Tag Archive | "arm"

Report: Mobile Computing at CES 2011 – The X-Over Year

Tags: , , ,


We’ve just recorded Meet:Mobility Podcast 62. If you haven’t tuned in, please do because JKK, Sascha and myself spend a long time talking in detail about what happened in mobile computing at CES and give you a good overview of the significant products we got to understand while were in Las Vegas. For me, the show brought me the biggest signal yet that the X-over is happening. That’s X86 and ARM platforms crossing over in the mobile and personal computing space.

We’ve seen many indicators before now that ARM-derived processing platforms and operating systems were capable of personal computing tasks. I tested the Compaq Airlife almost  a year ago (Fully Reviewed in May) and the Tegra2-based Toshiba AC100 that I still have for testing is everything needed for a good smart-book / PC experience except the software build but there hasn’t been a time when so many top-tier manufacturers have shown the same confidence by bringing out multiple X-over products. In that respect, CES 2011 is a very important year and I do believe that we’ll look back and say, yes, that was the start of the crossover.

Palazzo to CES Convention Center

It will be a turbulent year or whirlwind activities. We’re rising out of a depression and there’s new confidence that risks can be taken. Many of the products we’re seeing won’t’ succeed either due to being too early or by being side-swiped by other disruptive products.

Tablets came-of-age at CES 2011. The rising quality of devices and the number of top tier brands shows that there’s a big enough level on investment now that the segment is unlikely to fail to produce multi-million sales. Estimates range up to 40m units for 2011 which will match netbook sales. I agree. 40m is achievable, especially as prices drop like a ton of bricks.

06012011391

We’re not just talking about tablets though. There are a whole list of products we need to mention.

Motorola Atrix. Taking the prize for most-talked about device at CES is this dual-core Android phone with a big battery and lots of connectivity. The laptop ‘dock’ turned it into a desktop that made people stop and think. This idea of modular computing is exciting but there are lots of issues to consider. I have a Tegra 2 smart-book running Android and while it’s fun, it’s not productive. Processing power is short of what is needed and the apps are limited. Despite a full Firefox build being available on the Atrix, the limits I’ve experienced on the Toshiba AC100, will also apply to the Atrix until Honeycomb and a lot of ISV investment, solve the problem. I also imagine the cost of that set-up to be getting close to $800 or more. Who’s going to invest that much into a system that still won’t do 100% of personal computing activities? There’s also the issue of putting all your eggs in one basket. Smartphones have a tendency to get lost! The Atrix is a cool product and shows very clearly how desk-top computing will be possible with smartphone cores.

ASUS Eee Pad Transformer. Like the Atrix, this is a device aiming to be more than one computing solution. Unlike the Atrix, this one offers the tablet as the screen and computing core with a docking keyboard finishing off the ‘smart book’ look. The smartphone ‘core’ isn’t so obvious and this isn’t a device aiming to be totally convergent. In fact, it feels to me like it fits in with it’s target audience in a more comfortable way than the Atrix. A consumer, coffee-table tablet with an optional keyboard for ‘getting things done’ sounds perfect for the iPad generation. With Android offering great in-cloud synchronisation, a two-device Android strategy could work well, especially as this product will get the important Honeycomb operating system update. [More info on the ASUS products at CES available here]

Gemtech Zeus (1) Gemtek. Highlighting the progress that Intel are making with their non-Windows platforms is the Gemtek Zeus. While the product is targeted as a media phone, it’s worth stopping for a moment to look at exactly what’s happening here. This is Android version 2.2 on an Intel Moorestown platform. That’s an official Intel Android build (that is likely to become an official branch of Android) optimised for Intels always-on platform. The product is light, is said to have good battery life. It highlights just how close ARM and Intel products are in the mobile space – and this is just Intels first attempt!

