Home » smartphones » ARM Cortex-A12 CPU cores and Mali-T622 GPU cores with Process Optimization Packs (POPs), plus Mali-V500 video block for mid-range mobile devices of the end of 2014

ARM Cortex-A12 CPU cores and Mali-T622 GPU cores with Process Optimization Packs (POPs), plus Mali-V500 video block for mid-range mobile devices of the end of 2014

Prerequisites (June 2015⇒):

Welcome to technologies trend tracking for 2015⇒2019 !!! v0.7
5G: 2015⇒2019 5G Technologies for the New Era of Wireless Internet of the 2020’s and 2030’s
Networked Society—WTF ??? v0.5
Microsoft Cloud state-of-the-art v0.7
• Service/telco for Networked Society
• Cloud for Networked Society
• Chrome for Networked Society
• Windows for Networked Society

Opportunity for Microsoft and its Partners in FY17:

As progressed since FY15:

Or enter your email address to subscribe to this blog and receive notifications of new posts by email:

Join 95 other followers

2010 – the 1st grand year of:

3.5G...3.9G level mobile Internet
• system-on-a-chip (SoC) and
reflective display technologies

Why viewed most (till Feb 1):

Marvell SoC leadership
Android 2.3 & 3.0
Hanvon's strategy
Welcome! or Home pages
Treesaver (LATELY #2!) and
IMT-Advanced (4G)
MORE ON THE STATISTICS PAGE

Core information:

in order to cover (very competitively) the hole existing in ARM-based SoCs so far:

Arm unveiled the Cortex A12 processor during a news conference at Computex in Taipei on June 3, 2013.

AnandTech’s judgement about the Cortex-A12 announcement:

… The Cortex A9 is too slow to compete with the likes of Intel’s Atom and Qualcomm’s Krait 200/300 based SoCs. The Cortex A15 on the other hand outperforms both of those solutions, but at considerably higher power and die area requirements. … The Cortex A15 island in Samsung’s Exynos 5 Octa occupies 5x the die area as the A7 island, and consumes nearly 6x the power. In exchange for 5x the area and 6x the performance, the Cortex A15 offers under 4x the performance. It’s not exactly an area or power efficient solution, but a great option for anyone looking to push the performance envelope. Today, ARM is addressing that hole with the Cortex A12. …
Asked at a Taipei news conference about the future of Intel’s x86 architecture, rival Arm said it still sees life in the platform.

AnandTech’s judgement about Mali-T622 and Mali-V500 announcements:

… The Mali-T622 is a 2-core implementation of the 2nd generation Mali-T600 GPU architecture that we first learned about with the 8-core T628. Each shader core features two ALUs, an LSU and a texture unit. … On the video front, the Mali-V500 video encode/decode block is a multi-core engine used for all video acceleration. The V500 allegedly supports up to 100Mbps High Profile H.264, although details are scarce on more specifics. ARM claims support for up to 120 fps 4K video decode with an 8-core V500 implementation. Mali-V500 also features a protected video path, necessary for gaining content owner support for high-bitrate/high-resolution video decode. The V500 also supports ARM’s Frame Buffer Compression (AFBC), a lossless compression algorithm that can supposedly reduce memory bandwidth traffic by up to 50%. There’s presently no frame buffer compression in Mali GPUs today, but ARM expects to eventually roll AFBC out to Mali GPUs as well.

Announcement information from ARM:

image
POP IP for the Cortex-A12 processor core 
– The only implementation solution that is co-developed along with the processor itself
– The processor RTL and the POP implementation feed off each other and are thoroughly co-optimized
– Lower the risk to end customers and those designers starting from scratch with a new processor core
– Save months of effort optimizing the implementation
POP IP for Mali-T622 GPU core 
– Eliminates the iterative guess work required to find the most optimal implementation
– Enables best-in-class PPA and frames per second metrics coupled with highly flexible implementation
More information: POP IP for the Cortex-A12 Processor: Enabling the Next Billion Smartphones [June 3, 2013]

image
New ground-up design for mid-range mobile
– OoO, Dual-Issue, 11 stage dynamic length pipeline
– Tightly integrated, high-performance NEON and FPU units
Perfectly balanced design for best efficiency
– Highly optimized L1 and L2 memory sub-system
– Ideal for current and upcoming mobile workloads
Flexible interface options to adapt for use-case
– 128-bit AMBA ACE – System coherency with CPUs or GPUs
– Accelerator Coherency Port (ACP) – I/O coherency with DMA
– Peripheral Port – For low-latency peripherals elmiminating DDR traffic congestion
More information: Cortex-A12: Diversification in the Mobile Market – Serving the Mid-Range [June 3, 2013]

