Some responses to comments here: We do not have an FPGA implementation, although we are working on it. The reason getting to product is slow is that we are by

Some responses to comments here: We do not have an FPGA implementation, although we are working on it. The reason getting to product is slow is that we are by choice a bootstrap startup, with three full-timers and a dozen or so part-timers. Compare a CPU project at a major such as Intel or ARM, with hundreds of full-time engineers, big budgets, and five years of development before the product is even announced - all for minor modifications to an established and well understood design. The Mill architecture is notable primarily in what it takes out of the CPU, rather than what it puts in. Patents are a backbreaking amount of work. To economize in filing costs our attorneys have consolidated what we expected to be separate patents into single filings - no less work, just fewer/bigger and hence cheaper. So far the twenty filings that have gone in represent around two thirds of the core technology. And if you think that 80-odd pages of patentese is fun then you are more kinky than should be let out in public. "Architecture astronaut" is a cute term, not restricted to architecture; I have more than enough experience with those full of ideas - for someone else to do. We have done a lot of architecture over the decade, but a fair amount of work too. For a live demo the result of some of our work see http://millcomputing.com/docs/specification/ http://millcomputing.com/docs/specification/ While I have worked in academia, I am more comfortable in an environment making real things for real people (or trying to) rather than sheepskins. The purpose of our talks is to get professional validation of the design in a way we can afford. It is not to sell you something; it will be quite a while before we have something to sell, and you are not our customer. We have a private placement in process and welcome Reg-D qualified investors. If you do not know what that is then you probably aren't. For details: http://millcomputing.com/investor-list http://millcomputing.com/investor-list. We are not cruising Sand Hill Road. We estimate ~$6M to an FPGA, and $25M to a product. Heavy semi is serious industry. In early days I did my first compiler for a Bob Barton-designed machine, the Burroughs B6500. That compiler is still in use forty five years later. I wish Barton were still here today; the Mill owes a great deal to his design philosophy. Ivan