Oh I think it amounts to the same thing. Automation can help but it's fundamentally painstaking stuff. As a general example, imagine someone walking down the street, but you don't like the original recording for some reason and decide to replace it. All the footsteps have to be matched up and that's tedious. There are plugins that will match the peak sounds of one recording to another to save you aligning every footstep (or word if you're replacing dialog) but you often end up having to do things by hand. Then if it's one person alone the footsteps may sound too simple so you need to add some fabric swish for their clothes. Or if it's two people, you need to have two different sets of footsteps which is really really annoying. Three people walking is fine because nobody can keep up with that so you can completely fudge the timing and nobody will notice.

The big challenge in movie sound is not to distract from what's going on in the story, which often means throwing away sounds that are present in the real environment but with are excluded by the picture frame, and whose presence thus becomes confusing to the audience because they don't know where the sound is coming from. I hate shooting in restaurants for example, because most restaurants have very loud refrigerators and of course most of them can't be switched off for food safety reasons. When you eat dinner don't notice this because there's music, other diners, maybe the person(s) you went to have dinner with, and the sound of your own body chewing and swallowing - so your brain just filters all the background noise away and lets you focus on the conversation or whatever. But when you watch the same thing on a screen, your brain is like 'what's that big machine noise? why does the lead actress sound like a farm animal? why is it all so echoey?' Total nightmare :-)

I haven't seen Whiplash yet but I'm looking forward to it.