We fabricated a 16-kB cache macro using 0.35-µm quadruple-metal CMOS technology. This is a 285-MHz, two-port 16-kB (512256 b) cache macro that has a 2-ns access time. This high-speed performance is enabled by a hierarchical bit-line architecture that uses double global bit-line pairs (WGBs), and a high-speed timing-insensitive sense amplifier (ISA) that shortens the access time.

Digital neural networks are suitable for WSI implementation because their noise immunity is high, they have a fault tolerant structure, and the use of bus architecture can reduce the number of interconnections between neurons. To investigate the feasibility of WSIs, we integrated either 576 conventional neurons or 288 self-learning neurons on a 5-inch wafer, by using 0.8-µm CMOS technology and three metal layers. We also developed a new electron-beam direct-writing technology which enables easier fabrication of VLSI chips and wafer-level interconnections. We fabricated 288 self-learning neuron WSIs having as many as 230 good neurons.

There are various kinds of analog CMOS circuits in microprocessors. IOs, clock distribution circuits including PLL, memories are the main analog circuits. The circuit techniques to achieve low power dissipation combined with high performance in newest prototype chip in the Super H RISC engines are described. A TLB delay can be decreased by using a CAM with a differential amplifier to generate the match signal. The accelerator circuit also helps to speed up the TLB circuit, enabling single-cycle operation. A fabricated 96- mm 2 test chip with the super H architecture using 0. 35-µm four metal CMOS technology is capable of 167-MHz operation at 300 Dhrystone MIPS with 2. 0-W power dissipation.