On the eve of Computex, Taiwan’s big showpiece event where PC makers roll out the latest and best implementations of Intel CPUs, mobile rival ARM is announcing its own big news with the unveiling of a new generation of ARM CPUs and GPUs. Official today, the ARM Cortex-A75 is the new flagship-tier mobile processor design, with a claimed 22 percent improvement in performance over the incumbent A73. It’s joined by the new Cortex-A55, which has the highest power efficiency of any mid-range CPU ARM’s ever designed, and the Mali-G72 graphics processor, which also comes with a 25 percent improvement in efficiency relative to its predecessor G71.


The efficiency improvements are evolutionary and predictable, but the revolutionary aspects of this new lineup relate to artificial intelligence: this is the first set of processing components designed specifically to tackle the challenges of onboard AI and machine learning. Plus, last year’s updates to improve performance in the power-hugry tasks of augmented and virtual reality are being extended and elaborated.

Before we dive into the detail of this year’s changes, it’s worth recapping what ARM does and why it’s important. This English company, now owned by Japan’s SoftBank, is responsible for designing the processor architecture of practically every mobile device — you’ll have heard of Qualcomm’s Snapdragon, Samsung’s Exynos, and Apple’s A-series of mobile chips, all of which are built using ARM’s instruction sets and based on ARM’s design blueprints. When we talk about the oncoming wave of mobile AI, mobile VR, and smartphones that can perform machine-learning tasks without sending them off to processor farms up in the cloud, developing the capabilities for those tasks starts with ARM. ARM

The new Cortex-A75 and A55 are the first Dynamiq CPUs from ARM. Dynamiq is the branding chosen to describe a much more flexible set of design options for silicon vendors like Qualcomm. Where previously ARM allowed for designs that paired a cluster of so-called big CPUs (from its A7x class) and a matched number of little CPUs (from the A5x series), the new design makes it possible to spec a single, mixed-up cluster composed of both big and little CPUs, to a maximum of eight. Thus, chip makers can now have, for example, seven little A55 cores and just one big A75 one: for a favorable mix of long battery life, cost efficiency, and a high ceiling of single-threaded performance when it’s called for.

"50x improvement in AI performance over the next three to five years"

ARM marketing chief John Ronco says he anticipates a "50x improvement in AI performance over the next three to five years thanks to better architecture, micro-architecture, and software optimizations." ARM’s Dynamiq changes include a redesigned memory subsystem and tweaks to how CPU caches work — which has led to a doubling of memory streaming performance on the A55 relative to the A53 preceding it. Given that the A53 has shipped on 1.7 billion devices over the past three years, it’s truly the A55 that will make the biggest difference in achieving Ronco’s ambitious forecast. In most applications, the new mid-range core will be 10 to 30 percent better than previously, offering up to 15 percent better power efficiency and 18 percent better single-thread performance. But it’s the fact that the new chip designs will be 10 times more configurable, with up to 3,000 different configurations, that will allow chipmakers far greater flexibility to make the most of them by tailoring them to specific tasks. ARM
The Cortex-A75 makes double-figure performance improvements across the board

Interestingly, ARM won’t just be powering machine learning with its new chips, it’ll benefit from ML too. The new designs benefit from an improved branch predictor that uses neural network algorithms to improve data prefetching and overall performance.


The Cortex-A75 makes double-figure performance improvements across the board, with ARM claiming it’s on average 22 percent better than the A73, with 16 percent higher memory throughput, and a 34 percent improvement in its Geekbench score. Single-threaded performance, according to ARM’s Ronco, is up by 20 percent, purely by improving the instructions-per-clock efficiency. The A75 chip is roughly 2.5x the size of the A55, and its intended uses are for infrastructure, automotive, and rich mobile applications. Yes, that means VR, AR, and high-fidelity games, the latter of which ARM’s research has shown have been rapidly increasing in popularity.


A major architectural change with the A75 is the opening of a larger power envelope for chips using this core, scaling up to 2W of power consumption, and thus offering up to 30 percent of extra performance on larger-screen devices. This is entirely targeted at the upcoming Windows on ARM reboot, expected later this year. It’s worth noting that in ARM’s world a "large" screen basically amounts to a laptop — and the company set up a dedicated Large Screen Compute division a year and a half ago to more aggressively target the clamshell devices that Intel has been dominant in. ARM

As to the new Mali GPU, it has 32 shader cores, 25 percent higher energy efficiency, and a 20 percent better performance density (aka performance per mm² of space). The Mali-G72 is at the heart of ARM’s push toward improving machine learning efficiency, and ARM claims it’s showing itself to be 17 percent better than the G71 in ML benchmarks. The design optimizations from the company are tailored to accelerate inference engines rather than training engines — that’s to say ARM chips will be best at using accumulated ML capabilities rather than developing them, which makes perfect sense for mobile applications. Training AI will be a task better left to Nvidia and AMD graphics cards or Google’s custom TensorFlow TPUs.