OLPC XO-1.75 – One Laptop Per Child product was another win for ARM this week. Starting with an AMD CPU and then moving to VIA, another X86 CPU, they have now switched over to a Marvel ARMADA 610 ARMv7 core for the XO-1.75 which should go into production in mid 2011. The difference in battery life is likely to be very significant because in the past, they were using relatively old X86 CPUs. Always-on, screen-off standby is also now possible. It will be important to see the performance too. Compared to the very old AMD Geode CPU that was used in the original, there shouldn’t be any noticeable difference at all and yet the power envelope has been cut from 5W to 2W! There’s an interesting video available on this here. OLPC CTO Edward J. McNierney says in the video that the performance is now better!

xoom2 Honeycomb – Finally, the gun has been fired for a true large-screen version of Android. This is likely to be a branch of Android that will run parallel to version 2.x but the important thing is that it signals Googles commitment which, in turn, with give ISVs the confidence to invest in larger Android application projects. Serious productivity apps, video editing and ‘HD’ versions of existing applications. Motorola and Nvidia were the big winners as they have been chosen to provide the reference hardware and product. Soon after the Xoom launches you can expect to see more Honeycomb product announcements that will roll in the second half of 2011.Finally, we could see a productive ‘smart’ book although don’t forget that Intel will also be involved here. Honeycomb on i86 is announcement I expect to hear about soon.

Windows and Office on RISC SOCs. This was a huge announcement that gives ARM partners a reason to take a Cortex A15 license if they haven’t done so already. I’m not expecting to see a mobile product drop out of Microsoft onto ARM but again, it gives ISVs reason to create ARM versions of applications. That effort could spill over into Honeycomb-related work too. Interestingly, it puts Adobe in a great position as a runtime that will work across all of these platforms and operating systems. They could find themselves being used as an important bridge. Timescales for Windows, timescales for drivers, timescales for ported software are all in the 2-5 year timeframe although X86 emulation could speed that up? Developers could be given virtual ARM SDKs to aid development work which would explain why Microsoft took a full ARM license this year.

One of the interesting things about Windows on ARM is that, finally, it will give everyone the ability to benchmark ARM against X86 in like-for-like products. My money is on Intel having the processing power advantage and ARM offering battery life and price advantages. Differences, however, are likely to be minimal and it could all be decided on value-add features like security, wireless integration and application stores. OEDs are the ones that will make the decision here.

IMG_6338 Angry Birds on AppUp. Angry Birds migrated from ARM to X86 this week as Intel announced that the popular phone game was available on their AppUp store. That makes it available to some 100 million netbook and notebook customers and will have driven a large number of installations of the AppUp store that doesn’t yet come pre-installed on netbooks. Clearly Intel have bought-in the app to drive adoption but even so, it’s great to see and it won’t take many more of these wins before AppUp starts to drive its own adoption. Video demo here.

I was surprised not to hear any news concerning Windows 7 Compact. Microsoft still don’t have a consumer internet device operating system for the 4-10” segment. What’s going on there?

06012011412

Intel’s Oaktrail surprised me at CES. I saw a number of WIndows-based products that were significantly smaller than I expected and had logner battery life than I expected. At least the claims of battery life seem to be good anyway. Viliv, a company that has both an ARM/Android and Intel Oaktrail/Windows product in the same 7” screen casing proved that parity has almost been reached. The X70 Slate is some 35% lighter than the previous model and even increases the battery life from 6 to 6.5hrs. I can’t wait to see the performance on both Intel and ARM versions. The Samsung Gloria/PC7/TX100 was also an interesting product in terms of technology, size and battery life on Intel. Ocosmos are also working on an Oaktrail device. It’s tiny!

Nvidia announced project ‘Denver’ which aims to bring ARM to the desk-top. [More info] I suspect this is a Cortex A15 project and won’t see the light of day until 2013 but once again, there’s the confidence and investment in a crossover product. These are hugely expensive projects so the message is clear now – the risks are low enough and potential gains are high enough to get these projects underway.

Finally, there was another signal that crossover is starting to happen. I used a Galaxy Tab A LOT at CES. Wifi and 3G internet was hard to come-by but as my PIM, note-taking device, map and Twitter device it worked perfectly and preserved my phone battery , a Nokia N8, for photos and those voice-type things some people do! SMS were also handled on the N8. The netbook was with me most of the time and, like now, there’s no easier way to get a lot of text in a blog and video edited and posted. For bum-on-seat activities, I still need Windows but I surprised myself just how much I used the tablet. You’ll see me use it a lot more at MWC next month.

What Google’s Chrome OS Means For You.

Tags: , ,


Google finally launched Chrome OS at a press event in San Francisco yesterday after first introducing the concept back in July of 2009. Its a straight forward idea, your browser is the operating system and you use web applications for your daily needs.

