KurSix

4 ms·

1.
by KurSix 5d ago
HBM isn't dramatically better in every dimension. You get huge bandwidth and good energy efficiency per bit transferred, but not necessarily a meaningful latency improvement and capacity expansion becomes tied to the package
2.
by KurSix 5d ago
Yeah, the scale is kind of mind-bending. You're taking something that's already extremely fragile, grinding most of its thickness away, handling it through several more process steps, and somehow doing that cheaply enough for mass
3.
by KurSix 5d ago
Sanctions can absolutely create the incentive to build a domestic supply chain, but incentive doesn't automatically translate into catching up on the same timeline
4.
by KurSix 5d ago
China has had a very strong incentive to recreate lithography and fab equipment for years, but a lot of the specialized HBM packaging chain was probably much lower priority until relatively recently
5.
by KurSix 12d ago
You probably can't make this reliable across every modern car without breaking something, but that's different from technically impossible
6.
by KurSix 12d ago
It should be as mundane as pulling the SIM card from a phone (or must be)
7.
by KurSix 12d ago
At that point "opt out" starts to feel like a UI feature rather than an actual description of what the car is doing
8.
by KurSix 19d ago
If AI makes code more stylistically uniform, it might actually make those deeper choices more visible
9.
by KurSix 19d ago
If software becomes dramatically cheaper to produce, that may just give investors even more software to invest in
10.
by KurSix 21d ago
Japan is probably evidence that the surrounding food culture matters at least as much as the existence of UPFs. You can sell plenty of processed food without getting the same outcome if portion sizes, eating habits, school meals, and daily
11.
by KurSix 21d ago
Shelf-stable hummus and Doritos can both be "ultra-processed", but that doesn't tell you much about why one might encourage overconsumption. Hyperpalatability seems much closer to the actual claim the article is trying to mak
12.
by KurSix 29d ago
It's probably less "modern antidepressants are uniquely bad" and more "we got much better at starting people on safer drugs than we did at figuring out how to stop them"
13.
by KurSix 29d ago
For something prescribed to get someone through a temporary crisis, "how and when do we get you back off this safely?" should probably be discussed at the beginning, not years later when the patient decides to stop.
14.
by KurSix 29d ago
What seems most broken is that medicine apparently didn't have a great way to predict which group you’d fall into, or much of a plan for the people who had a really bad time stopping. Also, RIP the anti-spider benefits
15.
by KurSix 1mo ago
The sound especially feels almost impossible to date
16.
by KurSix 1mo ago
I wonder if part of that is that Blade Runner doesn't optimize for the first viewing. Once you already know where the plot is going, your attention is free to wander into the production design, sound, background activity
17.
by KurSix 1mo ago
Wow... The venue almost adds another layer to the movie. You're watching a 1982 vision of 2019, accompanied by live electronic-orchestral music, inside a theater built a couple thousand years ago...
18.
by KurSix 1mo ago
The sound is probably why the visuals feel so much larger than the sets actually were
19.
by KurSix 1mo ago
Soon the GPUs will just be the promotional gift you get for opening a sufficiently large Nvidia financing account
20.
by KurSix 1mo ago
I think they just mean "is expected to guarantee". Since the deal hasn't been signed, Reuters is hedging, but "may guarantee" is definitely awkward wording for something whose entire purpose is certainty...
21.
by KurSix 1mo ago
The permanent jobs number really puts it in perspective. You're not just dropping a giant data center into an existing community at that point, you're potentially reshaping the county around it
22.
by KurSix 1mo ago
The $500B number is hard to even contextualize. Calling it a "data center" almost undersells what is being proposed
23.
by KurSix 1mo ago
The numbers have become so large that normal corporate risk management starts looking quaint
24.
by KurSix 1mo ago
The harder test would be whether they can come up with the right invariants and abstraction in the first place
25.
by KurSix 2mo ago
"The market chose it" doesn't mean it was the best solution, just that it fit the incentives
26.
by KurSix 2mo ago
The problem may not be that LLM taste is bad, but that it is incredibly easy to converge on
27.
by KurSix 2mo ago
Yeah once taste can be fully reduced to rules, it stops being taste and becomes something you can optimize for
28.
by KurSix 2mo ago
Generating the first version is almost free now. Understanding why that version is subtly wrong, debugging it in production, maintaining it six months later and deciding whether the abstraction was a mistake are all still expensive...
29.
by KurSix 2mo ago
So like memey version is still pointing at the real mechanism
30.
by KurSix 2mo ago
A real flatulence event would probably qualify as a minor geological disturbance

More ›