Samsung Exynos 5 Octa SoC Uses PowerVR GPU Instead Of ARM’s Mali GPU

One of the highlights of this year CES was the announcement of an Octa-Core mobile SoC from Samsung – the Exynos 5 Octa. This absolute beast of an SoC makes use of ARM’s big.LITTLE implementation and packs 4 Cortex-A15 based cores, and 4 lower-power Cortex-A7 based cores, manufactured on the 28nm HKMG process.

The use of ARM’s big.LITTLE technology will allow this SoC to provide the best of both worlds – performance and battery life. On the GPU side, it was assumed that Samsung went with ARM’s Mali GPU, which was found inside the Galaxy S II and the Galaxy S III.

However, it looks like Samsung may have had a change of heart and has instead used PowerVR’s Imagination GPU on its latest SoC. AnandTech has confirmed from multiple sources that the Exynos 5 Octa SoC packs in three cores of PowerVR SGX543 GPU, with each core clocked at 533MHz.

This is the same GPU that Apple uses in its A5 and A5X SoC, with the latter being found inside the third generation iPad. Considering the PowerVR SGX543 inside the Exynos 5 Octa SoC is clocked so high, it should be able to beat the 3rd generation iPad in benchmarks. However, it won’t be able to surpass the performance of the A6/X SoC found inside the iPhone 5 and the fourth-generation iPad.

There is a very strong possibility that Samsung will end up using the Exynos 5 Octa SoC on its upcoming Galaxy S IV smartphone.

Intel vs. ARM: From the World of Servers

It is a well-known fact that Intel dominates the world of personal computers and ARM owns the world of mobiles. Moreover, while Intel has taken its first baby steps towards making mobile processors, ARM has also made a slow entry to the personal computing world. Nevertheless, there is another market where this Intel vs. ARM war can go, and it is the market for servers.

Recently, ARM started showing its head in the server market and as it turns out, there are some impressive reasons for using ARM over Intel for servers. Look at this presentation for some interesting facts on ARM servers. In a few months, Dell and HP are going to enter the ARM server market with all guns blazing. While Dell showcased an ARM based server product back in May, HP has the advantage of being the early adopter. Things have started moving in this competition, and to spice things up, our beloved Ubuntu is ready for this ARM revolution, from version 11.10.


However, the biggest piece of news came yesterday when Intel hit back HP’s ARM server partner Calxeda over a benchmark. According to a benchmark carried out by Calxeda, ARM’s Cortex A9 based ECX-1000 servers could outdo Intel’s Xeon processors in performance, and Intel was not ready to accept this. Therefore, it went ahead, performed its own benchmarks, and came back with another set of results in its own favor. Apparently, both Intel and Calxeda used Apache HTTP benchmarking, but Calxeda used Intel’s Sandy bridge Xeon E3-1240 processors to carry out the tests instead of the latest Ivy Bridge. Additionally, Calxeda limited the Xeon for I/O for the benchmark, and effectively, the CPU was running only at 14% of its total capacity.

Intel and ARM are fighting on multiple fronts, and this is just another battle in the ongoing war. While Intel has the advantage of high performance, ARM has the advantage of power efficiency. They are both masters of their fields, and this war will only leave us spoilt for choice.

No, Microsoft’s Entry-Level Surface Will Not Cost $1000

Microsoft’s announcement last month that they’re making their own tablet hardware — and directly competing with the very OEMs and partners that they license their software to — was huge in revealing a radical step forward for the company.

However, while it certainly is an important milestone and turning point for Microsoft, the event lacked plenty of important details pertaining to the product itself. We’re still in the dark about exact pricing and availability information, among other things, leaving many intrigued and mystified about key factors that could make or break the device.

Yesterday, WPCentral spotted that Swedish online store Webhallen listed the Microsoft Surface on their website, which sported some ludicrous pricing; the entry-level ARM-based 32GB Surface is priced at 6990 NOK, or roughly $1150 USD. What fascinates me is that there are posts aggregating this “story” that don’t immediately ridicule, but rather entertain the idea, as though it’s even plausible. On top of the outlandish pricing, we already know that the Surface will only be officially available through Microsoft Stores (and their online outlet.)