On the surface Chrome OS is virtually identical to Google’s Chrome browser but actually runs on a stripped down Linux core which promises to be lightweight and efficient.

cr48_top

As part of the announcement Google opened a pilot program to test Chrome OS on an x86 unbranded notebook called the CR-48. The Intel Atom based 12.1 inch notebook won’t be on sale to the consumers but does come with a speculated 8 hours in use battery life and an impressive 8 days standby. Whilst the CR-48 certainly doesn’t come under the Carrypad coverage I’m certain as the platform matures smaller ARM based devices will be available which will bring better portability, power efficiency and of course the all important full desktop browsing experience which we talk about so much. Couple this with the new Chrome Web Store providing web applications for both Chrome OS and browser and we could be provided some competition to the developing MeeGo ecosystem.

So, What Google’s Chrome OS means for you?… Nothing yet but watch this space.

Report: Timeline for MeeGo Netbooks, Tablets and Smartphones

Tags: ,


Just a few days after the first ever MeeGo conference I have the best chance ever to take what I’ve seen, heard and learnt to try and predict when and what will happen with MeeGo in 2011. When will versions ship? On what hardware? When will the applications store ship and finally, when will end-products ship both via Intel and Nokia funding and, importantly, through independent vendors.

First let’s remember that MeeGo is an offering to developers, OEMs, manufacturers and other non-end-user customers. Like Android open-source, it will comprise a complete core, a vanilla user interface and a basic set of core applications. Driver support will be limited to common open-source drivers, codecs for audio and video will be limited to free versions and there will be no applications store. While ‘hacker’ types may welcome the new OS, end-users are unlikely to be too thrilled. Journalists that don’t get the whole picture are likely to react with negative reports. Once again, remember that MeeGo on end products will be different to what you see coming out of the MeeGo project.

Let’s also remember what MeeGo is about. It’s an open-source project run by the Linux Foundation and funded by Intel and Nokia for a range of products from mobile phones, in-vehicle entertainment, TV, netbooks, ‘smart’ books and tablets. Intel need MeeGo for their new low-power platforms (Moorestown and Medfield, the handheld platforms, just won’t work without MeeGo although Android is also in the works for these platforms) and Nokia have committed to bringing their next flagship product out with MeeGo. The stakes are very very high for both companies. MeeGo will happen, products on MeeGo will happen and applications on products on MeeGo will happen. But when?

It is possible to get a product out using MeeGo today. The WeTab proves it’s possible but there’s a list of things that need to happen before ‘milestone’ products appear. I also refer to these products as ‘disruptive’ because they will be good enough to compete in the same space as the best-of-breed in their category. There’s also another category of important products and that is those that are not funded by the MeeGo partners. Here’s what needs to happen before the products appear…

Hardware platforms

For netbooks, the Intel Pinetrail platform works and is likely to be the only choice for most of 2011 and until the next generation of netbook platform is introduced. At that point we should expect a lean towards always-on and the addition of hardware video decoding and encoding. The Intel netbook platform should start to look more like the Oaktrail platform proposed for tablets. If we look at the ARM platforms, the dual-core Cortex A9 series of variants is looking interesting for ‘smart’ netbooks and focusing purely on hardware, it’s possible to build a netbook-style device on ARM today. The Toshiba AC100 is one example.

  • Timescales for Intel netbook hardware: Now
  • Timescales for ARM netbook-style hardware: Now

In terms of tablets, there’s a wide range of choices. Intel are offering Oaktrail which can support Windows for a desktop-like experience and MeeGo or Android for the always-on consumer-style experience. Interestingly Intel also offer Moorestown on which only MeeGo and Android will run to provide a consumer handheld experience right down to almost mobile phone sizes. in the ARM world we have a huge range of choices. We’ve seen MeeGo running on Ti (who work closely with Nokia and are likely to be providing the platform for the Nokia MeeGo products in 2011) and on Freescale, ST Ericsson and other ARM-based platforms. These platforms are targeted at the 4-10” segment for highly mobile devices and could potentially be used to make an ARM-based smartbook, just the the Intel Moorestown platform could.

  • Timescales for Intel Tablet hardware Q1-Q3 2011
  • Timescales for ARM tablet hardware: Now

For the mobile phone space, Medfield is the Intel platform that might appear in late 2011 (more likely 2012) and for ARM, lower-power and phone-oriented platforms are available now.

