hyperbovine

6 ms·

61.
by hyperbovine 8mo ago
would love to be able to teleport this thread to, oh, 5 years ago. people would think some sort of alien technology had landed.
62.
by hyperbovine 8mo ago
It’s more like, millennials got older and started drinking less (as happens), and Gen Z drinks different things like hard seltzer, and also drinks a bit less overall. Plus there were just way too many craft brewers making hoppy ipa to begin
63.
by hyperbovine 8mo ago
Same exact thing happened in tennis. There was a whole "lost generation" of amazingly talented players who just basically shat the bed whenever they stepped onto the court with Djokovic, Federer, or Nadal. It wasn't until muc
64.
by hyperbovine 8mo ago
What does % usable even mean?
65.
by hyperbovine 8mo ago
By pulling ten million people a year from farms into factories and ploughing 40% of GDP into infrastructure and education. Sounds like a sound analogy to me.
66.
by hyperbovine 8mo ago
I agree, and cloud compute is poised to become even more commoditized in the coming years (gazillion new data centers + AI plateauing + efficiency gains, the writing is on the wall). There’s no way this makes sense for most companies.
67.
by hyperbovine 8mo ago
> Sadly, instead of having better laws we get fallacy rhetoric by people who probably have never touched, much less fired a gun in their lives. Why is this the litmus test for being qualified to write gun legislation? Do we also expect o
68.
by hyperbovine 8mo ago
The ISS consumes roughly 90kW. That’s about *one* modern AI/ML server rack. To do that they need 1000 m^2 of radiator panels (EACTS). So that’s the math: every rack needs another square kilometer of stuff put into orbit. Doesn’t make s
69.
by hyperbovine 8mo ago
Alarm is a good guess. On average I can solve a wordle in 3.6 turns when I start with this guess.
70.
by hyperbovine 8mo ago
Sure there is, as long as your audience does.
71.
by hyperbovine 8mo ago
It’s this. There are many five letter words that are not “wordley”. Words such as, idk, bokeh, are technically part of the lexicon but would never appear as a solution. The wordle bot will even tell you this if you guess them — “good guess,
72.
by hyperbovine 8mo ago
Uhm, plenty of phrases like this existed before 12 months ago.
73.
by hyperbovine 8mo ago
printf() debugging is still considered a best practice in the eyes of many. I still remember being really surprised when I heard my famous (Turing award-winning) CS professor tell the class this for the first time. https://tedspe
74.
by hyperbovine 9mo ago
Hakeem Olajuwon famously started playing basketball in college. He had some … other gifts tho.
75.
by hyperbovine 9mo ago
That’s not irony. Interesting, perhaps, but not ironic. https://thereader.mitpress.mit.edu/what-irony-is-not/
76.
by hyperbovine 9mo ago
On the flip side, if the rumored AI crashdepressionapocalyspe does in fact materialize, those things will become super cheap.
77.
by hyperbovine 9mo ago
The word "stable" literally does not appear in the comment to which I was responding. Maybe I'm just scarred from laboring much too hard in the 90s and aughts to get desktop and laptop Linux working, but here is my current ta
78.
by hyperbovine 9mo ago
I don't understand why this comment is downvoted, it is undeniably true.
79.
by hyperbovine 9mo ago
But hey, he also owns Sees Candy.
80.
by hyperbovine 10mo ago
I still don't understand why certain performance aspects of the CUDA platform are so poorly documented. Why is successfully pushing the hw to its performance envelope considered a novel research result? Shouldn't I be able to look
81.
by hyperbovine 10mo ago
The bar is low.
82.
by hyperbovine 10mo ago
> 1. Gaming cards are their R&D pipeline for data center cards. Lots of innovation came from gaming cards. No way that is true any more. Five years ago, maybe. https://www.reddit.com/r/pcmasterrace/comments&
83.
by hyperbovine 10mo ago
LOL of course they don't want to own Anthropic, else they themselves would be responsible for coming up with the $10s of billions in Monopoly money that Anthropic has committed to pay AMZN for compute in the next few years. Better to t
84.
by hyperbovine 10mo ago
Agree.
85.
by hyperbovine 11mo ago
There's nothing to solve. The CoD kills you no matter what. P=NP or maybe quantum computing is the only hope of making serious progress on large-scale combinatorial optimization.
86.
by hyperbovine 11mo ago
Eh, those applications (incl. protein folding) existed for a decade-plus before LLMs came onto the scene, and there was absolutely nothing like the scale of capex that we're seeing right now. It's like literally 100-1000x larger t
87.
by hyperbovine 11mo ago
> The value of modern AI seems very high. That nobody knows how high, that they still haven't figured out applications, and that the technology and its tools are still far from refined, is normal for any new technology. The same cou
88.
by hyperbovine 11mo ago
jupytext completely solves this problem. Agree about the no hidden state comment from above however -- that's a source of constant footguns for us.
89.
by hyperbovine 11mo ago
Completely agree. I don't know anyone who isn't financially incentivized to see the Jupyter project fail who feels this way about Jupyter. This whole post stinks of "we're losing to Jupyter so let's throw up a ridic
90.
by hyperbovine 11mo ago
Code gen is the absolute best case scenario for LLMs though: highly structured language, loads of training data, the ability to automatically error check the responses, etc. If they could mimic reasoning anywhere it would be on this problem

More ›