image
Smallest GPU Compute solution in the market
– Renderscript Compute and OpenCL 1.1 Full Profile
50% energy-efficiency improvements over Mali-T600 series
Richest user experience with OpenGL ES 3.0
More information: Mali-T622 – Bringing Full Profile GPU Compute to mid-range devices [June 3, 2013]

image 
1080p60 HD encode/decode
Optimized for lowest cost and power
– AFBC gives 50% lower memory bandwidth
TrustZone secure video path
– Premium content protection
More information: A new branch for the Mali family tree: Mali Video, featuring the Mali-V500  [June 3, 2013]

ARM Targets 580 Million Mid-Range Mobile Devices with New Suite of IP [press release, June 3, 2013]

News Highlights:

  • Faster time to market and less design risk with suite of IP including: 
    ARM Cortex-A12 processor, Mali-T622 GPU, Mali-V500 video solution and POP IP technology;
  • 580 million mid-range smartphones and tablets are forecast to be sold in 2015
  • Cortex-A12 processor delivers 40 percent more performance than Cortex-A9 and brings premium features such as virtualization to the mid-range mobile device market; efficiency profile also makes it ideal for DTV and home networking;
  • Cortex-A12 processor brings optimum performance and maximum efficiency of big.LITTLE processing to mid-range smartphones and tablets;
  • Mali-T622 GPU offers an efficient and qualified OpenGL ES 3.0 solution and smallest Full Profile GPU Compute solution, putting even greater compute power into the hands of more mobile users;
  • Mali-V500 video IP solution reduces system bandwidth and power, while enabling the protection of premium video content with TrustZone support.

The essence is that the first Cortex-A12 based SoCs are expected by mid-2014
– for mid-range devices (smartphones and tablets) in the $200 … $350 price range by late 2014 to early 2015  
– with Cortex-A7/A15 architectural compatibity, thus in big.LITTLE configurations with either core, supporting 40-bit addressing (up to 1 TB) and virtualization
– plus providing the highest efficiency in pairing with Cortex-A7 core
– as the follow-up with +40% performance to the current SoCs for mid-range devices based on Cortex-A9 SoCs

The SoC ramp-up of about one year or so is compared to not less than two years ramp-up for Cortex-A9 based SoCs. This is the result of significant progress with Process Optimization Pack technology of ARM which was first time developed along with the processor and GPU cores themselves. It is available now for TSMC 28HPM process technology for lead partners. Six of them are already starting their SoC design. Moreover it will also be available at GLOBALFOUNDRIES 28-SLP HKMG process technology in Q4 2013. So it is also first time as such complete sourcing from two foundries will be available for SoC vendors so early on. GLOBALFOUNDRIES is even going to achieve up to 70 percent higher performance in comparison to a Cortex-A9 processor core using 40nm process technology. Competition between those 2 foundries will understandably be very strong as the 2015 mid-range smartphone and tablet market is expected to be not less than 580 million units.

In comparison the Cortex-A9 core was announced in October 2007 and released in 2008
now contributes to approximately one-third of all smartphone shipments worldwide
real development opportunities began in H2 2009 with possibility to go even against Intel Atom (source: Computex 2009 – Warren East Presentation [ARM Holdings, June 1, 2009]):
image
with improving Cortex-A9 performance on 45nm process achieved through:
– 56% improvement from processor and physical IP optimisations
– 44% improvement from other techniques
The first SoC products based on 45nm technology came in 2011, namely:
NXP PNX 847x/8x/9x set-top box SoCs sampling in January 2010. However a month later the business related to these products was sold to Trident Microsystems (see the PNX8490/PNX8491 datasheet of February 2010) and as Trident had experienced continuing operating losses it filed for bankruptcy in January 2012. Its set-top box SoC business had been taken over by Entropic Communications, Inc. in April 2012. Although only the PNX8475 is currently offered by Entropic the original Cortex-A9 related SoC know-how is flourishing quite well there (see also: 1, 2, 3 and 4).
Samsung Orion application processor, later renamed into Samsung Exynos 4210 then further into Exynos 4 Dual, announced in September 2010 for sampling in Q4 2010 and mass production in H1 2011. It first came out with the Samsung Galaxy S II smartphone announced in February 2011 for May 2011 delivery. Other Samsung smartphone and tablet products then followed.
Texas Instruments OMAP 4430 and OMAP 4440 (later renamed OMAP 4460) application processors announced in February 2009 for sampling in H2 2009 and expected production by the second half of 2010, but actually debuted a year later in February 2010 with sampling available and expected production in H2 2010. The first product based on OMAP 4430 was the BlackBerry PlayBook tablet announced in September 2010 for early 2011 availability but becoming available in June 2011 only. Smartphone products from Motorola (a lot, also a few tablets) and LG (a few) followed that, as well as a number of tablet products from Archos and most notably the Kindle Fire from Amazon, and the Nook from Barnes & Noble.