  • Timescales for Intel Handheld hardware: Q2 2011 (Oversize smartphones) 1H 2012 (smartphone)
  • Timescales for ARM handheld hardware: Now

Touch

For the phone and tablet market, touch is critical. The experience needs to be fast, multi-touch and up there with the best-in-class. For that, MeeGo 1.1 isn’t enough. MeeGo 1.2 is being built with multi-touch in mind and this is planned for April 2011. Products built on the Beta versions will not be ready for market until at least two months later so unless Nokia is doing their own private work on multi-touch with MeeGo 1.1, high quality products are unlikely before that. With the next MeeGo conference planned for end of May 2011, it is the perfect time to launch a product that will be available in June or July. Whether Nokia chooses to launch their product at this time is difficult to tell. With CDMA support not planned until MeeGo 1.3 a launch in the U.S. would have to be focused around AT&T or T-Mobile but with stronger support in Europe, it would appear likely that a separate Nokia event would happen in Europe for the launch of their products.

  • Timescales for single touch products: Now
  • Timescales for multi-touch products: Starting June 2011

Battery Life

Intel products will need to be built on Oaktrail or Moorestown for the best possible battery life experience. We are moving to a world where ‘always-on’ will be the standard. ARM-platforms are already capable of offering class-leading battery life.

  • Intel Tablet battery life timescales (Oaktrail/Moorestown) 1H 2011
  • Intel Netbook battery life timescales: (Oaktrail) 1H 2011 (Cedar View) 2H 2011
  • Intel smartphone battery life timescales: (Medfield) 2012
  • ARM battery life timescales: Now

Security

Security subsystems need to be in place for carriers and that didn’t happen in MeeGo 1.1. I’m hearing that 1.2 is critical for carriers so it hits the same timescale as those devices relying on the multi-touch user experience.

  • Carrier security subsystems in place: April 2011. Products. 2H 2011
  • Other products (non carrier): Now

Applications stores

As mentioned, OVI and AppUp are critical. MeeGo will only ship with a basic set of applications and for the best-of-breed consumer devices and to create the developer excitement that is, in-turn, a critical part of the application store, they need to be in place with payment systems.  With OVI expected only on the Nokia devices (Question: What application store will be available for ARM-MeeGo devices that are not from Nokia?) we know that it will be a Q3/Q4 timescale. For AppUp on MeeGo we are seeing some marketing campaigns starting now. Launch is likely to be on the Netbook platform first in order to capitalise on the existing Windows-based AppUp store and to enable MeeGo netbook variants to go out of the door as soon as possible. Remember, netbooks using MeeGo will not need any support for carriers, phone stacks, touch and other elements that can only be delivered with V1.2. Considering that V1.1 is available now and that we’ve already seen proposal OS builds from Linpus, we can assume that existing AppUp partners Acer, Asus, Samsung and Dell will be bringing out MeeGo options likely to drive lower-cost netbooks aimed at entry-level markets initially. Based on that, we should see AppUp for MeeGo netbook UX available in Q1.  For tablet/handheld user interfaces, this might not happen until V1.2 (It’s on that roadmap)

  • Application store timescale for Nokia (OVI/ARM): Delivered with first Nokia handheld product after June 2011
  • Application store timescale for non-Nokia/ARM: Unknown. Currently no support
  • Applications store timescale for Intel / Netbook: Q1 2011
  • Applications store timescale for Intel handheld/tablet: 2H 2011 (After MeeGo v1.2)

Applications

Applications are starting to work their way through already. In the MeeGo release itself, media player, email client, calendar, sync, browser and other applications are already being worked on and there are rumors that KOffice will also be offered but as any smartphone user knows, discovery, sharing, enhancement and customisation through 3rd party applications is critical now. There is already a way for developers to make apps for MeeGo on Intel and ARM ‘target’ devices(SDK available here) but there is no support for the Application store yet. Intel have already set up an AppUp/MeeGo portal though.

Preparing applications for OVI is another story. Ovi is accepting Qt applications which will work on some existing platforms and the Nokia MeeGo products when they are launched.

Monetisation (OVI, AppUp) will stimulate the developer ecosystem and this will happen in the timescales shown above.

Finally – When Will We See A Competitive Product?

