China’s Oppo Touts Smartphone Photo Breakthrough With New Chip

By BLOOMBERG

Oppo, one of three leading Chinese smartphone brands filling the void left by Huawei’s retreat, claimed a smartphone mobile photography first with a self-designed chip that delivers sharper and more faithful images.

Oppo unveiled the MariSilicon X Tuesday, which it says can process real-time RAW images — the raw storage format that contains the most information but also requires more computing power to handle. That’s a break from other smartphones that first compress the image captured before applying artificial intelligence to filter and improve picture quality, which saves time but may sacrifice fidelity.

The 6-nanometer part, produced by Taiwan Semiconductor Manufacturing Co., works as much as 20 times faster and halves energy consumption for some tasks relative to Oppo’s previous top-of-the-line smartphone, the company said in a statement. It will debut in the Chinese firm’s flagship Find X line in the first quarter of 2022.

Oppo joins a growing number of Chinese companies exploring advances in semiconductors, at a time Beijing is encouraging the growth of a domestic chip industry. It hopes to differentiate its top-end models from Xiaomi Corp. and Vivo, which are vying for the lead in China now that Huawei Technologies Co. has all but abandoned the arena because of U.S. sanctions. 

Custom silicon has become a key battlefield in particular among smartphone makers, with Vivo launching its own V1 image processor this year after Apple Inc. established the template with its A-series chips.

Video and picture quality are critical to users when selecting a phone. Oppo says its MariSilicon X can produce AI-assisted 4K night video with a high dynamic range, applying image enhancements typically seen on still images to moving pictures.

On Tuesday, Oppo also revealed its first augmented reality glasses, the Oppo Air Glass. The headset consists of a micro projector and a single lens — which it dubbed a “monocle waveguide design” — and takes input via touch, voice or head gestures. It’s also planned for release in the first quarter of next year.