Saturday, December 21, 2024

Arm Unveils Cortex-X2, Cortex-A710, Cortex-A510 CPUs, New Mali GPUs for Future Smartphones, Laptops

Arm has unveiled its first v9 Cortex CPUs that are designed and deliver next-generation performance. The new CPU range by the Cambridge, UK-based company will take on Intel and includes Arm Cortex-X2 as its flagship offering, which is meant not just for high-end smartphones but also laptops. There is also the Cortex-A710 for a broad range of smartphones, laptops, smart TVs, and smart home devices, as well as the lower-end Cortex-A510. In addition to the new CPUs, Arm has brought in a new Mali GPU family that consists of Mali-G710, Mali-G610, Mali-G510, and Mali-G310.

As announced in late March, the Armv9 CPU architecture is aimed at delivering better performance alongside enhancing machine learning, digital signal processor (DSP), and security experiences for customers.

The first and the top-notch option in the v9 CPU range is the Arm Cortex-X2, successor to the Cortex-X1 that is available on the latest chipsets including Qualcomm Snapdragon 888 and Samsung Exynos 2100. The CPU is designed specifically for premium smartphones and laptops. It is claimed to deliver more than 16 percent performance boost and two times faster machine learning processing over its predecessor. The single-thread performance of the new CPU is also claimed to be 40 percent higher than 2020 mainstream laptop CPUs. In terms of smartphones, the Cortex-X2 is claimed to be capable of delivering a 30 percent single-threaded performance improvement over the premium Android flagships of today’s market.

Arm is using its DynamIQ Shared Unit-110 (DSU-110) to scale up the performance of the Cortex-X2 CPU. The configurable cluster approach enables up to eight Cortex-X2 cores in a single cluster — along with a larger L3 cache support of up to 16MB. This will help reduce data latency.

Manufacturers will, of course, be able to adjust CPU configurations as per their requirements and market demand.

Unlike the Cortex-X2, which is meant for top-end devices, Arm also has the Cortex-A710 for a range of smartphones, laptops, and smart TVs, among other hardware. It is rated to deliver a 10 percent performance uplift at the same power envelope of the previous generation Arm Cortex-A78 CPU, alongside 30 percent enhanced energy efficiency and a two-fold increase in machine learning processing. The Cortex-A710 is the first Armv9 “big” core that is meant for chipsets based on big.LITTLE architecture.

For lower-end devices and the “LITTLE” core in the next-generation SoCs, Arm has brought the Cortex-A510. It is claimed to deliver a 35 percent performance increase over the previous generation Cortex-A55. The new CPU is also rated to be 20 percent more energy efficient and three times faster in machine learning processing when compared with its predecessor.

The Cortex-A510 has a three-wide in-order design that brings better energy as well as area efficiency. Manufacturers can even group two Cortex-A510 CPUs into a complex, with multiple complexes per CPU cluster, the company said.

Similar to the Cortex-X2, the Cortex-A710 and Cortex-A510 both have DSU-110 as the CPU cluster. It brings “sustained use cases” such as AAA gaming to the new Cortex-A CPUs. The three CPUs in the Armv9 range also have Secure-EL2 that brings a standard secure isolation mechanism for trusted services. The company has also provided its Memory Tagging Extension (MTE) to detect and prevent memory safety vulnerabilities. The MTE adoption is already in the works for Android.

“We are also addressing Control Flow Integrity with two new built-in features – Pointer Authentication (PAC) and Branch Target Identifiers (BTI),” the company said in a blog post. “These two hardware mechanisms enable a strong prevention of Return Orientated Programming (ROP) and Jump Orientated Programming (JOP) attacks. Based on our studies from enabling these two features, the number of gadgets available to an attacker in Glibc reduces by about 98 percent, with a code size increase of only around 2 percent.”

The new Armv9 CPUs also include Crypto instructions in NEON and SVE2 space and speculation barriers with micro-architectural built-in that are claimed to help mitigate side-channel attacks.

Alongside the new CPUs, Arm has also introduced the Mali-G710 GPU — its highest-performing GPU that delivers up to 20 percent improvement in performance and power efficiency, as well as a 35 percent machine learning uplift over the Mali-G78. The GPU replaces the existing job manager with a command stream frontend (CSF) that, according to Arm, helps meet the requirements of modern APIs, such as Vulkan, and future mobile gaming content trends. There is also a configurable number of cores, starting from seven cores and scaling up to 16 cores. The count is indeed less than the 24-core scalability available on the Mali-G78. However, Arm claims that the cores on the new GPU are bigger and better in performance and energy efficiency.

Arm has also launched the Mali-G610 as its sub-premium GPU with the new CSF similar to the Mali-G710. It is, however, aimed at sub-premium smartphones that are at a lower price point over the phones featuring the Mali-G710.

Additionally, the new portfolio has the Mali-G510 GPU that is claimed to be 22 percent more energy efficient over the Mali-G57 and enable a 100 percent machine learning boost. It is targeted at mid-range smartphones, premium DTVs, set-top boxes (STBs), and Chromebooks.

The Mali GPU range also has the Mali-G310, the lower-end option in stable that comes with six times better texturing performance, 4.5 times better Vulkan performance, and two times improvement in Android UI content rendering over the Mali-G31.

Arm has not provided a timeline on when we would see the new CPUs and GPUs in devices in the market. However, considering the fact that chipmakers would take some time to deploy the new technologies, we are likely to see some action in 2023. The arrival of the new Arm offerings will also do away with 32-bit chips and make 64-bit architecture more prominent.


Affiliate links may be automatically generated – see our ethics statement for details.

Source link

Related Articles

LEAVE A REPLY

Please enter your comment!
Please enter your name here

- Advertisement -spot_img

Latest Articles