Intel has reportedly decided to use TSMC’s 4nm EUV foundry node for its upcoming Arc Xe2 discrete GPUs based on the “Battlemage” graphics architecture. This move represents a significant advancement over the Arc “Alchemist” family, which was built on TSMC’s 6nm DUV process. The TSMC N4 node offers higher transistor densities, improved performance, and better power efficiency compared to the N6, allowing Intel to nearly double the Xe cores in its largest “Battlemage” variant. This, along with higher IPC, clock speeds, and other enhancements, is expected to make the “Battlemage” competitive with current AMD RDNA 3 and NVIDIA Ada gaming GPUs. It is interesting to note that the “Battlemage” Xe2 is not being built on the most advanced foundry node; the iGPU powering Intel’s Core Ultra 200V “Lunar Lake” processor is part of its Compute Module, which is being developed on the more advanced TSMC N3 (3nm) node.
This decision by Intel to use TSMC’s 4nm process for its Arc Xe2 discrete GPUs signifies a significant technological leap in terms of performance, efficiency, and competitiveness in the GPU market. By leveraging the capabilities of the TSMC N4 node, Intel aims to deliver a product that can rival the offerings from industry heavyweights such as AMD and NVIDIA. The increased transistor densities, improved performance, and power efficiency of the N4 node will enable Intel to enhance the Xe cores in its “Battlemage” GPUs, resulting in a more powerful and efficient graphics solution for consumers.
Moreover, the utilization of the TSMC N3 node for the iGPU in Intel’s Core Ultra 200V “Lunar Lake” processor further underlines the company’s commitment to innovation and pushing the boundaries of semiconductor technology. By leveraging the more advanced N3 node, Intel aims to deliver cutting-edge performance and efficiency in its integrated graphics solutions, catering to the growing demands of modern computing tasks and applications. This strategic decision to use different foundry nodes for its discrete and integrated graphics solutions showcases Intel’s versatile approach to product development and its focus on delivering high-performance solutions across a range of products and segments.
Overall, Intel’s choice to use TSMC’s 4nm and 3nm nodes for its upcoming Arc Xe2 discrete GPUs and Core Ultra 200V iGPU respectively reflects the company’s dedication to advancing semiconductor technology and delivering competitive products in the GPU market. By leveraging the advanced capabilities of TSMC’s latest foundry nodes, Intel aims to strengthen its position in the GPU market and provide consumers with cutting-edge graphics solutions that rival the offerings from established competitors. This strategic move underscores Intel’s commitment to innovation and technological leadership in the semiconductor industry, setting the stage for exciting developments in the realm of graphics processing technology.
Article Source
https://hardforum.com/threads/intel-arc-xe2-battlemage-discrete-gpus-made-on-tsmc-4-nm-process.2035769/post-1045911491