What we see coming out of the MeeGo project is a demonstrator. It’s a complete core with a functional user interface. It’s not what we’re likely to see on end products. In order to make a competitive product; One that has potential to seriously distrupt sales of other devices in the sector requires all of the above milestones to be met. Hardware, Application store, touch, battery life and something we haven’t spoken about yet. The customisation, optimisation and branding process takes months and for a class leading product, could easily take 6 months. Adding in codecs, optimising and branding the content stores, optismising the base applications, checking security, spicing up the interface, writing the drivers and testing is a 6 month to 1 year project. Lets assume that with 1.1, the teams were able to start the process of building a product around MeeGo. In April they will get the features needed to finalise the product and then, along with the integration of an application store, you’ve got another 2-4 months of work ahead. The first competitive products, driven by investment from the core partners, will only hit the market in June 2011 at the earliest. For products from other vendors, expect that timescale to go into Q3 2011 because they will definitely hold back to see what Nokia do, what Apple do, how Android develops, how Chrome OS develops, how the MeeGo application stores grow and even, how Android on Intel develops. The first MeeGo products, in all categories, need to be very special to secure trust from external companies. (Note: It’s likely that Intel and Nokia will invest huge amounts in external companies efforts to get MeeGo-based products to the market.)

Note on In-Vehicle0Entertainment and TV

I haven’t covered these two categories in this article as we’re focusing purely on the mobile/handheld/netbook computing market here.

Summary

We could see MeeGo netbooks with AppUp as early as January with ‘features’ such as quick-boot, lower cost, a simple-to-use operating system with a social-networking slant. We’re unlikely to see too much excitement around these early devices though because platforms and applications need to develop to create products with any major selling points. ARM do have an opportunity to get MeeGo on a netbook-style device in order to create an interesting long-battery-life product.

Tablets could appear in the early part of 2011 as 3rd-parties are already working on UI solutions based on MeeGo 1.1 but for interesting multi-touch products, with an application store, this won’t happen until around June 2011.

The first MeeGo smartphone requires MeeGo V1.2 and won’t happen, either on Intel or ARM until around June 2011. That phone is likely to be a Nokia product and its success will be critical to MeeGo.

Everything up until this Nokia/MeeGo phone can be called Phase-1 – led by Intel/Nokia investment. If these products show class-leading features and the developers start to create applications then we’ll start to see Phase 2 products created through independent investment that are true indicators of MeeGo momentum. That story starts in Q3 2011.

Footnote: All timescales are estimates based on current knowledge.

The next MeeGo conference has been announced for San Francisco on May 23rd-25th 2011.

Want a chance to learn more about AppUp? I’m at the Apps World expo in London on the 1st Dec and AppLab in Berlin on the 3rd Dec.

Coming To You from Ubuntu on the ARM-based AC100. (Update: Video)

Tags: ,


I’ve installed Ubuntu on my Toshiba AC100 ‘smartbook’ and I’m accessing my WordPress back-end via Firefox 3.6. This is a test to see if I can create and post an entry.

You should see a photo on the right (uploaded from the filesystem on the AC100)

Wifi is obviously working and considering i’m running this from an SD card the experience isn’t too bad. It’s locking-up from time to time as the OS works with the filesystem but i’m seeing some quite impressive CPU-related performance.

An interesting example of the performance is the SunSpider result i’ve just got from Firefox 3.6. Its the fastest result i’ve ever seen on an ARM-based device.

Firefox SunSpider Test on AC100.jpg

Click to enlarge that image and click here to view the gallery i’m creating as I go. (You’ll see OpenOffice is working!)

This feels like a seminal moment for productive smart-books. There’s lots to fix and improve but for a base install, i’m impressed.

Expect a video soon… Update: Video now available below.

Tricked by an ARM11 CPU

Tags: , ,


This is rather embarrassing

I’ve been a big proponent of ARMs Cortex CPU cores and have regularly highlighted that they are ‘entry level’ for any sort of Web work on an ARM platform. I’ve tested many ARM11 devices from smartphones to tablets and have never seen any browser speed or quality that would make me happy in a productive scenario. In today’s world of advanced smartphones, there’s every chance that the average consumer would notice too. The latest device that I’m testing, the SmartQ T7, is so slow that I’ve postponed my review until the company can confirm that there isn’t a networking fault. I’ve also been fairly public that the Nokia N8 with its ARM11 CPU won’t be a blazer when it comes to Web browsing and my early tests confirm it. I’ve also disputed cries that a new browser will make it better.

