AI Industry

AI Chip Duel: Apple A13 Bionic VS Huawei Kirin 990 5G

“A13 Bionic is the fastest chip ever in a smartphone, that quickly handles the tasks that matter most to people every day,” boasted Apple Senior Vice President of Worldwide Marketing Phil Schiller this morning at the Apple’s annual product conference in Cupertino, California.

Every September Huawei and Apple proudly unveil the new chipsets that will empower their high-end smartphones. Huawei debuted their Kirin 990 last week at IFA in Berlin; and today it was Apple’s turn to introduce the A13 Bionic, the latest iteration of its smartphone chipset. The A13 Bionic chipset will be embedded into Apple’s new iPhone 11, iPhone 11 Pro, and iPhone 11 Pro Max.

“A13 Bionic is the fastest chip ever in a smartphone, that quickly handles the tasks that matter most to people every day,” boasted Apple Senior Vice President of Worldwide Marketing Phil Schiller this morning at the Apple’s annual product conference in Cupertino, California.

Manufactured with TSMC’s second-generation 7nm process, the new A13 Bionic packs 8.5 billion transistors, approximately 23 percent more than A12’s 6.9 billion.

Apple Vice President of Engineering Sri Santhanam said the A13 Bionic’s standout features include its powerful machine learning capability and a low-power design — hundreds of voltage gates enable the chipset to activate only the subsection(s) required for a given process at a given time.

The A13 Bionic adds two machine learning accelerators to six CPU cores (two performance cores and four efficiency cores), enabling the chipset to perform six times faster matrix multiplication — the core operation for machine learning models — than its predecessor.

The A13 CPUs are 20 percent faster than the A12 while using 40 percent less power, and can reach one trillion operations per second. Apple Senior Director of iPhone Marketing Kaiann Drance said the A13’s CPUs significantly outperform the Snapdragon 855 and Kirin 980.

The A13 Bionic uses the same eight-core neural engine design as the A12 but is up to 20 percent faster and uses 15 percent less power.

The A13 Bionic also lays the foundation for the iPhone’s upcoming “Deep Fusion” image processing feature, which uses machine learning to improve low to medium light photography. Enabled by the A13 Bionic’s neural engine, Deep Fusion allows the smartphone to shoot four short images, four secondary images, and one long exposure image when users press the shutter button, assembling a final photo with enhanced details and noise reduction. Deep Fusion will launch this fall.

The chipset race is a tight one. Huawei’s new 7-nm Kirin 990 5G — which was unveiled last week and will empower the company’s upcoming flagship Mate 30 — contains 10.3 billion transistors, eight CPU cores, 16 GPU cores, and three neural processing units built on Huawei’s homegrown Da Vinci architecture. The Kirin 990 5G betters Qualcomm’s Snapdragon 855 by 4.7 times in AI performance, and scored 52403 on the AI Benchmark test, ahead of both the Unisoc Tiger T710 and Snapdragon 855 Plus.

A distinctive edge the Kirin 990 5G holds over the A13 Bionic is its Balong 5000 modem processor, which supports 5G networking and delivers a maximum download speed of 2.3Gbps and a maximum upload speed of 1.25Gbps.

The new A13-powered iPhone 11 lineup will be available for pre-order this Friday and will land on shelves September 20.


Journalist: Tony Peng | Editor: Michael Sarazen

8 comments on “AI Chip Duel: Apple A13 Bionic VS Huawei Kirin 990 5G

  1. Ajay Sharma

    Kirin A970 comparison also required

  2. James Hawdon

    So is it 40% less power or 15…. very horrible explanation indeed.

  3. Pingback: How AI is already reshaping the business world - TechHQ

  4. Pingback: How AI is already reshaping the business world – TechHQ – Ai Times

  5. Anonymous

    Soo…. the Kirin 980 is the clear winner?

  6. SnackonthatCheeseDude

    It’s newer, but no, raw Numbers mean nothing. It’s the practical side that will prove whether it is stronger or not.

Leave a Reply to Ajay Sharma Cancel reply

Your email address will not be published. Required fields are marked *