I am looking for some information about how "real" this soft CPU technology is. I'm working with someone who has become enamored with the "soft CPU" concept from the FPGA vendors. I have a number of what seem to me to be "gotta know" questions about this technology, and I don't know how to get them answered. There are big picture questions like:
- What are the compelling reasons to go this route?
- If we take this path, can it be made to *really* work, i.e. never fail from one in a billion type errors?
- How much longer will it take to do it this way compared to the "old" way of using a separate processor and FPGA?
Then I have small picture questions like:
- If you need to add peripherals (like UARTs, PWM contoller, etc., etc.) how well does this work?
- Is the whole development environment reasonable?
I looked around on the web, and there sure is a lot of marketing material, especially from Xilinx and Altera, but that's not what I'm looking for. Do you know anywhere I could get a description of how a real commercial project has gone for somebody, so that I can get some of my questions answered?
Thanks! Steve