gronglo

I thought I'd never played it until I read your comment, which brought back a memory. I guess I must have played it at least once.

Opus, in my opinion, is steps away from AGI. 4o doesn't come close.

You could have saved yourself another 2 hours by getting Claude 4 Opus to write out the specs for you first. Why did you need to write them at all?

I guess I figured that, if someone told me they were scared of dying, and I happened to possess the secret to immortality, it would be worthwhile to share it with them.

Can anyone explain what sort of "detector" is used to detect the photons? Is there a standard across all double-slit experiments?

I feel like the type of detector would have a significant impact on the outcome of the experiment, but explanations for layman always leave it out.

I get downvoted every time I praise Claude. But everyone in this thread is getting upvoted for saying the same things. Can someone explain to me the difference?

I think this should be renamed "I know when you're vibe coding with a last generation LLM".

Claude 4 Opus would not make any of the mistakes mentioned, because it knows and understands everything in your codebase. It would know when there's an existing utility function for that particular task, or that you prefer a functional rather than OOP approach. You say you want developers to care about the code they write, well, this is the next best thing; I believe it's as close as we'll see to a machine being able to "care" in my lifetime.

> But now, you're wondering if the answer the AI gave you is correct or something it hallucinated. Every time I find myself putting factual questions to AIs, it doesn't take long for it to give me a wrong answer.

I know you'll probably think I'm being facetious, but have you tried Claude 4 Opus? It really is a game changer.

Mark Zuckerberg himself predicted that AI would replace all mid-level engineers at Meta by the end of this year, with senior engineers continuing to work on tasks such as planning and architecture. With the release of Claude 4 it feels like, if anything, he was too conservative with his prediction — we're already there, and Mark must know it. So what's he playing at?

Why is Meta still interviewing SWEs, is my question. Surely, at this point, an agent can beat any human candidate at any task. What am I missing?

> I couldn't even imagine spending the time and effort into integrating someone else's assets into my game and keeping things balanced

Isn't this something that AI could take care of? I feel like if you gave Claude those same instructions you'd probably be pretty amazed.

> it's an insane amount of work to get an item in game A imported into game B

I understand why that would have been a road block in the last, but my hunch is that this type of problem (mapping visuals and item properties) is something AI would make quick work of.

> but if you want it all to actually work, you'll need to dive into the details eventually.

I don't think that's true. I'm wondering if the author has tried Claude 4 Opus.

Imagine a list of 10 items that gets shuffled so that the order is completely rearranged. With no animation, it would take a lot of additional cognitive load to find where each item went. With animation, your brain would track the general movement of each item much more easily.

Basically, animation provides additional information about object state. Removing that extra information increases cognitive load.

This isn't to say all animations are useful. Many animations are excessive or completely unnecessary, which is probably what has given you a negative view of them.