throwa356262

6 ms·

31.
by throwa356262 16d ago
IMO the key advantages are (1) connectivity and (2) self improvement. I am not the kind of guy who hands his personal email to an AI. But the self improvement part seems to be useful. Maybe something similar can be achieved by pi using an s
32.
by throwa356262 16d ago
What is going on? Why are everyone except OpenAI short on compute? Even Chinese ultra cheap providers are slowly reducing token plans or increasing prices.
33.
by throwa356262 17d ago
Theo is a very insightful guy, but also very opinionated. I think the truth is somewhere in between. Especially as more and more virtualization functions move into hardware, not using them as a second security barrier seems foolish.
34.
by throwa356262 18d ago
Short answer: no Long answer: some dude on youtube calculated that even if electricity was free it would take you 37 years to get back your investment. Slightly longer answer: ... but if you absolutely must have local LLM, a GPU with 24-34
35.
by throwa356262 19d ago
"Chinese official repeatedly told the student to destroy the SIM cards they used to communicate and stick to a Huawei phone." Good to know which phones NSA can't wiretap ;)
36.
by throwa356262 19d ago
BYD is becoming crazy good at execution. But I think they also do rigorous planing years in advance. The Chinese EV movement started actually 10-20 years ago.
37.
by throwa356262 19d ago
Even then it will be a self-dos: the video you uploaded won't be processed.
38.
by throwa356262 19d ago
Because an attacker would not gain anything he not already has. This is basically local self-DOS.
39.
by throwa356262 19d ago
I am sure the fuzzer is interesting. But this bug feels like something an LLM would flag as a major finding but turns out to be completely benign. Update: I tried to look into the fuzzer but it is hard to get past the AI blabb. Can someone
40.
by throwa356262 21d ago
Everyone is free to work on whatever they want, but... I wish the amount of time and brain power spent on this project was instead used to work on something more friendly to the Linux community. For example to improve Strix Halo GPU and NPU
41.
by throwa356262 21d ago
Not sure what you mean, but production silicon will be used in their phones next month.
42.
by throwa356262 21d ago
Yes, and it looked pretty good for an engineering sample. See this video from chineese youtubeer geerkerwan: https://m.youtube.com/watch?v=BCCJuOopLK8
43.
by throwa356262 22d ago
Or ICE. Half of the employees deported and a few shot dead just for good measures should be enough to bring Dario to the negotiations table.
44.
by throwa356262 22d ago
"The union representing those workers said that was news to them." Smells like IRL hallucination...
45.
by throwa356262 22d ago
Why is this a no brainer? There are both cheaper and faster options out there.
46.
by throwa356262 22d ago
Can you expand this a little bit? What makes this book different?
47.
by throwa356262 23d ago
I dont know, warm weather but not too much direct sunlight could actually be a good thing.
48.
by throwa356262 23d ago
O3 is supposed to draw 25% less than O1, which according to reviews of their tablets did not throttle. Their next Fold phone (TBA September) will use O3. I guess we will find out very soon.
49.
by throwa356262 23d ago
All Canada needs to do to become the center of innovation now is to... become 10°C warmer.
50.
by throwa356262 23d ago
Does it test hardware, or hardware plus software that can be different on different platforms?
51.
by throwa356262 23d ago
Linus Torvalds has in past used laptops from Apple, Lenovo and Dell and probably many more. His current laptop is an Intel laptop.
52.
by throwa356262 23d ago
After this news they might suddenly be.
53.
by throwa356262 23d ago
I dont think Geekbench is a good cross-platform benchmark, given what it measures
54.
by throwa356262 24d ago
If the hard cores are configured with incorrect cache coherency options you may not be able to correct the issue in fpga fabric. But this is pure speculation, I haven't checked Cyclone V documentation myself.
55.
by throwa356262 24d ago
This is about the hard macros in Altera Cyclone V FPGA. I suspect the issue is in how Altera configured the IP. They may simply have disabled some part to save on area.
56.
by throwa356262 24d ago
Thanks, now I too want a Lenovo Thinkstation PGX... I think it will be fairly easy to remove refusals from open models. Feels like a lost battle, so why does Alibaba even bother?
57.
by throwa356262 24d ago
That is correct, but if we compare the top 10 literally work of the last 20 years to the same period 500 years ago... Do you think the quality has improved? Do you think many people will read Stephen King in 500 years?
58.
by throwa356262 25d ago
When you think about it, humans have never been reading as much as we do right now thanks to Internet and digitalisation. Yet the quality has gone down when you compare it to, say, 1600's greatest works.
59.
by throwa356262 25d ago
"We were again surprised by the lack of novelty. The models clearly understand the objects they manipulate at a deep level, and yet very few genuinely new ideas emerge..." That was disappointing. I guess 2026 will not
60.
by throwa356262 26d ago
Correct link https://xcancel.com/deepseek_ai/status/2090730032574631962

More ›