Design Feature: March 1, 1996

The A236 has an SIMD architecture with four 16-bit vector processors and a 24-bit scalar processor. Each vector processor has a triple-port register stack with 64 registers. Software can access the registers in overlapping groups via register windows. Each vector processor has a 16-function ALU and a 16×16-bit combinatorial multiplier with a 40-bit accumulator. You can use the scalar processor for scalar arithmetic and Boolean operations; for program control; and for computation of data addresses, program addresses, and loop counts. The scalar processor has a triple-port register stack with 32 registers and an ALU. A crossbar switch connects all the processors to the data cache.

A 32-bit instruction unit implements a "super-scalar-vector" instruction set. Most instruction words are 32 bits long, with 64-bit-long extended instructions. As a result of its parallel architecture, the DSP can execute up to five instructions for every instruction word. The instruction cache has a 64-bit output so that the processors can access extended instructions in a single cycle.

The vector processors support the storage of 8- and 16-bit parallel data types to maximize memory utilization. The A236 also supports two 24-bit, software stack pointers, one for the vector processors and one for the scalar processor. To avoid cache conflicts, software can store scalar parameters and subroutine-calling parameters in the scalar stack, and vector parameters are stored in the vector stack.

The A236 has two 16-bit, bidirectional, asynchronous, double-buffered, video-aware, parallel DMA ports to load data for I/O and to pass information among multiple A236 chips. You can connect common video decoder and encoder chips to the A236 without any glue logic or frame buffers. A 32-bit-wide memory port with 64-byte bursts provides a 400-Mbyte/sec interface to SDRAMs. The A236 has a port that connects to a serial EEPROM that contains the chip's BIOS. Upon reset, the A236 reads the EEPROM and transfers the software to the system's SDRAMs. You can program the EEPROM with the A236.

You can use the A236's 16-bit-wide, bidirectional, asynchronous, double-buffered parallel DMA port as a command port for passing commands and data to and from a host processor. The A236's serial debug port supports the RS-232C protocol and is handled by an interrupt-service routine.

Addressing modes—Immediate and register direct.

Special instructions—The instruction set is tailored for programming in C. Oxford includes a set of conditional operations for program-flow control, including true and false conditions. Data movement among the processors is performed via memory, rather than between registers. The A236 also supports quad 8- and 16-bit words. The basic structure is the "quad," such as four pixels. The four vector processors can simultaneously operate on this quad, using a single instruction. An array of quads describes a video frame, which is operated on by a loop that operates on a series of quads.

Support—Oxford Computer provides an assembler, parallel C compiler, a linker, a loader, a simulator, and a debugger, all running under Windows 3.1x, 95, or NT. The company also provides a combined hardware- and software-evaluation kit that contains a video-processing system. This system contains an analog video input and output, a video encoder and decoder, the A236, 4 Mbytes of 100-MHz SDRAM, and an interface to an enhanced parallel port.