ARM is representing and projecting the evolution of the market since then as follows:image
More information about that was provided in:
Cortex-A12: Diversification in the Mobile Market – Serving the Mid-Range [ARM Smart Connected Devices blog, June 3, 2013]

Mobile devices have become indispensable in North America, Europe, and much of Asia, and are becoming the primary compute platforms for people in emerging markets. We are entering a new era of computing, the post-PC era. ARM® technology has been at the heart of the mobile revolution for over twenty years and continues to be the bedrock of all innovation and change in this space.
Mobile devices, such as smartphones and tablets, are connecting billions of people. In 2013, we are expecting:
– Over 1 billion smartphones forecasted to ship*
– Smartphones for <$50 and Tablets >$800
– Tablets out-ship notebook PCs
What becomes clear when looking at mobile devices is that we are seeing segmentation into multiple markets, which is an opportunity for growth for ARM partners:
– Premium devices: Price range > $400
– Mid-range devices: Price range between > $200 and < $350
– Entry-level devices: Price range up to $150image
Source: Mixture of ARM and Gartner Estimates
Premium smartphones and tablets receive a great deal of attention, but it is the entry-level and mid-range markets are expected to grow the fastest over the next years. ARM delivered the Cortex®-A7 processorin the fourth quarter of 2011, and it is now shipping in large volumes in low-cost, quad-core devices. It will be followed by the Cortex-A53 processor, which is soon to be released to lead partners. Both are high-efficiency processors, that are efficient by simple in-order eight stage pipelines which are highly efficient and tuned to deliver very good performance for their size. In the mid-range mobile device market, the industry had tremendous success with devices based on the higher-performance Cortex-A9 processor, which uses a partially out-of-order, nine stage pipeline to achieve high performance tuned to the power constraints of smartphones. The Cortex-A9 processor was released in 2008 and now contributes to approximately one-third of all smartphone shipments worldwide.
The market segmentation is driving the diversification in mobile and resulting in many different requirements needed to achieve the highest performance and lowest power within a sustained thermal envelope. These requirements make it mandatory to provide the functionality previously available only in premium devices, but within the power budgets of mid-range devices. Looking at how to serve those markets, it is clear that one size does not fit all anymore.
Today ARM is introducing the Cortex-A12 processor, the highest performance mid-range CPU that is specifically designed for the next-generation mid-range mobile market. The Cortex-A12 processor brings its own mix of high performance and energy efficiency to 2014 SoC designs: more performance than the Cortex-A9 processor with the same mobile-tuned power efficiency. The Cortex-A12 processor is designed to deliver the best mobile experience:
– Highest performance at lowest power consumption and cost
– Highest efficiency within mid-range thermal envelopes, i.e. achieve highest performance at uncompromised area
– Premium feature set in mid-range mobile
The Cortex-A12 processor is the successor to the Cortex-A9 processor and increases single-thread performance by 40 percent, while matching the best-in-class energy efficiency. Measured in 28nm, the Cortex-A12 processor is about 30 percent smaller in area compared to the Cortex-A9 processor in 40nm technology using the same configuration. Additionally, the Cortex-A12 processor brings today’s premium smartphone features into the mid-range, allowing new use cases and great mobile experiences. Some key added features include:
big.LITTLE™ processing enables the extension of the dynamic range of the Cortex-A12 processor with the addition of the Cortex-A7 processor
Virtualization and TrustZone® security support enabling new use cases like BYOD (bring your own device)
– 1TB addressable memory, providing close to no boundaries on memory space
