There's two, main attractions for Raptor: 1. All Intel and AMD CPU's are backdoored with management processors running black box software that's already had vu

There's two, main attractions for Raptor: 1. All Intel and AMD CPU's are backdoored with management processors running black box software that's already had vulnerabilities. They refuse to remove the backdoors for consumer segment for probably shady reasons. These POWER CPU's have open-source firmware, OpenBMC, which is a little better on trust side. Personally, I think they're still backdoored somehow since IBM is one of NSA's longest-running partners. Still a risk reduction if we go from anyone might hit this closed software to open software anyone can analyze and improve. https://hackaday.com/2017/12/11/what-you-need-to-know-about-the-intel-management-engine/ https://hackaday.com/2017/12/11/what-you-need-to-know-about-... 2. People that want POWER ISA or just a RISC ISA at x86's performance. The x86 ISA is pretty horrible to some of us. It also locks you into specific patterns of execution, like its stack architecture, that can make it harder to efficiently implement alternative schemes. POWER could be more flexible for alternative designs. 2.1. PowerPC, like sold by NXP/Freescale, is still used in a lot of safety-critical fields along with sections of the embedded sector. While invisible to desktop programmers, it kept getting better and better for niches like with QorIQ CPU's for telecoms. I speculated developing on a POWER architecture for POWER-like targets might be easier somehow (idk though). https://en.wikipedia.org/wiki/QorIQ https://en.wikipedia.org/wiki/QorIQ 2.2. The separation kernels like INTEGRITY-178B were all originally designed for PowerPC boards like Curtis-Wright makes for aerospace and military. They might be easier to port in evaluated configuration to a Raptor than to x86. There are also a lot of hardware/software architectures for improving security that work better when you know what the hardware is doing in the first place and/or can modify the firmware. OpenPOWER has more potential for these designs with the main drawback being CompSci folks usually not able to afford expensive, new computers. Somebody might port an existing idea, though, that they were doing on MIPS, Leon3, ARM, etc. 2.3. The Amiga people also use PowerPC-based machines. They're willing to pay a premium to maintain their nostalgia. Wouldn't surprise me if someone ports MorphOS or something like that to these machines. So, there's a few ideas.