xiphias2

7 ms·

1.
by xiphias2 4d ago
Sure, Dario needs to decide first if he sees China as an enemy orva friend. Limiting NVIDIA chips already turned China into silicon compete mode, and China is already far ahead in robotics.
2.
by xiphias2 8d ago
Try GPT5 and you will feel the difference. Not one from 2 months ago, but one from a year ago. And then you can get the idea of what happened in just 1 year and what you can expect in 1 year.
3.
by xiphias2 8d ago
I'm not so sure. We humans are from a lower intelligence form (some monkey like ancestor). If those monkeys knew that they are making higher intelligence, they would have collaborated to stop creating humans because they can control th
4.
by xiphias2 10d ago
https://x.com/wtsnz/status/2096815939027349574?s=20 Yes, I see people working on M4 and every few days it gets better ,,However, there's plenty more work to get it fully supported and working with Hyperland.&
5.
by xiphias2 10d ago
What you are talking about is just simply political polarization which is done on purpose by poiticians, which happened whether DHH or Ashahi Linux devs wanted or not. Also there is real human element in both, I'm just going to an Omar
6.
by xiphias2 10d ago
While the Ashahi engineers are working super hard reverse engineering the Apple Silicon, there has been a huge surge in Omarchy circles to get it all working with AI writing all the drivers and get all kinds of hardware in M4/M5 and la
7.
by xiphias2 14d ago
Pretty cool, it's missing how these concepts map to ThunderKittens code (my favorite CUDA toolkit)
8.
by xiphias2 18d ago
Much better at satisfying formal specifications (what they were trained for), but not yet good at long term maintenance. For long term maintenance you need to throw features out / delete code, which requires a deep understanding of the
9.
by xiphias2 18d ago
While it's true, reviews got much harder as generating code got easier, understanding it hasn't.
10.
by xiphias2 19d ago
We'll see, it will depend on the supreme court decision I guess. I'm quite confident the administration will bring it to higher court.
11.
by xiphias2 20d ago
Anthropic may win this but strategically it was just plain stupid. Mythos was blocked for just enough time for OpenAI to just catch up with Sol and get lots of enterprise customers.
12.
by xiphias2 22d ago
I wanted to try it but it works just on M1 and I have M4. I was also considering upgrading to M5/M6, again to take advantage of what I really love: access to the 2nm process. The performance you get with M5/M6 in theory is crazy (
13.
by xiphias2 22d ago
There were 1000 plugins created for Omarchy 4 in 2 days. That's why I don't feel it being hyper minimal anymore. It's still not well integrated of course as those plugins are from different people, but I at least don't f
14.
by xiphias2 22d ago
It's interesting because I just haven't felt the polish. For example when using PyTorch I wanted to try to speed up my NN kernel by 2x by just using half precision and haven't noticed any speedup at all. Also I was missing th
15.
by xiphias2 22d ago
Just Beelink, but it doesn't matter at all. I ordered an ASUS Zephyrus G16 with 5090 NVIDIA card + 1.9kg (quite an overkill, and I know that I will have to limit power output), but hasn't arrived yet. But what's fun is that I
16.
by xiphias2 22d ago
Apple still has the best hardware so I moved to it for the last few years, but the closed software ecosystem is terrible for taking advantage of it. I wasn't able to debug network errors (restartin my Mac worked), Metal was missing low
17.
by xiphias2 26d ago
I still don’t get where this ,,get 8M’’ is from. He put his own 1M into funding Omarchy and 7 people matched it.
18.
by xiphias2 26d ago
Have you read the announcement? He’s funding it not the other way. 7 people just matched it after he announced it.
19.
by xiphias2 26d ago
First put $1M of your own money into a project then talk
20.
by xiphias2 27d ago
It’s really cool that we have this proof that US companies are half year behind Chinese models in architecture.
21.
by xiphias2 27d ago
While reading this ,,paper'' I did some Learned Prompt Augmentation in my head about what I should comment, and realized that there's nothing interesting to write about it.
22.
by xiphias2 1mo ago
The scariest thing for me about companies not caring even minimally about conservation is that when AI gets more powerful than humanity (which is clearly a when not an if, even if there's a lot of uncertainty and differences in opinio
23.
by xiphias2 1mo ago
If RISC-V was good enough for AMD to use it in their controller for their GPUs and it became cheaper than ARM, and NVIDIA is using it in many places, it was better to build upon than getting a change in ARM/x86 licensed and approved by
24.
by xiphias2 1mo ago
,, additionally, moving limits to the end of a first-year course allows students to develop intuitions around the derivative first before seeing the formal proof of their validity’’ Waiting a year to get from intuition to theorems is a perf
25.
by xiphias2 1mo ago
Thanks, it's so boring and disappointing. LLVM has such a cool tutorial, eventhough I haven't written a programming language myself, it was excitung to go through it to see how static programming languages are lowered to SSA form.
26.
by xiphias2 1mo ago
Mojo may be interesting to me, but they should really split it from Max, which is some closed stuff I'm not interested in. They should show some performance comparisions between PyTorch and Mojo, PyTorch+kernel compilation + Triton vs
27.
by xiphias2 1mo ago
Exploitbench in itself isn’t that different from gain of function research in computer code, I can imagine the next being ExploitAndCopyYourselfBench.
28.
by xiphias2 2mo ago
Sure, and spent most of it experimenting with something nobody knew would work in production or not, and needed a lots of breakthrough ideas that are now known as the machines exist.
29.
by xiphias2 2mo ago
Not really, the only reason I (or any other programmer) haven't ever hacked into a system to make my life easier (not to do bad things) is because it's illegal. It was always easier than making a system secure.
30.
by xiphias2 2mo ago
Everybody who has rewritten software understands the current phase: it mostly works, but there are always things to fix to make sure that there are no regressions, and the pressure is huge for any release. I still believe it was the good de

More ›