I also reached out to Webhallen, and they issued a comment stating that existing prices on the site are not based on any word from Microsoft whatsoever:

Our customers are very interested in pre-ordering these products, so we have set a high preliminary pricing for the lineup so that they may be able to pre-order them.

Just to clarify, we have not recieved any pricing from Microsoft regarding MRSP or purchasing net cost, and any people who have booked the Surface at this high price will of course have their order adjusted before any product is shipped. So we’re not going to overcharge anyone for being an early adopter.

I understand that Microsoft does some pretty unusual things, but they’re not batshit crazy.

So in conclusion, here’s a recap of yesterday’s highly credible blog posts: Microsoft, the company that needs every advantage it can get to even gain a smidgen of ground in the tablet market will charge a few hundred dollars more than a 32GB WiFi+3G iPad for its entry-level model, and upwards of $2000 for an Intel-based Surface Pro which is essentially an Ultrabook/Macbook Air competitor. Riiighhttt.

[Post updated with comment from Webhallen.]

Why Does Windows 8 ARM Still Have The Desktop

When Microsoft showed Windows 8 on ARM there was a lot of enthusiasm. It was clear that Windows on ARM is Microsoft’s attempt of developing an OS for tablets that will consume less resources, have a long battery life, and provide a nice UI/UX.

One of the biggest concerns raised by critics was the presence of the traditional desktop in the tablet version. Like many, even I’d like to see Windows 8 tablets to not have the desktop UI. Office 15 shows why I don’t want the desktop interface on Windows tablets–it’s schizophrenic. Having to switch between the tablet and desktop interface is confusing and at many times annoying. Behind the intuitive Metro interface and Metro inspired apps lies the dormant ghost of Windows waiting to be unleashed every few hours when you use your tablet, haunting you as you hold that tablet.

Steven Sinofsky in his post about Windows 8 on ARM explains why the design decision was taken to include the desktop interface. Long story short, to give you the best of both worlds. Sinofsky and Microsoft believe that there are features and capabilities in Windows that users have become familiar with over the years and would want to have them going forward as well. Sinofsky explains:

[…] Enabling Windows to run super well on the ARM architecture is a significant engineering task. We undertook this work because when you look to the future you can see that so many of the capabilities that have been added to Windows over the years are things that customers will inevitably desire or require in the types of devices supported by today’s ARM-based products—changes in form factors and the desire for mobility only add to the scenarios and capabilities we all desire in our search for no-compromise PCs. While it is tempting to make bold statements about “starting over,” we believe in the evolution of technology assets when the foundation is strong. The foundation of Windows, the core, is the most solid, scalable, and secure one around. Our desire to deliver a no-compromise experience motivates our efforts. […]

[…] The availability of the Windows desktop is an important part of WOA. The desktop offers you a familiar place to interact with PCs, particularly files, storage, and networking, as well as a range of peripherals. You can use Windows Explorer, for example, to connect to external storage devices, transfer and manage files from a network share, or use multiple displays, and do all of this with or without an attached keyboard and mouse—your choice. This is all familiar, fast, efficient, and useful. […]

[…] Some have suggested we might remove the desktop from WOA in an effort to be pure, to break from the past, or to be more simplistic or expeditious in our approach. To us, giving up something useful that has little cost to customers was a compromise that we didn’t want to see in the evolution of PCs. […]

Sinofsky does make a valid argument. The iPad does not have a file explorer and is tied to iTunes. Microsoft sees the tablet as a mobile computing device somewhere between the phone and a laptop. By keeping the desktop in WOA, Microsoft is letting the end-user have the familiarity of good ol’ Windows while on a tablet.

Sinofsky says that having the traditional Windows desktop on ARM will not result in performance loss. The other way to look at it is, because Microsoft can.


