We have one or two of Patterson's books on computer architecture and his name does carry some weight for ARM. But RISC has been challenging x86 for quite some time in desktop/workstation/server and it has repeatedly lost. One could argue that software has a lot to do with that and there is some truth in that but it isn't like x86 processors can't do the job with ease today.
Patterson complained about the inefficiency with x86 in the 1990s and early 2000s. Is he now saying by omission that x86 is very efficient today?