Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I failed to recreate the 1996 Space Jam website with Claude

j0nah.com

Post reply on HN
Print view View original

31–40 of 483 posts

Re: I failed to recreate the 1996 Space Jam website with Claude

#31

Post by fnordpiglet »

Wowfunhappy wrote: post #3

Claude is not very good at using screenshots. The model may technically be multi-modal, but its strength is clearly in reading text. I'm not surprised it failed here.

Especially since it decomposes the image into a semantic vector space rather than the actual grid of pixels. Once the image is transformed into patch embeddings all sense of pixels is entirely destroyed. The author demonstrates a profound lack of understanding for how multimodal LLMs function that a simple query of one would elucidate immediately.

The right way to handle this is not to build it grids and whatnot, which all get blown away by the embedding encoding but to instruct it to build image processing tools of its own and to mandate their use in constructing the coordinates required and computing the eccentricity of the pattern etc in code and language space. Doing it this way you can even get it to write assertive tests comparing the original layout to the final among various image processing metrics. This would assuredly work better, take far less time, be more stable on iteration, and fits neatly into how a multimodal agentic programming tool actually functions.

Re: I failed to recreate the 1996 Space Jam website with Claude

#32

Post by Lerc »

daemonologist wrote: post #24

Interesting - these models are all trained to do pixel-level(ish) measurement now, for bounding boxes and such. I wonder if you could railroad it into being accurate with the right prompt.

What models are good at this? I have tried passing images to models and asking them for coordinates for specific features, then overlaid dots on those points and passed that image back to the model so it has a perception of how far out it was. It had a tendency to be consistently off by a fixed amount without getting closer.

I don't doubt that it is possible eventually, but I haven't had much luck.

Something that seemed to assist was drawing a multi coloured transparent chequerboard, if the AI knows the position of the grid colours it can pick out some relative information from the grid.

Re: I failed to recreate the 1996 Space Jam website with Claude

#33

Post by GeoAtreides »

999900000999 wrote: post #9

Space Jam website design as an LLM benchmark. This article is a bit negative. Claude gets close , it just can't get the order right which is something OP can manually fix. I prefer GitHub Copilot because it's cheaper and integrates with GitHub directly. I'll have times where it'll get it right, and times when I have to try 3 or 4 times.

>which is something OP can manually fix

what if the LLM gets something wrong that the operator (a junior dev perhaps) doesn't even know it's wrong? that's the main issue: if it fails here, it will fail with other things, in not such obvious ways.

Re: I failed to recreate the 1996 Space Jam website with Claude

#35

Post by fishtoaster »

johncoatesdev wrote: post #30

You last-minute cancelled coffee with your friends to work on this? I'm not sure how I would feel if a friend did that to me.

Based on the later life updates, I suspect this was being humorous.

> After these zoom attempts, I didn't have any new moves left. I was being evicted. The bank repo'd my car. So I wrapped it there.

Re: I failed to recreate the 1996 Space Jam website with Claude

#36

Post by monsieurbanana »

turnsout wrote: post #25
sigseg1v wrote: post #19

Curious if you've tested something such as: - "First, calculate the orbital radius. To do this accurately, measure the average diameter of each planet, p, and the average distance from the center of the image to the outer edge of the planets, x, and calculate the orbital radius r = x - p" - "Next, write a unit test script that we will run that reads the rendered page and confirms that each planet is on the orbital ra…

Yes, this is a key step when working with an agent—if they're able to check their work, they can iterate pretty quickly. If you're in the loop, something is wrong. That said, I love this project. haha

I'm trying to understand why this comment got downvoted. My best guess is that "if you're in the loop, something is wrong" is interpreted as there should be no human involvement at all.

The loop here, imo, refers to the feedback loop. And it's true that ideally there should be no human involvement there. A tight feedback loop is as important for llms as it is for humans. The more automated you make it, the better.

Re: I failed to recreate the 1996 Space Jam website with Claude

#37

Post by GeoAtreides »

ChrisArchitect wrote: post #29
dreadnip wrote: post #15

Why involve an LLM in this? Just download the site?

Yeah, Internet Archive has lots of copies https://web.archive.org/web/20250000000000*/https://www.spac... also What's with the panicked pleas and need to preserve the site, assuming locally...?

The post is clearly about something else than preserving https://www.spacejam.com/1996/

It seems to me the post is about how Claude fails to recreate a very simple website from 1996.

Re: I failed to recreate the 1996 Space Jam website with Claude

#38

Post by Aurornis »

sigseg1v wrote: post #19

Curious if you've tested something such as: - "First, calculate the orbital radius. To do this accurately, measure the average diameter of each planet, p, and the average distance from the center of the image to the outer edge of the planets, x, and calculate the orbital radius r = x - p" - "Next, write a unit test script that we will run that reads the rendered page and confirms that each planet is on the orbital ra…

This is my experience with using LLMs for complex tasks: If you're lucky they'll figure it out from a simple description, but to get most things done the way you expect requires a lot of explicit direction, test creation, iteration, and tokens.

One of the keys to being productive with LLMs is learning how to recognize when it's going to take much more effort to babysit the LLM into getting the right result as opposed to simply doing the work yourself.

Post reply on HN