I think it's bizarre! Despite lo budget indie filmmakers generally having a hundred (or more) times less budget and resources than commercial/professional filmmakers, they are commonly far more wasteful.
Averaging 35 takes per scene is almost unbelievable, let alone 42. For a very long, complex scene comprising a number of different shots this number of takes might be slightly more reasonable, only slightly though. However, the scene in question is a short and simple two shot. For example:
"Take 4 (2:16 p.m.): I know I’m doing the scene badly, but I can’t figure out how to do it well." - Huh? This problem should have been identified and solved during rehearsals, well before take 1, let alone not even identifying the problem until take 4!
"Take 14 (2:44 p.m.): It’s beginning to take shape. The rhythm is kicking in." - Great! However, with proper preparation this point should have been achieved by take 1!
"In the final cut of the film, Noah used Take 29." - Why do another 13 takes then? Sure, a safety take after take 29 but why the other 12?
In other words, the shooting should have effectively been from take 14 to take 29 (plus a safety), although take 18 should have been sorted before filming and takes 19 and 24 are probably non-issues considering how many alt takes there are to play with! And even with 42 takes, they may still have needed some ADR (because the paper rustling continues over the dialogue)!
It's just all so wasteful. I can maybe understand the actors and director needing over an hour to rehearse this scene but why pay for the DP, camera-operator, PSM, any additional personnel and the location for all this rehearsal time? And, it's not just the time/cost of the location crew, the director and editor will have to review 42 takes, then the dialogue editor will probably have to review them all again and there's also a time/cost for all the additional file management, transfer, back-up and storage.
If it's spot on it is spot on.
Indeed, but when is it ever "spot on"? I can't remember the last time I edited the dialogue for a scene in a dramatic production where all the dialogue used in the final mix was from a single production take. And, in 20 years, I'm pretty sure I've never edited or mixed dialogue for a scene which didn't need at least some corrective treatment.
Additionally of course, I'm just talking about the production sound, the chance of all the other crafts (along with the production sound) simultaneously getting it "spot on" is effectively zero.
The only things we can debate is whether you can tell during shooting or find out in the cutting room.
That wouldn't be much of a debate! A PSM for example, has neither the equipment nor the time to perfectly judge if the dialogue is "spot on".
And that it's always good to get an extra take. Just in case the file gets corrupted or something stupid happened that nobody noticed. That's talking about best practises.
I wouldn't call a safety take an extra take, a take after the safety take would be an extra take. Best practises exist because many decades of film making have proven that averaged out, it is the most efficient practise. In other words, sure, you can often get away without a safety take but on average you will be bitten in the ass harder by not doing safety takes than by the time it takes to do them. Best practise is by definition the best way, otherwise it would be called an optional practise. Also, although a safety take does provide a level of protection against file corruption and missing something stupid, these are by far the least common uses of a safety take. The most common use in my experience is for something which was fairly innocuous/subtle or even imperceptible that happened on set.
G