Taiwan Semiconductor Manufacturing Company (TSMC) has introduced the foundry industry's first multi-layer mask service (MLM) for 90nm, 80nm and 65nm advanced process technologies. MLM is useful for prototyping single or multi-chip verifications or small volume production runs. It reduces the number of masks used by placing images with the same mask grade from several mask layers onto one reticle. Customers only pay for the actual number of reticles made, plus the data processing of each mask layer. A special in-fab IT system automates the reticle handling and the wafer process flow. The same wafer acceptance criteria used in normal mask sets are applied so wafers achieve identical process and quality results. "Configuring four mask layers onto a single reticle generates the ideal cost-to-performance ratio," Ed Chen, senior director of Strategic and Service Marketing at TSMC, said in a statement. "The savings in mask costs enables our customers to enter advanced technologies with lower NRE investment. We emphasize flexibility so that companies can launch tape-out at anytime and, subsequently, can carry out full package or board level development if needed." As of the fourth quarter of 2007, TSMC has successfully completed and delivered over a score of MLM service runs in 90nm, 80nm and 65nm across different product segments. TSMC also offers multi-project wafer prototyping service like CyberShuttle, as well as normal production tape-out.
Comments