ARM Sees Profits Rise 45% in Q4 2011

ARM Holdings, the company whose chip designs powers a majority of the world’s mobile devices, has posted its Q4 2011 results, and they are very much in line with expectations.

ARM’s revenue grew 21% year-over-year to $217 million, with profit before tax growing 45% to $109 million. ARM grew its operating margin in Q4 2011 to a hefty 48.2%. ARM is able to maintain such high margins because it doesn’t produce chips itself, it just designs them and licenses those designs to chip manufacturers like Qualcomm, Texas Instruments etc.

While its revenues are just a rounding error compared to Intel’s, ARM is proving to be a very tough competitor. While Intel tries to break into the mobile processor market, ARM is trying to get its foot in the door when it comes to desktop processors. With Windows 8 supporting ARM processors, ARM may be able to capture a significant share of the desktop processor market in 2012.

Warren East, ARM’s CEO said:

“In Q4 and throughout 2011 ARM has seen strong licensing growth, driven by market-leading semiconductor companies increasing their commitment to ARM technology, and more new customers choosing ARM technology for the first time. We have also seen our royalty revenue continue to grow faster than industry revenues as the ARM Partnership gains share in our target markets.

2012 will bring exciting opportunities and challenges as ARM enters competitive new markets where we are well positioned to succeed with leading technology, an innovative business model and a thriving ecosystem of Partners. As our customers are designing more ARM technology into their widening product portfolios, ARM is investing in the development of new products. These products will drive further long-term growth in our revenues, profits and cash.”

ARM’s chips are now used not only in mobile devices, but also other consumer electronics technology like digital TVs and embedded devices.

Around 1.2 billion chips for smartphones and mobile devices and 1 billion chips for embedded digital devices were shipped in Q4, implying 10% and 40% growth respectively.

Intel’s Future Depends on the Success of Ultrabooks

Intel, which has been unable to crack the mobile processor market yet, has been betting on ultrabooks, to revive notebook sales which have been slowing down, and also steal the limelight away from tablets, which have seen sales explode in the past year. In August, its venture capital arm, Intel Capital started the $300 million Ultrabook Fund to encourage manufacturers to invest in the research and development of ultrabooks.

UltrabookIts efforts seem to have bore fruit. Most major notebook manufacturers have either launched ultrabooks, or are working on them. Lenovo, HP, Dell are reportedly planning to launch ultrabooks soon. Samsung has announced its intentions to ditch the netbook business, and focus on ultrabooks completely.

Acer, Asus and Toshiba recently launched their ultrabooks, but none of them have been able to make much of an impact, as they don’t offer any significant price advantage over the MacBook Air.

For ultrabooks to go mass-market, they must first become much cheaper than they currently are. Most manufacturers are trying to bring down costs to below $1000, which is proving to be difficult due to the high cost of processors and SSDs.

A report by Digitimes says that ultrabook prices may drop by 5-10% in Q1 2012. Intel is reportedly offering a $100 marketing subsidy for every ultrabook to some manufacturers llike Acer, Asus and Toshiba, and also helping them with their marketing efforts in a bid to make ultrabooks popular.

Analysts expect ultrabooks to be available for $600-$800 in 2012. The launch of Windows 8 in mid-2012 could also lead to a boost in ultrabook sales.

With the smartphone and tablet market being almost completely dominated by ARM, and the PC market shrinking, the ultrabook market seems to be the only source of hope for Intel’s topline growth.

How Apple’s iPad Disrupted an Entire Industry

“The Enlarged iPod Touch”

When Apple unveiled the iPad in January 2010, the first reaction from most of us was one of ridicule. “It’s only a bigger iPhone / iPod Touch” is what I first thought. In a way, I was right, but the iPad was much more for most of its target user base.

It was a huge hit; Apple has sold more than 40 million iPads to date, generating more than $25 billion in revenue from its tablet business. It revived the almost dead tablet industry, which many had tried and failed to do before. Scores of Android tablets have been launched since the iPad’s launch, but most of them have failed miserably. Android for tablets – Honeycomb – was a major disappointment. Other tablet platforms did even worse. HP was forced to discontinue webOS as no one wanted to buy the HP TouchPad, not at the price it was originally launched at. Even the Blackberry Playbook by RIM has been a failure.

