Hacker Newsnew | past | comments | ask | show | jobs | submit | flimflamm's commentslogin

It would be interesting to see what would be good input for LLMs to do the optimizations / trials on that automatically. Is the visual representation best or something else?

It's actually trivially easy to add a "automated report" feature to buildprof due to its architecture (just a few SQL queries on top of Perfetto's trace processor). And I'm sure AI could hill climb the build time based on this.

But I still think the visual view is invaluable for human understanding and for the "why is this even happening" problems which I think AI is still bad at seeing.


This is just how hacker news is now, huh?

“I did this cool useful thing”

“Great! Please describe how I can get a machine to replace you.”


yh mate sorry for living in a capitalist society

What prevents forgers from generating such magnetic fields? That might be worth while if people start to blindly believing that "pottery frozen magnetic fields are not possible to counterfeit". Note that "not possible to counterfeit" is not a claim in the article.

From TFA:

> “So it’s hard for us to be ahead of the forgers since they can read our papers and figure out ways to trick our methods.”


> What prevents forgers from ...

Cost. Same as them fooling carbon dating, electron microscopes, isotope analysis, and a host of other techniques.

Forgery has always been about how cheaply you could make fakes, vs. the odds and consequences of them getting enough scrutiny to be identified as such.


It’s the banknote thing: at some point a great forgery costs more than the real thing.

I would love to see such counterfeiting. The gear they would need to get and the scientists.

Some wire and DC? and Claude et al

It's one of several techniques they can apply, at least. I for one don't even know why there's a forgery market for ancient pottery, or that there's a market for it.

> I for one don't even know why there's a forgery market for ancient pottery

People have paid top dollar for archeological artifacts for centuries. Humanity lost a lot of knowledge from graverobbers, illegal artifact hunting and unreported thefts of stuff getting exposed after stuff like forest fires.


> Humanity lost a lot of knowledge from graverobbers […]

Don’t forget the major scale destruction wrought by wars (religious and otherwise) - eg in the Middle East.

You might also want to count cultural theft. (Where are the gates of Babylon now??)


> I for one don't even know why there's a forgery market for ancient pottery

Because people desire owning ancient pottery and are therefore willing to pay for it.


Forgery is so ancient that we have forgery ancient items made in ancient times

So we must move to 5 hour working days. Thanks OpenAI!


probably less - I sometimes manage to burn like most of my 5h credits on Claude 5x Max plan in 2h.

The annoying things is that you gave bigger tasks and is such session finished maybe 70% and after 5h pool resets cache will expire so that if you "please continue" it will wipe out again a lot of your 5h pool)


Lets normalize working 5 hours a day. :)


Excellent unless you have a mudflap in your back wheel in countries where it rains.


Thanks for this comparison. Very interesting and surprising!


To whom did you call with your server?


I on the other hand think that I have now found the smoking gun!


This is a genuinely impressive result — and an interesting one as well.


Just be careful not to step on it!


I configure it by copy pasting the pages to chatgpt and asking questions. Dumb? Maybe but seems to work.


If you use codex or Claude code or similar you can just have it ssh in.


We humans have a really good system to do this in practice. We use _imagination_ to simulate how things work and ... then we understand it.


However, there was only 57 however's in the paper it self.


> A new paper provides a major step in that direction, however.

It may have gone unnoticed if used only used once in the article, however.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: