It’s Day 0 of Computex 2023 and Arm had an announcement this morning and unveiled the new Total Compute Solutions 2023 (TCS23) which is something like a reference design for all chipset designers like MediaTek to use and customize to their needs before fabricating into a chipset.
There are a few new highlights for today – namely the new 5th generation GPU architecture, new Armv9 Cortex CPU for more performance gains, and the TCS23 also focuses on next-gen AI applications too.
New Immortalis-G720 GPU
Let’s talk about the GPU first – the new Arm Immortalis-G720 GPU is made to provide even better performance for your mobile gaming experience, including ray-tracing (which is still not fully taken advantage of now) and it also has a new feature called Deferred Vertex Shading (DVS) which will change the dataflow, hence further improving performance or lowering power consumption. Arm said that games like Genshin Impact and Fortnite are benefitting from this.
Arm also highlights that this GPU architecture delivers 15% performance and efficiency improvements over the previous generation, and 40% uplift in system-level efficiency.
Armv9 Cortex CPU
For this generation, Arm introduces the 4th-generation Cortex-X core, which is the highest-performing core within the Arm SoCs. It claims to be the fastest CPU that they have ever built (obviously), and brings 15% more performance compared to the Cortex-X3 while being 40% more power-efficient at the same time.
The big and LITTLE cores got an upgrade too – and they are the new Cortex-A720 and Cortex-A520 cores. The former is aimed to provide a higher sustained performance while the latter is Arm’s most performant high-efficiency CPU core ever. For your average Joe, Arm claims that it can provide around 20% of power efficiency over previous generations.
The Cortex-X4 will also be fabricated using the TSMC N3E process – which is the industry’s first.
Let’s talk AI
Everyone knows AI is the new big thing since the boom of ChatGPT a few months ago. But, there is a way to do it better. To have AI processes done more efficiently and more securely, you’ll have to do it locally on your device. That’s why Arm is leading the way to support developers to develop AI and machine learning (ML) apps today.
Arm NN and Arm Compute Library are being used by Google’s apps right now – and they will be opened up to other 3rd-party app developers soon.
What’s next?
Arm is not going to stop. They’ve shown us a sneak peek of the CPU and GPU roadmap and we have something called the Krake GPU and Blackhawk CPU architectures coming for TCS24 – which is next year.
Like our Facebook Page here at NasiLemakTech.com for more news and in-depth reviews! Also, join our Facebook Group for insightful information and memes!
Subscribe to our YouTube channel too!