With the iPad, Apple has offered the best tablet experience, at a surprisingly (by Apple standards) low price. For an entire year, no one was able to beat Apple on price, which enabled Apple to capture a majority market share in the exploding tablet market with virtually no serious competition.

The Winners and the Losers

Apple has been the obvious winner, ringing up billions of dollars in sales, thanks to the iPad.

ARM has been another major winner. With the iPhone in 2007, and the iPad in 2010, Apple redefined two industries – smartphones and tablets. Even the standard smartphones and tablets available these days are much more powerful than the average computer was a decade ago. We now have quad-core tablets and dual-core smartphones, all of which have one common factor – ARM.

Most desktop and notebook processors are based on the X86 architecture, and are made by Intel. However, Intel’s chips have traditionally been power guzzlers, which is something you absolutely don’t want in any mobile device.

Almost every mobile chip is based on ARM’s reference designs, and manufactured by Qualcomm, Nvidia or Texas Instruments. Intel hasn’t yet been able to crack the mobile processor market. Its Atom processors failed miserably when it came to efficient power consumption and performance. It’s working on a new set of processors for tablets, and is trying to push the Ultrabook onto consumers, as an alternative to tablets. It even launched a $300 million fund to encourage manufacturers to build ultrabooks.

Apple iPad

Microsoft has been another casualty of the tablet revolution. It was late to the smartphone party with Windows Phone, and won’t be launching Windows 8 until mid-2012, leaving the market open for Apple and Google to conquer.

It has also been affected indirectly by the growing popularity of tablets. Desktop sales in the last couple years have almost plateaued, and even notebook sales have slowed down considerably.

As Windows sales account for a major portion of Microsoft’s revenues, and are directly linked to global desktop and notebook sales, Microsoft has seen a slump in its revenue growth in the past few quarters, which is expected to continue until the launch of Windows 8, which will run on tablets as well as computers.

According to a recent report by Bloomberg, even DRAM manufacturers have been facing losses as sales have decreased in lockstep with notebook sales.

On the other hand, flash memory manufacturers like Samsung have seen a jump in revenue, as flash memory is being increasingly used in tablets, smartphones and notebooks (thanks to the popularity of the MacBook Air, many notebook manufacturers have starting using SSDs).

Tablets may very well be the future of computing. Desktops are almost dead, and notebooks are becoming more and more like tablets – increasingly slim, portable and fast with flash storage and excellent battery life.

Winners: Apple, Samsung & ARM
Losers: Intel & Microsoft

Touch Me! Microsoft gets ready for BUILD


We were greeted to Windows 95’s launch by The Rolling Stones’ Start Me Up, a reminder of the new, but now iconic Start button in Windows. Maybe for Windows 8, Microsoft should use The Doors’ Touch Me.


We have been waiting anxiously for this day to arrive. Tomorrow, after months of keeping a tight leash (leaks notwithstanding) on the progress of or the details about Windows 8, Microsoft will reveal its newest operating system to the world at BUILD.

BUILD is Microsoft’s new developer-focused conference, a combination of PDC (Professional Developers’ Conference) and WinHEC (Windows Hardware Engineering Conference). It is being held at the Anaheim Convention Center, Anaheim, CA.

What we know

Ever since Steven Sinofsky and Julie Larson-Green revealed Windows 8 at All Things D’s D9 conference in June this year, the anticipation and expectations have gone up for what Windows 8 will be. Windows 8 sports a brand new Metro style interface with its big tiles. This interface is obviously suited to touch gestures and along with the upcoming Xbox dashboard update, it completes the trifecta of Metro styled interfaces from phones (Windows Phone 7) where it started, to PCs and TVs. Recently, Microsoft started a new blog dubbed Building Windows 8, where they have revealed (or confirmed rumors regarding):

