Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
lelag
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
13 ms
·
91.
▲
by
lelag
3y ago
I'm sure Meta will be very happy helping Apple selling more iPhones.
92.
▲
by
lelag
3y ago
The Quest 3 cannot do that, its depth sensor is far too low resolution to scan a face with a good fidelity level. Unfortunately, it also means the quest 3 probably will probably never be very good at dynamic occlusion in mixed reality (ie:
93.
▲
by
lelag
3y ago
The pattern of MS releasing badly received OS every other major release is well known. XP -> hit, Vista -> miss, Windows -> 7 hit, Windows 8 -> miss, Windows 10 -> hit, Windows 11 miss. With this track record, it's not s
94.
▲
by
lelag
3y ago
> Initially designed to last five years. I imagine that it was only sold like this in order to be able to call the mission a success after 5 years. I imagine that the engineers that created the probes, designed them to last as long as po
95.
▲
by
lelag
3y ago
I totally agree with your assessment, Nerf and Gaussian are much more interesting and I'm sure they will be huge in the future. But I don't think it will ever be easy to capture a clean scene randomly, at least not with anything
96.
▲
by
lelag
3y ago
It would be a terrible experience for most users. It would not be like a 360 video where you can direct your gaze in the direction you want. It would be a stereoscopic 110 degrees videos where the viewpoint keeps changing to follow the orig
97.
▲
by
lelag
3y ago
I clearly have "non-educated eyes" as I can't see any meaningful differences personally.
98.
▲
by
lelag
3y ago
This type of 360 photo media experience will soon be replaced by 3d gaussian / Nerf realtime render as demonstrated by the Smerf project which was trending yesterday: https://news.ycombinator.com/item?id=38632492 Add a
99.
▲
by
lelag
3y ago
> For a Golden SAML attack, an adversary must first compromise the AD FS service account on the AD FS server. Then it was game-over from the beginning... Nothing surprising about this and it's not really a weakness of SAML either...
100.
▲
by
lelag
3y ago
And it is getting faster... Consider that it's been 229 days since Mistral AI was founded and they have open models rivalling the OpenAI models.
101.
▲
by
lelag
3y ago
I wish they would have said something about the Antikyrthera mechanism, an analog computer from the 2nd century BC capable of complex astronomical predictions and that was about one and half thousand years early compared to the next time we
102.
▲
by
lelag
3y ago
If the model is closed and served through an API, probably not. But at the moment Mistral seems to focus on smaller and efficient models that you can fine-tuned and run on your own infrastructure. Mistral 7b was free but if they come up wit
103.
▲
by
lelag
3y ago
Weights seem available at https://huggingface.co/togethercomputer/StripedHyena-Nous-7B... .
104.
▲
by
lelag
3y ago
There was also this project that was posted a couple days ago: https://blog.metaphysic.ai/controllable-deepfakes-with-gauss... Why they call a virtual avatar a deepfake beats me though....
105.
▲
by
lelag
3y ago
The paper and effect is really cool but I feel calling it "a deepfake" is very clickbaity. They are demonstrating a tech that allow to animate a photorealistic avatar model of someone. It's similar to what Zuckerberg demonstr
106.
▲
by
lelag
3y ago
> This applies to goods rather than labor to a much higher degree. Permanently owning the labour of 10 people working 16 hours per days is worth a lot more today than 10 slaves back then. Yes, but that's also because of the producti
107.
▲
by
lelag
3y ago
I think it's fair to argue than using modern value for goods that you can produce more efficiently thanks to the availability of cheap energy and phosphate is not really fair. 750 sheeps or 1200 gallons of olive oil were certainly wort
108.
▲
by
lelag
3y ago
It depends on your battery capacity. With a 50-60kwh battery, you will get less than 200 km from 80% to 20% at highway speed, which could render necessary to charge to 100% to get safely to the next charger in a situation where the charging
109.
▲
by
lelag
3y ago
The Turing Test in the past: can it look smart enough to be a human. The Turing Test soonish: can it look dumb enough to be a human.
110.
▲
by
lelag
3y ago
Everyone's grandma has heard about ChatGPT by now. My hairdresser told me she uses it. You can bet that no participant in the original study had heard about computer software capable of simulating a conversation, let alone ELIZA itself
111.
▲
by
lelag
3y ago
It's quite absurd to assert that ELIZA from 1966 outperformed GPT-3.5. Sure, it deceived 27% of the participants at the time, but that was largely because those participants were unaware that such a program could exist. I would bet goo
112.
▲
by
lelag
3y ago
> After a Dutch lab modified the H5N1 bird-flu virus, five years ago, making it super contagious, the chance of a lethal synthetic virus being released in the next twenty years became, well, nonzero. He must have felt pretty vindicated a
113.
▲
by
lelag
3y ago
I run a bug bounty program and I don't mind report for small issues. It's true that most report from "beg bounty" hunters are noise, but we've acted on some reports a few time. One time, in particular, a researcher
114.
▲
by
lelag
3y ago
Interesting... Then choosing the best codecs might depends on the specific use case and the amount of motion in particular. I have a 3080 so I hadn't the chance to test AV1 yet. Is the latency difference significant as reported by VD ?
115.
▲
by
lelag
3y ago
I would add that broader AV1 support is also a good news for low latency application like Cloud gaming or streaming to VR headsets. HW accelerated encoding/decoding improve upon HEVC on every metrics so you can get near real-time strea
116.
▲
by
lelag
3y ago
I thought you were trolling... but no, you are not. For those who want to see more lens shots, you can scroll to "Product Specs" and click the 2nd tab "Navigation".
117.
▲
by
lelag
3y ago
Since a single inference is limited by context length, a multiple agents model is able to process more context at each steps of the reasoning chain, which might improve the overall quality. However, given how easy it is getting to fine tun
118.
▲
by
lelag
3y ago
But then bypassing the filter would just require to change a few bits here and there.
119.
▲
by
lelag
3y ago
I've developed tinnitus in my early 30's for no apparent reasons. I don't think I ever even abused my ears that much. It just happened... Initially, I found it very annoying but you get desensitized over time. Now, I hear it
120.
▲
by
lelag
3y ago
It's not very surprising they are taking this route. It was reported earlier this year that they were having an epiphany after taking apart a Model Y which their engineer described as a work of art... Toyota is all about efficiency and
More ›