The Cortex-A75 and A55 designs were released to ARM’s partners at the end of 2016, so by this point, they’ve all had a few months to decide what to do with them. ARM says a "realistic time window" for new mobile devices powered by its latest designs would be the first quarter of 2018 — though the company is also conscious of a new phenomenon it describes as "China speed," where Chinese phone vendors will put its designs into products almost immediately. The Huawei Mate 9, for example, was released just eight months after ARM distributed the Mali-G71 to partners. This faster Chinese cadence could lead to some A75- and A55-based designs this year, but then bulk of them are likely to arrive with the usual smartphone refresh cycle early next year.

ARM has unveiled a set of new processors to provide the brainpower for our mobile devices to cope with advanced artificial intelligence (AI), virtual reality (VR), and mixed reality (MR) technologies.
On Monday, the British semiconductor giant said the new Cortex-A75 and Cortex-A55 processors, alongside the new Mali-G72 graphics processor, have been designed to "address the changing nature of computers driven by AI and other more human-like experiences."
"Distributed intelligence" is at the heart of this trend, which includes connecting AI and the cloud, on-device learning, enhanced security and privacy, and the use of 4K, HDR, and 5G for more "human-like" interfaces.
ARM says that by providing low-power, efficient and powerful processors, device vendors will be able to explore the possibilities of distributed intelligence, and the new Cortex-A architecture enables system-on-a-chip (SoC) architecture designers to scale up to eight cores in a single cluster.
The Cortex-A75 and Cortex-A55 have been designed with this concept in mind. Based on ARM's DynamIQ big.LITTLE technology, the first in line to use DynamIQ, each core is used as the "right processor for the right task," diverting commands and control depending on application and system tasks, thereby improving overall power usage and efficiency.
As each core can have different performance and power characteristics -- and the Cortex-A processors also contain dedicated processor instructions for machine learning and AI tasks -- ARM says it is possible for AI performance to increase by over 50 times during the next three to five years.
The Cortex-A75 will deliver a 50 percent increase in performance in comparison to current devices such as the Cortex-A73. Designed for laptops and large-screen mobile devices, ARM says the ML and VR-ready processor is also suitable for autonomous vehicle use, and in particular when speedy responses are required.
The Cortex-A55, a processor suitable for Internet of Things (IoT) devices and systems which connect IoT to cloud gateways, is a "breakthrough" in power efficiency standards, according to the chip maker. In comparison to the Cortex-A53 found in 28nm devices, ARM says the new processor is capable of delivering up to 2.5 times the power efficiency than older models.

In addition, due to DynamIQ, this chip can also be used in autonomous vehicle applications.
The new Mali-G72 graphics processor has been designed from the ground up for high-fidelity and power-consuming mobile gaming. ARM hopes the Mali-G72 will deliver an advanced mobile VR and gaming experience for users, and having also been optimized for machine learning, the chip reduces write bandwidth requirements which may improve frame rates, streaming, and responsiveness.

According to ARM, one billion Mali GPUs were shipped last year and with the inclusion of a new processor which improves performance by 1.4 times in comparison to 2017 designs, the company hopes adoption will only increase.
In comparison to the Mali-G71, the G72 offers an improvement of 25 percent in power efficiency, a 17 percent ML efficiency gain, and 20 percent better area efficiency, and also includes improved support for mobile multiview systems, multisampling anti-aliasing, foveated rendering and Adaptive Scalable Texture Compression (ASTC), which is used to increase image compression quality in order to save power.
The processors will be shipped in devices from the start of 2018.
First up is the Cortex-A75 CPU core, which the company says can deliver laptop-level performance without burning through any more power than existing mobile processors. ARM is promising a 50 percent boost in performance compared to the older A73 core, which should lend itself well to machine learning processes that run right on your devices. Remember: we're starting to see more smartphones optimize their performance on the fly based on behaviors sussed out by these kinds of algorithms. As these chipsets get more efficient at machine learning, we'll benefit more from the deep insights that get unearthed.
Meanwhile, ARM's A55 CPU is a little less interesting. It certainly seems like a capable update, though — it's said to be 2.5 times as power efficient as the existing A53, a notable gain for a mid-range CPU. The thing to remember is that both of these processors use ARM's relatively new "DynamIQ" foundation, an updated design that allows for them to be used more flexibly.
See, ARM's older big.LITTLE architecture typically paired an equal number of high-powered CPU cores with less powerful ones use for tasks that aren't all that intense. DynamIQ, meanwhile, allows for up to eight completely different cores to be used — you could team up one very powerful core with mixed bag of mid-range and low-power cores, depending on what the cluster is meant for.
And then there's the new Mali-G72 graphics core, an updated take on ARM's work with last year's G71. (You might remember it from devices like Huawei's pretty-damned-good Mate 9.) If you thought the A75 was big news for on-device AI, the G72 may be even more important.
Not only is it more power efficient than the G71; ARM says the GPU is 17 percent more efficient at machine learning processes than the processor it replaces. This is obviously great news for device makers looking to embed more intelligence into their work, but don't forget about the other benefits: the G72 should also handle new, more taxing games and VR experiences very well, too. The only real downside to all of these announcements? We'll have to wait until 2018 before these processor cores start seeing use in devices we can actually buy.

Source: Google News