Imagine my surprise today when a tablet I’ve been very positive about in terms of speed turns out to be an ARM11-based device. Initial reports from Viewsonic indicated that the Viewpad 7 it would be a Snapdragon-based device but the specifications were changed and the unit will be based on the Qualcomm MSM7227 CPU. I only found out because all the other devices based on the same OED (Camangi FM600, Olive Pad, Spice Mi700) are all advertising the Qualcomm MSM7227. Viewsonic confirmed to me today that the device will now ship with the MSM7227 and interestingly, that the unit I tested at IFA and am basing my positive thoughts on was the ARM11 model. [Gulps]

Of course it doesn’t change the product one bit (unless you have an affinity for Cortex cores) but it does mean I have to be more careful with my analysis in the future. In my defence I was told that it was Snapdragon and it was actually pretty fast but I apologise for this mistake and hope you’ll respect my openness. The relevant product specification pages have all been updated. Learnt from this: Android 2.2 is an extremely efficient OS and it’s possible that Symbian^3 could also be well-optimised in terms of its latest browser software.

At the end of the day our ‘Open Review’ sessions and videos catch any failings and issues and we’ll be planning just that for the Viewsonic products (Viewpad 7 and Viewpad 10) so stick with us for the real story.

Note: I’ve also added a bunch of early review links to the Viewsonic Viewpad 7 product pages.

ARM’s A9 goes to 2Ghz in NuFront CPU Targeting the Desk.

Tags: ,


IMG_5061

While we wait for full tablet and smartphone implementations of the Tegra 2, dual-core Cortex A9-based CPU, NuFront have been working on a high-end implementation aimed at every-day computing.

The NuSmartâ„¢ 2816 Computer System Chip is testing in various versions up to 2Ghz. Average power drain (not idle, not TDP) is 2W which puts it in the same envelope as a 1.6Ghz Z6xx Atom CPU (TDP 3.7W)

At a demonstration of the CPU we watched hi-def movies and Android running smoothly but were left with the impression that this chip would work better with a slick desktop-style operating system. In terms of raw power you’re looking at a netbook/nettop style experience although without the possibility of Windows desktop operating systems, it will need some strong marketing and targeting to slightly tighter niches.

The final versions of the NuFront 2816 are expected soon with products following up in Asia shortly after that. We’ll be very interested to see if someone can build on the smartbook theme with a super-light Ubuntu-based solution. Given that Android isn’t ready for these devices, the Ubuntu solution could be better.

Press release PDF is available here.

ARM Cortex A15 – 5x Performance for More Than Just Mobile

Tags: , ,


Just one week before Intel’s IDF will detail more about their mobile CPU roadmap for 2011/2012, ARM have announced some details about their own 2012 high power mobile silicon. Cortex A15 (Codename: Eagle)

Based on a 28/32 nm process (ref: news about 28nm process here) this quad-core design will clock up to 2.5Ghz. That’s 4x cores and 2x clock compared to todays ‘best of.’ ARM say it will provide 5x computing performance. You’ll see it run up to 1.5Ghz in smartphones but I suspect there are some other ‘smart’ opportunities in the netbook space here.

‘Eagle’ Cortex A15 News release.

cortexA15

With the dual-core A9 already reaching up into Intel’s performance space, solutions based on this Cortex A15 will put it directly head-to-head with Intel’s 32nm Medfield platform.

Expect similar power and performance profiles for A15 and Medfield’s Atom core which means the choice of CPU wont be that important for customers any more. The choice lies with the software, designer and, increasingly, the marketing and customer-relations teams.

Intel and ARM have 2 years to get a suite of operating systems ready for these products that range from real-time, through mobile to desktop productivity. I see ARM having a lot of potential with Android but with a lack of productivity applications and a potential patent problem, Intel are in a good position with MeeGo. Having said that, MeeGo will run on ARM CPUs to so in my eyes, what we have here is simply a great range of platforms and operating systems lined up for an amazingly mobile 2012.

A15

Roadmap for Eagle web

More detailed information available here.

Via Slashgear

Expect more details at ARM’s Techcon event in November.

The Question Marks That Remain over Q4 Tablets.

Tags: , , ,


smartdevices Bob Morris, head of the mobile computing division at ARM, is telling us that the Dell Streak is just the first in a line of more tablets that will arrive from various vendors in time for Christmas. I guess if anyone should have the inside info on this it’s Bob so it’s a good sign.

We’re clearly looking at Android as the de-facto Q4/Q1 2011 operating system solution for most of these tablets and although Froyo with Flash 10.1 is a great starting point, there’s still a significant number of big question marks that keep me sceptical. I know Nvidia, ARM and others have talked about waiting for Flash and ‘fall’ but there’s more to it than that.

