Many programs still administer pretty general qualifying exams.
Mine was about two-thirds systems neuroscience (close to my thesis topic), with some reinforcement learning (in the Sutton and Barto sense) and genetics of autism on the side. You were strongly encourage to pick at least one topic that would be "mind-expanding" rather than just thesis prep work. My spouse's department instead had a thesis/anti-thesis structure. They wrote two proposals, the real one that they intended to carry out and a second one on a totally different topic. They then defended both.
Both were written + oral exams. I found the oral part helpful (fun, even) because the conversation keeps the candidate and reviewer from getting hung up on minor points, while probing how they think. Plus, I think there's a bit of tradition in making the candidate sweat by seeing how far you can ratchet up the difficulty.
What I am surprised about is the "practical" component. I've never heard of a hands-on exam!
I'm guessing we just missed each other in grad school?
Yeah, I think we just missed each other since your name sounds familiar.
Those are some pretty major misses for a PhD candidate physicist. He would have been forwarded of this since he was forced to take one of the exam panels courses!
I think part of the reason is that the knowledge tends to be "use it or lose it" on a multi-year timescale. If you work with physicists who have been pidgeonholed in industry for decades then this is probably to be expected.
PhD graduates are extremely diverse in every possible measure. Some are deep generalists. Others are hyper specialized. Some can walk on water. Others leave you wondering how they managed to graduate.