The Cortex-A12 processor extends the performance capability in mid-range devices without sacrificing energy efficiency when combined with the Cortex-A7 processor as a big.LITTLE CPU subsystem. big.LITTLE processing provides a highly efficient, high-performance processing solution that can scale to many different use cases. The first iterations of big.LITTLE processing featured the Cortex-A15 and Cortex-A7 processors for high-end solutions. Now, the Cortex-A12 processor is bringing big.LITTLE processing to increase the dynamic range of the mid-range by enabling SoC designers to push the Cortex-A12 processor further while using the Cortex-A7 processor to reduce power well below levels of the Cortex-A9 Processor. This results in an ideal combination of compute resource for efficient workload distribution, running lightweight tasks on the Cortex-A7 processor and high-performance tasks on the Cortex-A12 processor. Early results show up to 2x increased efficiency.
Even though it is designed for mid-range smartphone and tablet devices, the Cortex-A12 processor leads with an excellent efficiency profile, making it an ideal fit for other use cases like home networking, residential gateway and auto infotainment systems.
ARM has also designed the Cortex-A12 processor to work efficiently with a complimentary family of high performance, low power ARM CoreLink™ System IP components:
image 
The system diagram shown above illustrates the system IP components that will typically support the Cortex-A12 processor in a mobile SoC. To deliver effortless 1080p30 graphics with 1080p encode/decode the system also features a Mali™-T622 GPU supporting OpenGL/ES 3.0 and a Mali-V500 video accelerator.
The CoreLink CCI-400 cache coherent interconnect provides an IO coherent channel with Mali and opens up a number of exciting possibilities for offload and acceleration of tasks. When combined with a Cortex-A7 processor (not shown) on the ACE port, CCI-400 also allows big.LITTLE operation with full L2 cache coherency between the Cortex-A12 and Cortex-A7 processors. Efficient voltage scaling and power management is enabled with the CoreLink ADB-400 enabling efficient DVFS control of the Cortex-A12 processor.
CoreLink MMU-500 provides a hardware accelerated, common memory view for all SoC components and minimizes software overhead for virtual machines to get on with other system management functions. In this system, the Cortex-A12 processor also enjoys a secure, optimized path to memory to further enhance its market-leading performance with the aid of CoreLink TZC-400 TrustZone address space controller and DMC solution. All interconnect components and the ARM DMC guarantee bandwidth and latency requirements by utilizing in-built dynamic QoS mechanisms.
ARM POP™ IP supports the physical implementation of the Cortex-A12 processor and Mali GPU to enable best power, performance, and area so critical to success in the highly competitive mid-range SoC market. ARM CoreSight™ debug and trace on-chip hardware, coupled with the ARM DS-5™ software development toolchain, enable the debug of random, time-related software bugs, and the non-intrusive analysis of critical areas of software. The ARM Development Studio 5 (DS-5TM) toolchain also makes use of performance counters embedded in the processor, graphics processor and interconnect to enable system-wide optimization.
The ARM Cortex-A12 processor is the highest-performance, mid-range CPU. It is specifically designed for the mid-range mobile market, and is broadly supported by a range of other ARM technology IP including ARM system IP, POP IP and development tools to enable ARM Powered® solutions that contribute to the very best user experience in terms of responsiveness and battery life. At the same time, it allows ARM partners to accelerate time to market for mid-range SoCs, while freeing development time to add their own differentiation. The Cortex-A12 is a highly tuned processor that will bring the performance of high-end mobile devices into mid-range smartphone and tablets, as well as into other great market opportunities we haven’t even considered.
* Source: Bank of America
Related Blogs:

ARM and GLOBALFOUNDRIES to Optimize Next-Generation ARM Mobile Processors for 28nm-SLP Process Technology [press release, June 3, 2013]

