Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
chazeon
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
34 ms
·
1.
▲
by
chazeon
2mo ago
As many have discussed earlier, if you have an accent, macOS transcribing is not so accurate; on top of that, Apple's transcribing also don't work well if you have a lot of technical terms (for example, if you are transcribing a t
2.
▲
by
chazeon
2mo ago
the place where zotero and alike become useful is when you start to need a tree-like structure
3.
▲
by
chazeon
2mo ago
Why does this lab test only have marketing materials? They are not even installing linux on a supposedly LLM dev machine
4.
▲
by
chazeon
2mo ago
Well, when 5.5 first came out, it was kind of OK, but now it's almost noticeably worse. It can be done, but requires a lot of effort, just more and more round, and actually, the Gemini Pro on the web (which should be 3.1 Pro) is actual
5.
▲
by
chazeon
2mo ago
Even without stats i know it went bad. In the pass two month barely can do any good scientific writing lately, which of course rely on reasoning. It just writing for gods sake. And it show how far we are from AGI.
6.
▲
by
chazeon
3mo ago
I remember years ago, Ubuntu, Kubuntu, etc., could all be obtained via CD for free; years later, everything become a limited print. https://web.archive.org/web/20090219122023/https://shipit.ub...
7.
▲
by
chazeon
3mo ago
I have always been thinking LUKS was supposed to be enrolled in TPM, so you should not have to enter this key manually; this is just to prevent someone from unplugging the hard drive and reading on another machine. Of course, this depends o
8.
▲
by
chazeon
3mo ago
But if you do this, don't you have to enter two passwords each time you wake? One for LUKS, one for the system login?
9.
▲
by
chazeon
3mo ago
Isn't this the company that make the LLM become a degenerate when it comes to bioscience?
10.
▲
by
chazeon
3mo ago
Mac mini Pro line is doomed, they never made enough of it; skipped M5 Pro, now skipping M6 Pro, it is like 2014-2018 again. Now ordering a custom M4 Pro build take 3 months+ to ship with an increased price.
11.
▲
by
chazeon
3mo ago
GPT-5.5 must have serious issues; it is fast, but quality-wise, it is just not good. It read one LaTeX paper (which is not long) and can spell my name wrong. This is GPT-5.5-high.
12.
▲
by
chazeon
6mo ago
Most CA households may be, but obviously not everywhere
13.
▲
by
chazeon
8mo ago
My understanding of safety concern around Linux is mainly the security, namely, full disk encryption. Full disk encryption avoid an attacker unplug your harddrive and directly read the data. While it is possible, it is definitely not easy t
14.
▲
by
chazeon
9mo ago
Well that’s a very misleading thing. If the US immigration policy wasn’t this hostile to populous countries, more Chinese will want to stay.
15.
▲
by
chazeon
9mo ago
Nvidia's private driver seems to deliver 4k@120Hz just fine.
16.
▲
by
chazeon
10mo ago
Well the seemingly cheap comes with significantly degraded performance, particular for agentic use. Have you tried replacing Claude Code with some locally deployed model, say, on 4090 or 5090? I have. It is not usable.
17.
▲
by
chazeon
10mo ago
Seem images on GitHub web also not showing
18.
▲
by
chazeon
10mo ago
The one thing I wish it has is 3.5mm audio jack. Both Xbox and SONY's dualsense controller have this. But SONY don't support audio via Bluetooth. The Xbox one need a USB adapter but its build is not as good as SONY's. SONY do
19.
▲
SAxense – DualSense Haptics over Bluetooth (POC)
(github.com)
1 points
by
chazeon
10mo ago
|
0 comments
20.
▲
by
chazeon
11mo ago
Gemini is the only model that can provide consistent solution to theoretical physics problems and output it into LaTeX document.
21.
▲
by
chazeon
11mo ago
you can just buy a nanokvm
22.
▲
by
chazeon
11mo ago
Aliexpress has them
23.
▲
by
chazeon
1y ago
Infuse is good, but it does not feel so well-polished for the desktop, for example, some windows for pop-up could have been a real window, but were a pop-up that blocks the main player.
24.
▲
by
chazeon
1y ago
I want to note that: long prompts are good only if the model is optimized for it. I have tried to swap the underlying model for Claude Code. Most local models, even those claimed to work with long context and tool use, don't work well
25.
▲
by
chazeon
1y ago
I do japanese transcription + gemini translations. It’s worse than fansub, but its much much better than nothing. First thing that could struggle is actually the vad, then is special names and places, prompting can help but not always. Fina
26.
▲
by
chazeon
1y ago
Recently, I visited the Pennsylvania Railroad Museum and was fascinated to learn that when steel railcars were first introduced—despite being far safer than their wooden predecessors, which could easily be crushed—many people feared they mi
27.
▲
by
chazeon
1y ago
Boards are low tech and low profit, does American company and workers even want to do it?
28.
▲
by
chazeon
1y ago
Actually recent email innovation I enjoyed is Mimestream, the macOS native client for Gmail. Apple’s smart inbox is half baked but better than nothing. Cloudflare now also has a pretty good email forwarding service.
29.
▲
by
chazeon
1y ago
The thing is, the publisher are not obligated to show all recent post, some may show 1 some may show 5, but they might publish more than what they show in your refresh interval (e.g. 1 week) In this case, using a stateless reader, you will
30.
▲
by
chazeon
1y ago
Yeah and of course it will be depend on your personality and risk model. Compared to other things I don’t want to risk my data, whether leaked or damaged. And I make mistakes, a lot. If you are very meticulous and can ensure that you can pu
More ›