Following its strategic move into the discrete GPU market, Moore Threads is now setting its sights on the APU sector with the introduction of its high-performance System on Chip (SoC) designed specifically for laptops.
Moore Threads Unveils Revolutionary Laptop Chip with Competitive Edge AI Performance
Moore Threads has garnered attention for innovative offerings, establishing a reputation for their unique technology solutions. The Chinese GPU manufacturer previously made headlines by presenting a PCIe 5.0 GPU, although its actual performance lagged behind that of NVIDIA’s GTX 1050 Ti. Now, they have moved into the laptop SoC arena with their proprietary ‘Yangtze’ architecture, which integrates a GPU, software functionalities, and CPU microarchitectures.
Technically speaking, the Yangtze chip boasts a robust 12-core design built on the ‘ARMv8′ architecture, coupled with an integrated GPU that utilizes Moore Threads’ proprietary MUSA architecture. The chip runs at a base frequency of 2.65 GHz and is featured in the MTT AIBOOK, an original equipment manufacturer (OEM) product from Moore Threads. The laptop comes equipped with impressive specifications, including 32 GB of high-speed DDR5-7500 RAM and a 1 TB SSD, complemented by an OLED display that supports a refresh rate of 120 Hz.
The AIBOOK represents a compelling choice for users engaged in mid-to-high-performance tasks. Notably, the in-house chip has seen significant upgrades; earlier iterations featured only an 8-core configuration. Another key attribute is its integrated Neural Processing Unit (NPU), which delivers an impressive performance of 50 TOPS (INT8) in a multi-core setup, specifically designed for advanced speech and image recognition tasks. Below are the main features of the chip:
- 12-core CPU
- Boost Clock of 2.65 GHz
- High-end integrated GPU based on proprietary IP
- 50 TOPS NPU
- Support for H.265, H.264, and AV1 codecs
- Geared towards the AI PC market
Moore Threads’ AIBOOK aims to leverage the burgeoning AI PC trend within China, especially as the demand for local hardware capable of deploying large language models (LLMs) grows. While the chip looks promising on paper, the true test will be its performance in real-world applications, as benchmarks will provide insight beyond theoretical specifications.
The MTT AIBOOK is priced around $1, 440 on China’s JD.com, a significant investment reflecting Moore Threads’ inaugural venture into the APU market.
For more detailed information, visit the news source: Videocardz.
Images and further insights can be found at: Source & Images.
Related Articles:
Microsoft Introduces Xbox Mode for All Windows 11 PCs, Merging Console and Desktop Gaming Experiences
20:22
Samsung Ends LPDDR4 Memory Production Amid 50x Profit Surge, Predicts Increasing Memory Shortages by 2027
19:07
Ultimate Guide to Windrose PC Performance Analysis and Tuning for Optimal PC Experience
18:18
Leave a Reply