  • Support for ARM architecture
  • System requirements for Windows 8 will be the same or less than Windows 7 requirements which means the hundreds of millions of PC’s being used today can be upgraded to Windows 8 without the need for further investment
  • The teamswithin Windows 8, which in some ways confirmed rumors such as existence of Hyper-V in the Windows 8 client and an App Store for Windows.
  • USB 3.0 support
  • New file copy/move/delete experience in Windows Explorer, along with a new conflict resolution user experience
  • Ribbon-ized Windows Explorer
  • Native support for accessing ISO and VHD files
  • Hyper-V in Windows 8 client
  • Extremely fast boot times in Windows 8

From what is explicitly mentioned in the blog and what was demonstrated at D9, we also know that Windows 8 will have two user interfaces. The first being the Metro style, tile-based, interface and the other being the classicWindows 7-style interface. Both these interfaces, Microsoft claims, are an effort to have no compromise. By no compromise, they are implying that just because an interface has touch-first design, does not mean it will not support keyboard and mouse. Microsoft realizes that a large portion of its user base uses Windows in an enterprise where the tile-based, touch-first interface may not be the most optimum. Hence, instead of ditching the past and starting afresh with the new paradigm, Microsoft is now at a stage where it has to explain how the two interfaces will co-exist. This co-existence leads to many more questions, which brings me to my next topic.

ARM Support in the Linux Kernel, the Unseen and Untold Truth

ARM support in the Linux kernel has been a debated issue for too long and today, it stands at a point where it is making more compromises. Every device with its own code for ARM support creates a bloatware out of the entire ARM section in the Linux kernel. This is a huge dilemma because if these codes are not submitted at the end of the day, it will (probably) be termed as a violation of GPL v2 and if they are submitted, they are too complex to include into the kernel. So they just lay there.
With a mini community of independent agents formed inside the Linux kernel developer community itself, these device manufacturers are finding it hard to get their ARM changes upstream into the mainline kernel. The reason?

  1. There are too many of them
  2. They are highly complex in their own way
  3. Most of them are just redundant

In short, there is utter chaos when it comes to ARM support in the Linux kernel and it was best left ignored until now.

The scenario is taking a turn and  attempts  are being made to standardize the process. ARM has moved to a separate Git tree but it still annoyed the maintainer all the more. Torvalds is rightfully annoyed here, as he would not include every bit of code that some device manufacturer somewhere has written to support some hardware that few people use!

This is a strong but a welcome decision because in the long run, it will keep hardware vendors from breaking the Linux ecosystem and acting in a more co-operative and a less competitive way.

The state of ARM in Linux kernel can still be ignored all right but we have seen how Microsoft is talking of a Windows 8 tablet now. ARM is indeed important for the future of portable and mobile computing and undoubtedly, Linux plays a major role in its future. The sooner they marry, the better it is for both of them.

Canonical Builds A 42-Core Ubuntu ARM Build Machine

With the low power ARM processors becoming very popular because of smartphones and tablets, Canonical is trying to expand the architectures which Ubuntu supports by including ARM as well. Technically, Ubuntu can be run on ARM machines as well but the Canonical servers does not build ARM packages. So, anyone who wants to use Ubuntu on ARM have to manually build them.

With Ubuntu planning to officially support ARM, they need to build packages for the ARM architecture as well. The job of building an Ubuntu ARM build machine was given to David Mandalla.

The ARM cluster server that Mandalla is building makes use of the relatively cheap PandaBoard. Each PandaBoard software development platform has a dual-core ARM Cortex A9 processors running at 1GHz and a low power 1 GB DDR2 RAM.

The ARM build machine that Mandalla is developing has 21 PandaBoards with each board connected to a 300GB hard drive. This brings final specifications of the machine to 42-core ARM processor, 42 GB DDR2 RAM and 6.3 TB of storage.

Out of the 21 boards, 20 will be used to build Ubuntu packages. One board will be used as the master board to allocate the build requests received from users to the other 20 boards.

Mandalla is documenting the build on his blog. You can read more at his blog.