In the late 80s and early 90s, a group of IBM-PC clone
manufacturers, including Compaq and Hewlett-Packard, among others,
defined an extended ISA bus that included a 32-bit data bus with
very fast DMA capabilities. This bus was named EISA (for what else,
Extended ISA). The EISA bus was built into machines for a few years,
but never became a major player. By the mid-90s, PCI had driven EISA
into oblivion.
During that time, I was called by a small company that
specialized in Bus Bridges and Extenders. This company was fully
staffed with competent electrical engineers and technicians, but
they found themselves short-handed. They were working with a large
company that was bidding on a contract to develop and install
communications systems on U.S. Navy ships. One element the prime
contractor needed was a transparent interface between the EISA bus
and the VME bus. This interface had to handle streaming DMA data
transfers, and also provide a transparent programmed I/O interface
from either side. My client already had an interface protocol
implemented in FPGAs, and they already had the VME side of the
product. They needed the EISA side, and they needed it quickly.
Although I had no prior experience with EISA, neither did they,
nor did anyone else, for that matter. I took the job. I immediately
wrote a detailed specification, and we held a spec review with the
prime contractor. We worked out the details and I used my
specification to design the board.
It was no small task. There were no EISA chip sets available so
all of the bus interface logic was implemented with fixed and
programmable logic. I also used about eight Xilinx 3000 series
FPGAs. Over all, I generated approximately 100 schematic sheets,
including the Xilinx designs. In case you are wondering, Xilinx had
just released the 3000 series and had only just begun to talk about
the 4000 series; and VHDL or Verilog was not around for commercial
FPGAs yet either.
My customer took care of the PCB layout and assembly, and I
debugged the first boards in my shop. I met all the deadlines set by
the prime contractor and my client. The project was a success, the
prime contractor won the bid, and my client won the job of supplying
the bus bridges.
This project is a good example of my experience with project
management, product definition, specification creation, logic system
architecture, logic design and debug, and product documentation.
After this project was completed, my client asked me to study the
feasibility of a transparent
EISA Bus Extender product.