How about Google Market? This is becoming more secretive than Adsense or Google’s Search algorithm and one wonders just how much money Google are now making from it. The Dell Streak got Market by being a large well-branded company that effectively designed a Android smartphone but what about the others? Every device that didn’t have marketplace/Gmail/contacts/maps so far has been highlighted as an incomplete Android product. Sideloading and 3rd party app stores aren’t the fix either. The second problem is that there needs to be a new suite of >=WVGA, large screen (mdpi-large in Android speak) apps before the first reviews start otherwise the whole Android tablet ecosystem will be tainted with poor early reviews. Bad news never seems to fade from search engine results so Google needs to re-build their app suite for mdpi-large (or even mdpi-maxi as ‘large’ only goes to 5.8” screens.) If Android is to have a chance at getting more productive applications in the store (as Apple have already done) Google also need to give developers a chance to prepare new versions of their apps. That can only happen if Google stimulates the developers by announcing Android 3.0 or a new phase of tablet-focused work. Give us a sign Google. Apple gave some devs a three-month head start before the product was launched. Although this was a restricted program, it was instrumental in creating a good day-1, week-1 buzz.

Link: Overview of tablets available, announced and expected

If I was an Android Tablet OEM right now I’d be considering waiting for even more than the above.

  • Cheaper Cortex A9 platforms and proven Android hardware builds. Cortex A8 is still good enough but to make a serious marketing splash, dual-core A9 is now needed.
  • Clarification on what the hell is going on with Chrome OS (touchscreen support looks likely)
  • Concrete information about Android 3.0 (Apps suite, developer take-up, information about ARM-optimised kernels)
  • IDF (Sept) and MeeGo 1.1 (Oct) (To asses competing product timescales)

Racing to get a product out for Christmas sales could be too risky and the whole ‘smart’ tablet market could suffer if a big name gets it wrong. Like you, I want products NOW but i have the feeling that the iPad will be a year old before we see any serious competitors.

Ti Announces License for ARMs ‘Eagle.’ Cortex A10 / OMAP 5 for 2013?

Tags: , , ,


hdr_ti_logo No-one really knows much about ARMs Eagle CPU design yet. I spotted it on a slide at Computex earlier this year but the only real info available is from a presentation back in 2009. It’s the next-next generation high-end core and likely to be produced in a 28nm process but based on the standard lead-time for an ARM architecture to get into an end product I’d say we’re looking at 2012 and 2013 for this one. Cortex A9 hasn’t been introduced in any products yet and Cortex A8 still has a way to go too. To give you an idea of the end-end timescales for a new CPU design, Ti are only now announcing that they have the license. The deal was made in 2009!

If we project our thoughts 3 years into the future we’re going to be in a time-period where Intel CPU’s are likely to be in smartphones and ARM are likely to be in laptops. We could be looking at MeeGo and WebOS or something completely new. Google OS perhaps? Will Windows still be as dominant or will Android or IOS develop to become a truly productive operating system? I wonder if I’ll still be blogging?

Ti’s announcement.

MeeGo Demolishes Android 2.2 in Javascript Test

Tags: , , , ,


Note: Article title is a tounge-in-cheek reference to a recent ARSTechnica article. See below for the real story. It’s actually all about the browser engine and platform, not the OS.

As I begin a series of tests on the Aava prototype phone today, the first thing I wanted to get out of the way was a Sunspider javascript test. It’s CPU-intensive and forms an important part of the chain of events that take place to get a web application displayed on your browser. Yes, it’s one of many variables but it’s a good indicator of CPU performance.

I’ve been recording SunSpider tests for a long time now (feel free to contribute to that list) and have been very impressed by the way that both hardware and software improvements have brought the figures down. In the ARM-based world, the best-in-class devices are producing SunSpider results in less than 10 seconds (iPad, iPhone 4) with some new devices even reaching down to sub 6 seconds [See title reference.] In the world of Intel, netbooks are producing Sunspider results of about 2 seconds (using the latest Chrome build.)

SunScript - Netbook_ Morestown Phone.JPG

Click to enlarge.

