may or may not be presently benchmarking: the Radeon HD 6990. Check out the gallery for the pics.

Update: AMD accidentally gave us a shot of the Radeon HD 6990 without the fan shroud attached. Apparently that image reveals a bit too much about the product and AMD asked us to remove it. Sorry guys :)

The last time we discussed CUDA and Tesla in depth was in September of 2010. At the time NVIDIA had just recently launched their lineup of Fermi-powered Tesla products, and was using the occasion to announce the 3.2 version of their CUDA GPGPU toolchain. And though when we’re discussing the fast pace of the GPU industry we’re normally referring to NVIDIA’s massive consumer GPU products arm, the Tesla and Quadro businesses are not to be underestimated. An aggressive 6 month refresh schedule is not just good for consumer products it seems, but it’s good for the professional side too. NVIDIA’s CUDA team seems to have taken this to heart, and so here we are just shy of 6 months later with NVIDIA preparing to launch the next version of CUDA: CUDA 4.0.

Back at IDF 2010, we wrote about Intel Light Peak nearing its eventual launch in 2011. Back then, the story was a 10 Gbps or faster physical link tunneling virtually every protocol under the sun over optical fiber. Though an optical physical layer provided the speed, in reality the connector and physical layer itself wasn’t as important as the tunneling and signaling going on beneath it. Daisy chain devices together, and connect everything with one unified connector and port. 

That dream lives on today, but sans optical fiber and under a different name. Intel’s codename “Light Peak” is now named Thunderbolt. 

As expected, Apple today unveiled a range of speed and functionality improvements for its MacBook Pro lineup. The update was unusually quiet for Apple. There was no scheduled press event and nothing more than a press release announcing the specs and availability. Apple retail stores received stock prior to today and began selling product immediately. The Apple online store also has immediate availability.

No mere speed bump, these new MacBooks bring Intel’s new Sandy Bridge processors chipsets to the entire line, replacing the previous Arrandale processors and finally retiring the aging Core 2 Duo from service in the 13-inch model.

Contrary to earlier reports, there are no default SSD configurations although the solid state offerings are still optional. The big new feature (outside of Sandy Bridge) is support for the first incarnation of Intel’s Light Peak interface technology, now called Thunderbolt.

On Day 0 of this year's Mobile World Congress Samsung and NVIDIA announced that the new Galaxy Tab 10.1 will come to market with NVIDIA's Tegra 2 (T20) SoC. At the same time, the two quietly announced they would be working on a new superphone together also based on Tegra 2. At Samsung's press conference however all we saw was the Galaxy Tab 10.1 and the Galaxy S II, the latter using Samsung's own Exynos SoC.

So what happened to the NVIDIA based smartphone and why would Samsung bother with using Tegra if it already had an Exynos based smartphone? To understand why we need to look at the Galaxy S. At its MWC press conference Samsung mentioned that it sold 10 million Galaxy S phones in 2010. The Galaxy S II should sell at least as much, if not more, once it's officially introduced.

Exynos however is a brand new SoC, with a brand new GPU for Samsung. Meeting demand for the Galaxy S II in all markets across the world with an SoC that Samsung has never shipped is risky at best. If you saw our benchmarks yesterday you'll note that NVIDIA's Tegra 2 is a near equivalent in terms of CPU performance and notably better in GPU and Flash performance. In other words, Tegra 2 isn't a bad alternative.

Meet the GT-I9103:

The GT-I9100 is the normal Exynos based Galaxy S II, the I9103 is the Tegra 2 edition. As one of our readers (thanks sarge78), Samsung lists its own dual-core Application Processor in the Galaxy S II as not being used in all regions. It's too early to tell if that means that we'll get Tegra 2 or Exynos depending on physical region.

I suspect Samsung didn't want to confuse users by announcing both a Tegra 2 and an Exynos based superphone at MWC. An unknown user managed to benchmark the GT-I903 at MWC and submitted the data to the GLBenchmark database. The GT-I9103's performance looks comparable to the Atrix 4G, meaning it's going to deliver the same experience we've seen in our Optimus 2X and Atrix 4G articles.

On the first day of MWC 2011, LG announced its four-part smartphone and tablet strategy for 2011 as part of what it's calling a "new beginning" for becoming a key contender in the mobile space.

