Night_Thastus

8 ms·

91.
by Night_Thastus 7mo ago
They're garbage. They're bad enough that If you have an Nvidia GPU, it's borderline impractical to game on Linux. You can, but you'll be cutting framerates in half or more in many cases.
92.
by Night_Thastus 7mo ago
I'll be very interested to see how this plays out with final 3rd-party benchmarks. Now if we can just get some decent Nvidia drivers......
93.
by Night_Thastus 7mo ago
IIRC most motion-sensing home devices are tuned to ignore pets, as best as they are able. They don't get it right 100% of the time though.
94.
by Night_Thastus 7mo ago
I am so frustrated that every company in the world seems to treat user support as a tax that they must must MUST eliminate at all costs. Microsoft just straight up doesn't have phone service anymore - at least for non-enterprise custom
95.
by Night_Thastus 7mo ago
You have it backwards. Suburb infrastructure is expensive and the land pulls in little tax money by comparison. They're almost always a net loss on the city's budget.
96.
by Night_Thastus 7mo ago
The other thing people often forget is that the 'land value' is also a measure of the city's well-being. Those big spikes you see in the center? They cost very little for the city to maintain, and generate oodles of tax mon
97.
by Night_Thastus 7mo ago
These days, I assume every web page is actively hostile until proven otherwise. Use as many layers of protection (uBlock, PiHole, various other extensions, etc.) as possible, filter out content when possible, and bail if those don't wo
98.
by Night_Thastus 7mo ago
Their post gives the impression that clearly AMD's branch prediction is better, because this one number is bigger. "Once more I am disappointed by Intel" While it could very well be true that the AMD branch predictor is strai
99.
by Night_Thastus 7mo ago
CAMM2 has definite advantages over normal sticks. HOWEVER, it all comes down to adoption, adoption, adoption. Until several board makers are making boards with CAMM2 support and manufacturers are making CAMM2 format memory - it'll st
100.
by Night_Thastus 7mo ago
AMD CPUs have been killing it lately, but this benchmark feels quite artificial. It's a tiny, trivial example with 1 branch that behaves in a pseudo-random way (random, but fixed seed). I'm not sure that's a really good examp
101.
by Night_Thastus 7mo ago
I accidentally had it flipped backwards in the original comment, and edited it to fix that. So the person you're responding to was right at the time.
102.
by Night_Thastus 7mo ago
>It is vanishingly rare for users to scroll up when reading content unless they want to reach the top This assumption is the problem. No, it is not rare for users to scroll up while reading. People are not perfect machines that read ever
103.
by Night_Thastus 7mo ago
Something else scroll-related I personally hate: Sticky 'headers' that disappear when you scroll down, and appear when you scroll up. I hate them so much. It hurts my brain to see the stupid thing appear and disappear constantly i
104.
by Night_Thastus 7mo ago
>Long time before our electrical systems to be able to compensate for that kind of whole sale change. Will be at least 20 years if it ever happens. There's little to no reason that the electrical grid itself needs to change for the
105.
by Night_Thastus 8mo ago
The diet of most people in the US is pretty horrific. Absurd amounts of sugar, little to no vegetables, little fiber, lots of heavily processed foods. That certainly does not help the situation. Whether it's correlational or causal I
106.
by Night_Thastus 8mo ago
Most people don't care. What you're talking about is way more work than 99% of people want to put in. They have a TV, it has a youtube button on it. They click, they watch. Anything else is irrelevant.
107.
by Night_Thastus 8mo ago
>0 is too much, in my view. Ads are a toxic weight on everyone who views them. I use ad-blocking as much as possible, along with things like SponsorBlock to skip such content within the video itself.
108.
by Night_Thastus 8mo ago
That is not even close to feasible with today's level of technology, and will not be for quite some time.
109.
by Night_Thastus 8mo ago
That site could use a little more. Maybe a count of how many in the current month and year, tallies for each year, maybe even trends. Could be nice. :)
110.
by Night_Thastus 8mo ago
The average consumer wants a phone with their personal balance of: * Feature * Price * Looks/status Everything else is completely irrelevant. Now, what features and what looks varies over time. But something as intangible as being
111.
by Night_Thastus 8mo ago
Yeah I don't think the common consumer thinks that the leap from a regular Android or Apple smartphone to something 'more open' is the leap from a horse-drawn carriage to an automobile. That's fantasy thinking.
112.
by Night_Thastus 8mo ago
They'll stay as long as it's cheap. The moment any attempt is made to raise the price, the number will crater.
113.
by Night_Thastus 8mo ago
*Was That era is long dead and gone, regardless of the current state of such a light.
114.
by Night_Thastus 8mo ago
It may have features, but you don't need to use them - and at least for me it starts up very quickly and none of those extras get in the way.
115.
by Night_Thastus 8mo ago
I agree. If the LLM were truly an intelligence, it would be able to ask about this nonsense question. It would be able to ask "Why is walking even an option? Can you please explain how you imagine that would work? Do you mean hand-wa
116.
by Night_Thastus 8mo ago
There's something to be said for the social losses associated with it. People who are truly boring don't tend to get noticed for social events, dates, promotions, etc. It can be hard for someone to realize they're not all t
117.
by Night_Thastus 8mo ago
That's the population project, not the fertility projection - which means it will appear smooth. However, I can no longer find the original data I was looking at, so I can't back myself up either.
118.
by Night_Thastus 8mo ago
Long compiles are always going to be more friction. No project stays small and quick to compile forever. Get better hardware, and then one day the project is large enough that you're back with long compile times.
119.
by Night_Thastus 8mo ago
Personally, as someone in C and C++ for the last few years, memory access is almost never the root bug. It's almost always logic errors. Not accounting for all paths, not handling edge cases, not being able to handle certain combinatio
120.
by Night_Thastus 8mo ago
Exactly. People forget that all the output of these LLMs is only as good as the input data. The moment they stop getting fed the latest data, they get stale. If there's no data to scrape about [LANGUAGE] - then the LLM can't help

More ›