JAKARTA - Apple is reportedly developing various technological innovations to celebrate iPhone's 20th anniversary in 2027. One of the key technologies being considered is Mobile High Bandwidth Memory (HBM), according to a recent report from ETNews.
HBM is a type of DRAM memory that is vertically compiled and connected through Through-Silicon Vias (TSVs), a small path that allows very fast signal transmission. This technology is commonly used on AI servers because of its ability to support AI processing together with GPUs, so it is often referred to as an "AI memory".
Mobile HBM is an HBM variant specifically designed for mobile devices, offering high data transfer speed while remaining power efficient and saving physical space. The report states that Apple is considering connecting Mobile HBM with GPU iPhone units to improve AI capabilities locally on devices.
This technology is considered ideal to run a large AI model directly on devices, such as large management models or advanced visual processing, without draining batteries or adding latency.
SEE ALSO:
Apple is reportedly already in talks with major memory manufacturers such as Samsung Electronics and SK hinix, who are both developing their own versions of Mobile HBM.
Samsung is said to use a packaging approach called Vertical Cu-post Stack (VCS)
SK hinix developed the Vertical wire Fan-Out (VFO) method The two companies are targeting mass production of Mobile HBM after 2026.
Even so, this technology faces serious challenges, including production costs that are much higher than the current LPDDR memory, as well as hot constraints on thin devices such as iPhones. The 3D stacking process and TSV also require very sophisticated packaging and production management.
If Apple really adopted Mobile HBM for the 2027 iPhone line, this would be another big step in commemorating the 20th anniversary of the iPhone which is also rumored to come with a complete no-bezel design and a curved screen on all four sides
The English, Chinese, Japanese, Arabic, and French versions are automatically generated by the AI. So there may still be inaccuracies in translating, please always see Indonesian as our main language. (system supported by DigitalSiber.id)