The strategy is simple and four-pronged. The first is dual-core and multi-core processing for performance. We've already taken a look at the LG Optimus 2X, the first dual-core phone, and found that it brought Android an appreciable speed increase. The software on the Optimus 2X is getting more stable - we're told the instability we saw in our review unit is totally fixed in newer builds, which we're having flashed. 

The second part is display innovation, which no doubt is a reference to the LG Optimus Black and its super-bright NOVA display, or the inclusion of 3D stereoscopic displays on other LG smartphones. We got to look at the NOVA briefly at CES and were impressed with how bright and contrasty the display is, and we'll look at it in due time. Hopefully NOVA displays will move into more displays beyond just the Black. 

The third arm is 3D - which brings us to the LG Optimus 3D. As we already mentioned, the LG Optimus 3D is OMAP4 based with an OMAP4 4430 and features a glasses-free autostereoscopic 3D display and stereoscopic camera. The 3D camera system consists of two rear-facing 4 MP cameras spaced 24 mm apart. Recording in 3D also includes a live 3D preview with real-time adjustible depth control. We got a few minutes to look at the Optimus 3D and came away decently impressed with the glasses-free 3D screen. You can't see it above, but that golf game is entirely 3D on the Optimus 3D.

LG is aggressively  marketing fact that the Optimus 3D is both dual-core and dual-channel. By dual-channel, LG is referencing OMAP4's dual-channel LPDDR2 memory interface. The Optimus 3D is listed as having 4 Gigabits of RAM translating to 512 MB. LG showed off the overall quadrant score of the 3D, and though we're not completely sold on Quadrant being a competent benchmark that truly reflects performance, the 2958 score is impressive next to 2670 which we saw on the Optimus 2X. The Optimus 3D captures 3D at 720P and 2D in 1080P. We look forward to benchmarking OMAP4 in due time. 

LG has secured a strategic partnership with YouTube to enable the Optimus 3D to both upload and locally play 3D content. YouTube feels that the Optimus 3D provides a leading 3D experience by enabling record and playback, and and in this way will catalyze 3D adoption.

The Optimus 3D also has a dedicated 3D interface, including some preloaded 3D games from gameloft. It looks like NOVA (as in the game, not the display - how's that for a namespace collision) will be an included HD 3D title. 

The fourth part of LG's strategy is getting into the "true tablet war" with the Optimus Pad. 

LG wants the Optimus Pad to be an optimal balance between portability and productivity with an 8.9" size, and thinks it has the form factor right. The size is indeed an interesting balance between the 7" and 10" form factors we've seen thus far. LG made numerous illustrative comparisons with the Galaxy Tab and iPad.

The Optimus Pad is NVIDIA Tegra 2 powered, running Android 3.0 Honeycomb, and also incorporates a 3D camera system consisting of two 5 MP cameras. There's a host of preview modes, including mixed, anaglyph, and side-by-side viewing. There's also HDMI 1.4 output to playback that 3D on appropriate displays. The Optimus Pad also brings a 2 MP front facing camera (which seems to be the new standard) for HD video conferencing. We got some time with the Optimus Pad and plan to publish our impressions of the tablets we've seen so far in the next few days of MWC.

The Optimus Pad will be available in April, the Optimus 3D in May.

Sony Ericsson’s big news from Barcelona is obviously the Xperia Play, the mysterious and oft-rumoured PSP-phone. But they’ve also announced two new additions to the Xperia line - the Pro and the Neo. Between today’s new phones and the Xperia Arc from CES, SE has a complete stable of new devices to take on the rest of the smartphone world.

Read on for more details.

At MWC 2011, Microsoft just announced a number of improvements for WP7 including the timeline for when its first update will be pushed to devices.

The update with app-loading performance improvements and copy paste support we demoed back at CES finally got a more specific release date - it's slated to launch in early March. A number of other updates are slated for 2011, though Microsoft wouldn't assign a specific date. Those updates include multitasking support for 3rd party applications, better cloud synchronization, more web-services built into hubs, and a completely new web browser.