New ARM POP technology provides core-hardening acceleration for Cortex-A12 and Cortex-A7 processors
Milpitas, Calif. and Cambridge, UK, June 3, 2013 – In conjunction with the launch of the ARM®  Cortex®-A12 processor, ARM and GLOBALFOUNDRIES today announced new power, performance and cost-optimized POP™ technology offerings for the ARM Cortex-A12 and Cortex-A7 processors for GLOBALFOUNDRIES 28nm-SLP High-K Metal Gate (HKMG) process technology. The Cortex-A12 processor was introduced by ARM today as part of a suite of IP targeting the rapidly growing market for mid-range mobile devices.
The companies will combine ARM’s next-generation mobile processor and POP IP with GLOBALFOUNDRIES 28nm-SLP HKMG process solution, enabling a new level of system performance and power efficiency with the optimum economics necessary to serve the mid-range mobile device market.   The new initiative builds on the existing robust ARM Artisan® physical IP platform and POP IP for the Cortex-A9 processor already available on GLOBALFOUNDRIES 28nm-SLP, signifying another milestone in the multi-year collaboration between ARM and GLOBALFOUNDRIES.
Central to this increase in functionality for mid-range mobile devices is the new ARM Cortex-A12 processor. The Cortex-A12 processor provides a 40 percent performance uplift and direct upgrade path from the incredibly successful Cortex-A9 processor, while matching the energy efficiency of its predecessor. The Cortex-A12 processor provides best-in-class efficiency as a standalone solution, but additionally supports the innovative big.LITTLE™ processing technology with the Cortex-A7 processor, bringing this energy-efficient technology to the mid-range.  GLOBALFOUNDRIES 28nm-SLP process technology and associated ARM POP IP for the Cortex-A12 processor enables up to 70 percent higher performance (measured single-thread performance) and up to 2x better power efficiency in comparison to a Cortex-A9 processor using 40nm process technology. Designers can achieve even higher performance by trading off for lower power efficiency, depending on their application needs. Click here for more information on the Cortex-A12 processor.
The newest POP technology enables customers to accelerate core-hardening of Cortex-A12 and Cortex-A7 processors on GLOBALFOUNDRIES 28nm-SLP HKMG process. POP IP for Cortex processors has successfully enabled ARM-based SoCs with more than 30 different licenses since being introduced over three years ago. POP IP is composed of three elements necessary to achieve an optimized ARM processor implementation: core-specific tuned Artisan physical IP logic libraries and memory instances, comprehensive benchmarking reports, and implementation knowledge that detail the methodology used to achieve the result, to enable the end customer to achieve the same implementation quickly and at low risk.
“With 580 million mid-range smartphones and tablets forecast to be sold in 2015[i], consumers are increasingly looking for the right combination of performance, low power and cost effectiveness,” said Dr. Dipesh Patel, executive vice president and general manager, Physical IP Division at ARM. “With the Cortex-A12 processor and suite of IP announced today, ARM is delivering an optimized system solution leveraging the most innovative technologies available for this market. The POP IP solution on GLOBALFOUNDRIES 28nm-SLP helps designers balance the performance, power and cost tradeoffs to achieve their targets for this growing market.”
GLOBALFOUNDRIES 28nm-SLP technology is ideally suited for the next generation of smart mobile devices, enabling designs with faster processing speeds, smaller feature sizes, lower standby power and longer battery life. The technology is based on GLOBALFOUNDRIES’ “Gate First” approach to High-K Metal Gate (HKMG), which has been in volume production for more than two years. The technology offers a combination of performance, power efficiency and cost that is ideally suited for the mid-range mobile market.
“GLOBALFOUNDRIES is committed to a deep relationship with ARM to enable best-in-class solutions for our mutual customers. Our collaboration on the ARM Cortex-A12 processor implementation is a direct result of this focus and collaboration,” said Mike Noonen, executive vice president of Marketing, Sales, Design and Quality at GLOBALFOUNDRIES.
GLOBALFOUNDRIES’ next-generation 14nm-XM FinFET technology is expected to bring another dimension of enhanced power, performance and area for ARM mobile processors. A Cortex-A9 processor implemented on 14nm-XM technology, using 9-track libraries, is projected to enable a greater than 60 percent increase in frequency at constant power, or a decrease of more than 60 percent in power consumption at constant performance, when compared to implementation on 28nm-SLP technology using 12-track libraries. Similar results are expected for Cortex-A12 processor implementations. Click here for more details on GLOBALFOUNDRIES’ 14nm-XM FinFet technology.
For further discussions about GLOBALFOUNDRIES process technologies or ARM IP offerings please visit the companies’ respective exhibits at the Design Automation Conference (DAC), June 3-5, 2013 in Austin, Texas. ARM is located in booth 931, and GLOBALFOUNDRIES can be found at booth 1314.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: