The whole issue is that Entropic is basically a 1 product company and thus the company lives and dies by the life cycle of the product. This is very typical in semiconductor startups (but system startups as well).
The challenge is that the best time to sell the company is before the major revenue cycle when the technology has promise and the opportunities are not run through yet.
The only question is now does Entropic sell itself before it becomes worthless (see Transwitch) or does it sell itself before things get quite that bad (see Broadlight). But the time to sell this company was probably 2006 - 7.