Microsoft recognizes that the browser is probably the single most important mobile application. It needs to be speedy, needs to be usable, and needs to have full feature parity with desktop browsers. Though the WP7 browser feels very fast translating and zooming around, JavaScript was slow and HTML5 support was nonexistent. It was based on the older Trident engine used in pre-IE9 browsers, and never felt like it really belonged in the Web 2.0 or HTML5 era. Coming in 2011, WP7 will get a major update with an IE9 browser complete with the same level of GPU acceleration and HTML5 compatibility of its desktop brother. WP7's mobile IE9 browser will use the same core rendering engine as IE9's. 

Mobile IE9 includes HTML5 support for the video tag as well. That's the same H.264 support as IE9 has on the desktop.

Next, WP7 will better leverage Microsoft's cloud assets, including SkyDrive support in the office hub. Other hubs are getting updates that include faster cloud sync. One of the major curiosities when WP7 launched was the absence of twitter support. Also part of 2011 will be an update that brings twitter integration in the people hub. 

Probably the most major update, however will be inclusion of 3rd party multitasking support. Fast application re-hydration and a task manager menu are the key changes. The new multitasking interface allows for task switching using the back button. Pressing and holding the back button brings up a ring-switcher interface with screenshots of running applications.

Finally, Microsoft demonstrated an impressive example of cross WP7 and Xbox connectivity with a demo of users controlling the placement of obstacles in Kinect Adventures. Microsoft needs to move quickly to deliver WP7 updates into the hands of existing users to keep the platform competitive as Android and iOS also roll out updates. 

Though we still like to think of Intel first and foremost as a computer CPU company, the fact of the matter is the company is trying its hardest to expand their horizons. Among their expansion efforts are a push in to the smartphone space, and to further that Intel is at Mobile World Congress 2011 making their latest smartphone-related announcements.

The first announcement, and of course the one nearest and dearest to our hearts, is on the CPU side of things. Medfield – Intel’s next-generation Atom-based smartphone SoC is now sampling and will ship later this year. Intel still hasn’t thrown out a solid timeframe for when Medfield will ship, but Q4 is as good a guess as you’re going to get.

Medfield is the follow-up to Moorestown, Intel’s first Atom smartphone-sized SoC design that was launched only 9 months ago, and did not ship until the later half of last year. Moorestown has not had any major design wins, so while it’s out there you probably never have and never will see a Moorestown powered smartphone. As Intel’s first foray in to smartphone SoCs Moorestown had its teething issues – the principle platform was a 2 chip family, with only the Z6xx CPU + GPU + MC manufactured in-house at 45nm, while the MP20 PCH containing the camera, audio, I/O, and other supporting hardware was a 65nm product manufactured at TSMC.

The importance of Medfield in Intel’s product lineup is that it should resolve Moorestown’s teething issues. The CPU and PCH are being integrated on to a single chip, and the entire product is being built on Intel’s 32nm process, which will allow handset makers to more easily fit Medfield in to phones thanks to the reduced chip count. Architecturally Medfield is not a significant overhaul – we’re still looking at a power optimized in-order Atom – but a die shrink for the CPU and effectively two die shrinks for the PCH should go a long way towards increasing performance; the last thing we heard in this respect is that GPU performance should double, while CPU performance has not been commented on. In any case at 32nm by the end of this year, Intel will have a process advantage over its SoC competition, who will still be on 4Xnm until they transition to 28nm some time next year.

Of course Medfield is not an entire smartphone on its own. Additional supporting chips – chiefly a modem – are necessary. As you may recall, Intel picked up Infineon’s wireless solutions business back in August of 2010, giving them modem technology to go with their Atom SoCs. Down the line we’ll see Infineon-derived modems integrated in to Atom SoCs, but for now Intel is still using separate modems developed by the new Intel Mobile Communications group, which is the basis of the other major piece of news coming from Intel today.

Intel announced their first compact, low-power multi-mode LTE modem (LTE/3G/2G), the XMM 7060 platform, powered by the X-GOLD 706 baseband processor. It will begin sampling in Q3 this year, and will ship roughly a year later in H2 of 2012. The multi-mode modem is important both for Intel and for Infineon’s traditional customers. For Intel it’s something to sell alongside Medfield, while for customers after just the modem it’s going to be among the first low-power LTE modems on the market. With the additional complexity of LTE, LTE modems had to be similarly beefed up compared to their 3G brethren, which in turn can hurt battery life. Low-power modems should bring power consumption back in balance with today’s 3G modems.

