Tessera is what patent trolls see themselves as being. It really does invent critical technology to license. Only a few patents (and hardly any trade secrets) make good clubs to compel payment, and the ones Tessera had been accustomed to use have expired (there remains a lot of money owed from before they expired). The fight over a new generation of microelectronics technology is just taking shape. Tessera clearly has some valuable technology (which they're using in the xFD challenger to pseudo-standard DDR4), but infringement of much of it can only be proved by industrial espionage. This will take a while to settle out.
The optical situation is different (hence the divisionalization). It was clear early-on that most of the key IP in MEMS manufacture was unprotectable by threat of infringement lawsuits. Tessera chose to protect it by keeping it in-house. They developed medium-volume capabilities organically and also bought a high-volume manufacturer whole. There ARE people on-board with experience from top to bottom of the business they are getting into.
It's the center that's a mess. When "life was easy," Tessera developed an open, aggressive corporate culture (reminiscent of 3M, and probably intentionally so). Now, with the Invensas part scrambling and the optical part negotiating with very publicity-shy customers, the move is to a closed financial-like (or troll-like) culture. This is annoying almost everyone. It MAY be necessary in the short run, but longer-term it harms customer acquisition, talent acquisition and even corporate financing.
From what I can tell, this is a patent troll with a lot of expiring patents. They bought up some companies in optics and mems thinking that the IP would be worth something. It looks like there are no major licensees so they are trying to enter the manufacturing business. Does anyone on the executive team or board have any substantial experience with HVM? Proceed with caution.