Hacker News Reader

Comment by gorgoiler | Hacker News Reader

Impressively high level of continuity. The only errors I could really call out are:

1/ 0m23s: The moon polo players begin with the red coat rider putting on a pair of gloves, but they are not wearing gloves in the left-vs-right charge-down.

2/ 1m05s: The dragon flies up the coast with the cliffs on one side, but then the close-up has the direction of flight reversed. Also, the person speaking seemingly has their back to the direction of flight. (And a stripy instead of plain shirt and a harness that wasn’t visible before.)

3/ 1m45s: The ducks aren't taking the right hand corner into the straightaway. They are heading into the wall.

I do wonder what the workflow will be for fixing any more challenging continuity errors.

View on HN

The video was slam cut together to avoid continuity problems. There was a lot of fast camera motion and unconnected scenes.

Particularly bad was the snowmobile sequence. It was literally a different snowmobile in every cut.

The racing pool duck scene was a different pool in every shot.

About the only consistent thing was the faces that were spliced into the scenes.

I do not really see anything super significant in the demo. It looks like this suffers from all the same problems of AI generated video. They just hid it by avoiding more then 5 seconds in the same setting.

The whole pool the ducks are racing at is a completely different pool when Sam starts talking.

The snowmobiles were different in each cut. The shape, color, and style of the lights were different.

The fact that this is their demo to the world and it's full of errors implies that average users will only get worse results.

I’m wary of being that damning, this early. What I want to know is, should my video have these kinds of continuity errors, how easily can I fix them?

It’s ok for this to be a fun toy. (And fun toy while also being an astonishing piece of engineering.) But if it wants to push beyond fun toy then it would be interesting to see how that process works.

Will Sora2 help me sketch out a movie for me, doing 10% of the work where I have to reshoot the other 90% for real, or will it get me 90% there leaving me only 10% left to do “by hand”?

(This is the exact same question, I believe, which is being asked of the maintenance burden imposed by vibe coded products. They get you 90% then fail spectacularly leaving you having to do the bulk of the work again? Or they get you 90% of the way and you int have to fill in the gaps to reach a stable long term product?)

I don't see how this is usable for making like a feature film. Editing will be impossible. At best it will be for ads. At worst for making social media slop.

It is not even just the errors. These video models are really impressive as long as you don't actually have something in your head you want to make. Then the laughable limitations are on full display.

I will believe it when I see because Sora 1 is probably the most disappointing technology given what I thought it was going to be that I can even think of. I waited forever for it and then barely used it because it sucks.

Not sure if it counts as a continuity error, but in the example "Prompt: Martial artist doing a bo-staff kata waist-deep in a koi pond", his wooden staff changes shape several times, resembling a bow at points. That was the first example I noticed as "clearly AI."

The Bo staff in the koi pond also seems to involve some impossible wrist movements

It definitely seems like a state-of-the-art model, but the sound having an underwater effect is the biggest tell. How long can it keep a scene going without things falling apart?

Very first frame of the video: green digital text is messed up. Stopped watching after that :)

Reply on news.ycombinator.com