Above is the result for the Aava phone I’m testing at the moment. 4215ms. It runs on Intel’s Moorestown platform at 1.5Ghz and uses the same CPU as you find in netbooks. Running Sunspider on the latest Firefox build on the netbook and the Firefox Mobile build on the MeeGo-based Aava phone you can see the difference is marginal. 4.2 seconds for a prototype phone is quite impressive. The figures confirm the CPU speed and also confirm that the sunspider score is likely to drop to around 2 seconds with the latest Chrome or Android browser build. Try Sunspider on your phone or PC browser to see what score you get. For reference, I use a ultra mobile PC as my desktop PC. It runs a 1.6Ghz Atom CPU and returns a score of about 4.1s. True desktop PCs will come in at below 0.5s

We really are talking PC-power here but there are two things we must not forget. 1) Multi-core ARM-based devices are round the corner too. A dual-core 1.2Ghz Snapdragon platform is likely to reach right down to the 2-3s range. 2) We must also remember that if you were to run this test continuously on these high-power platforms, you are likely to have a dead phone in just a few hours!

Tegra 2 Javascript Benchmark Shows a Lot of Promise.

Tags: , , ,


At an ARM event in Taipei this week I was fortunate enough to have a few minutes with a Tegra 2-based tablet prototype known as ‘Harmony.’ It’s a fairly standard tablet with a capacitive touchscreen and weighing in at the 700gm iPad-weight. You’ll see more details in the video below.

The interesting thing was the SunSpider javascript benchmark I did. As you might know, SunSpider is a well-recognized test of a browser’s ability to run javascript and it serves as a good data point for working out how fast a processor is. As far as I know, most, if not all, the javascript processing is done by the CPU.

You’ll see a result of 9.8 seconds on the video which is about the same as you’d see on the iPad. The iPad uses what we believe is an ARM A8 core, or at least something very similar. However, that doesn’t mean that  dual-core Tegra 2 is only as powerful as a single-core A4 CPU because the two browsers are vastly different. The javascript engine on the iPad is super-fast where the engine used on the Android 2.1 browser isn’t.

Harmony SunSpider

  • Android 2.1 on Tegra 2 – 9 seconds (Tegra prototype shown in video)
  • Android 2.2 on Snapdragon 1Ghz – 6 seconds. (Nexus One, Google V8 engine)
  • Android 1.6 on Snapdragon 1Ghz – 24 seconds (Xperia X10)
  • Android 1.6 on Snapdragon 1Ghz -  54 seconds (Dell Streak prototype)
  • Chrome on Atom 1.6 – 2 seconds (average netbook, Google V8 engine)

As you can see there, the Android browser is at least 4 times less efficient in processing javascript on version 1.6 than it is on 2.2. I suspect that version 2.1 is close to 1.6 in its efficiency too which means that the Tegra 2 is over 200% faster than Snapdragon. Given that it has two cores, it’s not surprising. A single A9 core (in my estimate) brings about 20% more raw CPU performance over A8 so the figures look correct.

Add in the 4X improvement that you’ll see with Froyo’s Google V8 processing engine and there you go! Tegra2 with Froyo will be able to process javascript as well as a 1.6Ghz netbook. Just imagine what the 1.2Ghz dual-core Snapdragon will do!

More information about high-end ARM platforms in our primer, here.

We’ve already seen that ARM can beat Atom in certain low-end Atom scenarios and cross referencing to some other figures I have seem to confirm that ARM is going to challenge Atom in raw CPU. If only they could sort the operating system out and get some productivity apps going, we’d have some interesting smartbooks out there.

Chippy Sidenote: 2 cores needs 2x power! It would be interesting to work out the CPU drain figures under these javascript test conditions. Also note that system performance is not directly related to CPU performance. A lot of work has to go into a lot of other hardware and a lot of other software to get a system running ‘fast.’ Please remember that when Intel and ARM enter the clock-speed game over the next few years!

ARM Press Conference. ‘Eagle’ leaked as next-gen core.

Tags:


A summary of the ARM Press conference at Computex today. Speaker: Tudor Brown, President of ARM

 Capture_00001

 

We’ve become used to ‘always on, anywhere.’

ARM go on to highlight the number of devices in it’s ecosystem.

Capture_00010

‘Eagle’ appears on Roadmap above Cortex A9.

Update: I asked what it was and ARM confirmed it’s the next-gen core for high-end platforms.

Capture_00011

Capture_00014  

ARM goes on to highlight where ARM fits in in the home entertainment ecosystem and talks about the Digital Connected TV.

Mali appears

Capture_00016

Highlights Android and Froyo and the ARM-specific optimizations.

Capture_00017

Q&A

I asked about Android for the emerging tablet space and the lack of . The response from ARM is that Google a looking at the option of marketplace and Google apps ‘very seriously.’

That’s it for now. I’ll try and get an interview later.