That kind of thing doesn't bother me. I don't think a dog could even comprehend a wheel. Even within the same species, major advances can look like magic to earlier generations. A top-tier scientist from the 19th century may well have pronounced just about all of today's digital devices impossible science fiction. "Billions of calculations per second in something in your pocket that runs on a battery you can recharge?" Lots of stretches there for a scientist from 1820. There are still people (who are not stupid) who don't believe we went to the moon.
Hard sci-fi just means a story that doesn't instantly fail undergrad physics and whose speculative physics are credible enough that a typical physicist won't roll their eyes.
Wormholes and warp drive, for example, have not been conclusively ruled out. Most physicists doubt you could build them in the foreseeable future (or maybe ever), but nobody's shown a reason they would "break physics."
Faster than light motion or reactionless drives that generate conventional momentum are however believed to be impossible.
It leads to some counterintuitive stuff. Giant massive ships that silently hover in the sky without blowtorching whatever is beneath them are strongly believed to be less plausible than something superficially more far fetched like the Alcubierre Drive. That's because reactionless drives that create conventional momentum "break physics," while so far we have not been able to rule out altering space-time curvature. If I were writing a hard sci-fi story with an Alcubierre Drive in it, I'd still have the astronauts going up to their warp ship in chemical rockets. I'd also have my warp ship using chemical or ion propulsion to move around when not using its warp drive, and the thing would have to have massive heat sinks that would glow red hot when warp is powering up to discharge all the power plant's waste heat (second law of thermodynamics).
Credible speculation in physics usually means that there exist mathematical solutions or models that suggest the possibility of something, but where we do not yet currently know if that math is "real" let alone "engineerable." There may also be things that are technically possible but that are so hard to do that they recede behind a kind of "probability horizon," becoming effectively impossible. An example might be a warp drive that requires so much energy and is so hard to build and control that the probability of someone doing it anywhere in the universe between now and heat death is near zero even if you assume that intelligent life is common.