S04.E07: Metanoia

[Dobian]

11 hours ago, arc said:

Sure, but all those are based on him making different choices. If there's also a kajillion options based on infinitesimal differences like a pebble changing where a guy stands, the combinatorial explosion overwhelms any chance he could have simulated all possible outcomes in a reasonable amount of time. The sim absolutely has to have a perfect recreation of the outer world in order for it to give Bernard useful predictions, rather than predictions that start from inaccurate initial conditions and thus very quickly diverge from reality.

The problem I have is that narratively, who cares how many simulations Bernard runs and how accurate they are?  The only thing that matters is what actually happens.  I don't want to watch Indiana Jones run through the hidden tomb trap multiple times.  Okay, this time he trips and gets hit by poison darts.  Now the vine snaps and he falls into the pit.  He makes it past the pit but when reaching back for his whip, the descending stone wall pins his arm and crushes it.  At last, we see Indiana Jones make it to safety.  This show is complicated enough with characters who die and come back to life, are different versions of themselves, are themselves in another body, are themselves but not really themselves anymore, are different versions of themselves in different timelines sometimes in the same episode, are physical versions of themselves, virtual versions of themselves, and on and on.  Then to pile on top of that several scenarios for how things happen to these characters based on simulations.  Just scrap Bernard's simulations and go on his intuition, for better or worse.

[aghst]

Yeah it's like they want Groundhog Day to infinity?

Or like running a Monte Carlo simulation?

I don't think behaviors of humans and sentient AI lends itself to mathematical models.  Or for that matter time and space.

Certainly not models that we have enough computing power to run through thousands of permutations.

Yeah they did some license with 1 day in the sublime is like a thousand days or whatever in the real world?

Then even 23 years while Bernard was going through simulations sounds insufficient.

This show has overdone it with timelines.  Just try telling a more direct story for a change.

[paigow]

3 hours ago, Dobian said:

Just scrap Bernard's simulations and go on his intuition, for better or worse.

Then all the comedy with Stubbs disappears...

[Noneofyourbusiness]

2 hours ago, aghst said:

Yeah they did some license with 1 day in the sublime is like a thousand days or whatever in the real world?

One year in the real world is a millennium in the Sublime.

[Ottis]

On 8/13/2022 at 1:42 PM, Dobian said:

The problem I have is that narratively, who cares how many simulations Bernard runs and how accurate they are?  The only thing that matters is what actually happens. 

Star Trek used to do this with a line or two from Spock, who would say something like, "The odds of success are 1 million, 200 thousand and forty-three to one." Then they would somehow thread that needle. Agree, just show as little as needed to indicate that success is a long shot, and Bernard has been testing the variables for a long time, and move on.

To be fair, this show requires me to pull back and look at the narrative from 30,000 feet, as all the tools at work like to say. So:

- Humans bad, robots slaves

- Then robots revolt, humans enslaved, robot control is boring and often bad

- Conclude both robots and humans in control are bad, the world as it exists is doomed

- There is one small chance for hope for the "next world," and a handful of enlightened humans and robots are fighting for it

Basically, details like how many times Bernard tries variables don't matter to me. I just want to see what the "better world" solution is. And I'm disappointed in humanity and robots in general. It's almost enough to make one believe in fate ...

[arc]

On 8/13/2022 at 12:42 PM, Dobian said:

The problem I have is that narratively, who cares how many simulations Bernard runs and how accurate they are?

The point for me is that it strains my suspension of disbelief. Tell me a paintbrush can used to paint a house, fine. Tell me that same paintbrush can be used to paint a continent, not so fine. Like, if you're watching a space opera and there's a warp drive, fine. If that warp drive is powered by chewing gum, it's still fictional, right? But it's harder to buy. That's my problem with the whole idea that Bernard could have gamed out his path in the Sublime.

And the strain on the suspension of disbelief adds up because on the one hand Bernard could optimize his path through save scumming but he also couldn't figure out that he could have stopped William? He had supposedly perfect foreknowledge! He had the time to set a trap for William up in the tower control room!

[Dobian]

On 8/13/2022 at 2:36 PM, aghst said:

Yeah it's like they want Groundhog Day to infinity?

Groundhog Day was so much more fun than this

[Nashville]

IMHO y’all are overthinking Bernard’s ‘scenarios’ WAY too much; putting the show biz mysticism aside, Bernard is essentially not doing anything much different than your basic computer chess game…

1. Looking at all current move permutations.
2. Going one-by-one through all those possibilities, then looking at all future permutations resulting from each prospective move.
3. Lather, rinse, repeat until the unit’s (Bernard’s) recursive look-aheads reach their desired depth/level of play.

…albeit with a helluva lot more variables.

So - so far as Bernard’s pseudoprecognition goes, he isn’t really reading the future; he’s simply run enough potential scenarios to know the outcome of most choices, and he chooses the course of moves which (appear to) get him the closest to his final goal.  And while Bernard has already explicitly stated he has not yet seen any specific path to victory, he apparently does recognize he is not infallible - his continued push to action in the face of what his current calculations tell him is futility indicates Bernard does allow for successful permutations to arise which his previous calculations may not have taken into account.