On a side note, given that Intel only recently acquired Infineon's wireless group, it shouldn't come as a big surprise that the X-GOLD 706 is being fabbed out of house. Intel says it will be a 40nm product, which means it's likely being fabbed over at TSMC.

While I was scheduling my Mobile World Congress meetings I got an email request from Intel. It wanted to give me a quick tour of the latest MeeGo UI for tablets. MeeGo, as you may remember, was the combination of Intel's Moblin OS and Nokia's own efforts. While MeeGo isn't completely abandoned by Nokia, it's looking unlikely that Nokia will be a major player in it going forward considering the fresh partnership with Microsoft.

Intel is still trucking away with MeeGo and unfortunately appears to be retaining the less-than-ideal name despite the recent shakeup with its partner (at least Moblin sounded respectable, MeeGo sounds like something you say before using the bathroom).

MeeGo is designed to be a mobile OS that can be used across a wide variety of devices. Cars, netbooks, smartphones, tablets and even TVs are supposed to be built around the totally open OS. Intel hopes that MeeGo will be the truly open alternative to Android. It's a lofty goal to say the least. Google officially introduced Android 3.0 earlier in the month and devices based on it are expected to be shipping in the next couple of months. The version of MeeGo Intel demoed for me however is a meager 1.2. While I know that comparing version numbers isn't the most scientific thing in the world, it's the easiest way for me to point out that MeeGo is no where near the maturity level of Android.

The point of today's demo was to showcase the foundation of MeeGo's tablet user experience. Intel's focus here is multitasking. What Intel wants to do away with is the concept of going back to an app launcher to do something else with your tablet. Instead, Intel believes you should be able to launch apps based on what you're doing. It's a subtle difference, but one that's best described by a screenshot.

This is the MeeGo tablet user experience. The screen is treated as a viewport into an infinitely wide and infinitely tall desktop. You scroll from left to right to view more panels, and up/down to view more information within a panel.

Each panel is grouped according to an overall function. The My Tablet panel includes your top applications as well as device settings for when you absolutely need to do something the old fashioned way. The Friends panel aggregates all of your communications between you and your friends. This could be in the form of tweets, Facebook status updates as well as emails and other messages. Everything appears in this one view. If you tap on any of the items in the view, the associated application will launch (e.g. tap an email you received from one of your friends to fire up the email app).

There are also panels for music, photos and websites.

Intel believes that launching applications to later access data with them is silly and you should instead be presented with the data you want and it should launch any necessary application for you. It's a shift from the app centric model of today's smartphone/tablet OSes to a data centric model. Intel feels that this approach will reduce the number of taps necessary to efficiently multitask, which will obviously encourage heavier usage models and ultimately require faster SoCs to run everything.

If the infinitely wide/tall viewport sounds a lot like Microsoft's Windows Phone 7 UI then you're not alone. The concept seems very similar, although it's not nearly as pretty/polished in MeeGo today. For not supporting full blown multitasking out of the box, Microsoft's WP7 UI is very efficient at moving between apps. If MeeGo can duplicate that efficiency and expand upon it, I'm interested.

Microsoft's Metro UI

The bigger issue I see with MeeGo is the huge disconnect between Intel's aspirations and the current state of affairs. If Android is the target, MeeGo needs to make a great deal of progress in a relatively short period of time. These consumer facing smartphone/tablet OSes have to be ridiculously polished, they need to make mechanical toasters look difficult to use, and MeeGo just isn't there yet.

MeeGo currently supports Swype

I like the concept. A truly open OS that allows for infinite customization and supports everything from smartphones and tablets to PCs and TVs. We need unification. The last thing I want is to have a HP tablet that can't run the same apps as my iOS smartphone and my Google TV. Unfortunately for Intel I feel like the deck is stacked against it on this. Intel does an amazing job executing on its microprocessor designs, but its extra curricular activities are rarely met with a similar amount of success.

I'm all for more competition so let's see where MeeGo takes us over the coming years.