Reading was a hobby most people chose not to engage in that much. If you read books/novels etc for 6 hours per day, people would remark on that like "he reads a lot", often asking you to put down your books to join them in whatever activity.
Few people would have had their own TVs in their room 30 years ago. That wasn't common. They were huge, expensive, and not remotely interesting enough to capture the attention of most people for prolonged periods. It was common to have family rituals where there was about 2-3 hours of watching TV during/after dinner together. That was when they aired a movie after some news.
Even game consoles, if you could afford them, really wouldn't capture your attention that much. Nobody plays Super Mario every day for hours weeks on end. And at least to us that was just another social activity anyways. We didn't play these by ourselves.
But I think all that misses the point. You would be doing pretty much none of these in place of another social activity. They either were a social activity, or they filled in otherwise dead time.
When you're having dinner with your friends or family and everyone is looking at their phone, that is replacing something. I remember getting playing cards and chatting at the dinner table when I was young. Nowadays people just get out their phone or disappear to other personal devices as soon as they are done eating if there's any dinner ritual left at all.
> Few people would have had their own TVs in their room 30 years ago. That wasn't common. They were huge, expensive, and not remotely interesting enough to capture the attention of most people for prolonged periods. It was common to have family rituals where there was about 2-3 hours of watching TV during/after dinner together. That was when they aired a movie after some news.
Depends on where one is from. In my country (U.S.A.), even many lower-middle-class kids tended to have at least a small portable TV (or, more often, the former family TV that had been replaced by a newer one in the living room) in at least their end of the house or apartment, if not their own room, ’way back in the late 1960s to early 1970s. What was common for kids in other countries at that time is, of course, a different matter. As for watching the TV together as a family rather than on separate TV sets: that often depended more on whether the family TV was a newer color model and the kids' room TV was an older black-and-white model --- or, as kids grew older and their viewing preferences changed from their parents’, which shows were on opposite one another. Sometimes it even came down to which room made it easier to watch TV while you were doing homework, talking to a friend who was visiting you from down the street, etc.
Reading used to be super common, including among working class. They used to read what was called "junk literature", basically written equivalents of fun tv.
> The recent MacBook Pros are every bit as fast as my Zen 5 desktop for most tasks like compiling.
Bad example. That's highly parallel, so a higher core-count die is going to destroy the base M5 here.
I don't typically compile Linux on my M5, so I don't really care, but at least online available clang benchmarks put it at roughly half the LOC/s of a 9950X, which released in 2024.
Anything single threaded it should match or even edge ahead though.
It gets for worse for multi threaded perf if you leave behind consumer-grade hardware and compare professional/workhorse level CPUs like EPYC/Threadripper/Xeon to Apple's "pro" lines. That's just a slaughter. They're roughly 3x a 9950X die for these kinds of workloads.
I don't compile Linux or other large C projects on my M5 (why would I). The only thing I have numbers for on both desktop and mobile is your typical JS/TypeScript/webpack shitshow that struggles to keep a high core count CPU remotely busy. Might as well do that on the M5.
There's a large C++ codebase I need to compile, but it can't compile/run on OSX in the first place, hence the desktop that I use remotely for that. Since it's also kind of a shitshow, that one has really terrible compile times: up to 15 minutes on a high powered Intel ThinkPad I no longer use, ~2 minutes on desktop.
I could do it in a VM as well, but let's be real: running it on the M5 in front of me is going to be nowhere near as nice as running it on the water cooled desktop under my desk.
For batch jobs there isn't much competition. 9995wx has 3 to 4x throughput of M5 max.
And then, if your laptop is busy, your machine is occupied - I hate that feeling. I never run heavy software on my laptop. My machine is in the cellar, I connect over ssh. My desktop and my laptop are different machines. I don't want to have to keep my laptop open and running. And I don't want to drag an expensive piece of hardware everywhere.
And then you need to use macOS. I'm not a macOS person.
> For batch jobs there isn't much competition. 9995wx has 3 to 4x throughput of M5 max.
I would hope so, given that you can buy multiple M5 laptops for the price of that CPU alone.
I made a comment about how impressive the M5 laptops were above, so these comments trying to debunk it by comparing to $12,000 CPUs (before building the rest of the system) are kind of an admission that the M5 is rather powerful. If you have to spend 3-4X as much to build something that competes, what are we even talking about any more?
Let's obliterate your phone's battery life running billion parameter models locally at snails pace, saving a couple kilobytes of data.
Sounds useful. I'm sure I'll at least think about maybe making use of that once. Even if I won't, thinking that I could will make me feel like such a savvy consumer anyways.
Invoking both Nazism and Fascism over something like this has to rank among the dumbest rhetorical fumbles of all time, besides being a contemptible way to engage in debate.
Which is sad, because it is followed by some points that are genuinely worth pointing out.
If they set their own place on fire, they're also homeless. Just as self-inflicted, but significantly less dangerous to third parties than driving drunk.
Driving while drunk is not a silly little mistake. A third of all fatal crashes involve drunk drivers. Letting these people drive at all even with a breathalyzer is an abomination. You can expect them to have a similar disregard for other fundamentals of safe driving.
I'm not commenting on the morality of drunk driving. I'm commenting on the effectiveness of just fucking them over, that being, not effective at all.
There's this thing in the mainstream where people feel like the best way to handle people doing bad things is to just pummel them into the ground as much as possible.
While that might feel the most justified, that doesn't actually solve the problem. Suspending licenses doesn't stop drunk people from driving, because cars are more or less a necessity.
So, knowing it's a necessity, we have to design the car around that and enforce safe operation by an alcoholic.
Which is a stop-gap solution. A better solution is making cars not a necessity. But until then, we should do the stop-gap.
Stop-gap is restricting the driving to work schedule then. Everything else is optional and you can learn to work within that system. We try to put up industrial solutions to everything. Why not keep the laws cut and dry. This action = this consequence. You determine your actions you must accept x consequence. Or better yet jail time for 6 months then you will be fed and you will lose everything. There are options
That's how the legal system worked a couple centuries ago. You might be familiar with some of the literature written about it, like Les Misérables. I don't know about you, but returning to the world of 18th century French penal codes sounds pretty dystopian.
Light sources in video games and such. If you have a light source with a very large falloff range illuminating a large area, you'll have noticable steps in the gradient.
Ordered dithering is a very cheap solution to this.
Truly random per-frame noise looks bad and grainy (imho), but various noise functions work well, yes.
Many implementations just sample some noise texture, possibly because that's cheaper - but hardware is so fast nowadays that even sampling some non-trivial noise function many times per pixel hardly registers.
A deferred 2.5D renderer I wrote some while ago just does this screen-wide on the entire framebuffer in a post process step and that pretty much hides all banding already:
You might call this random noise (though it's static). It's enough if you're operating in high precision framebuffers for most rendering steps, and will do a decent job at hiding banding when things are downsampled to whatever the screen supports.
If you can't afford to just rgba16float everything you might have to be smarter about what you do on an individual light level. Probably using some fancier noise/making sure overlapping lights don't amplify the noise.
The time you'd need to compare products also has value. Saving that time by buying a trusted brand is not inefficiency. You'd be poorer if you actually went out of your way comparing thousands of products.
If your model of an ideal market suggests that the realistic and practical approach is inefficient, i.e. your model fails when confronted by reality, your model is horseshit.
Also brand recognition and trust has real value because clearly people are willing to pay for it. Value isn't something intrinsic in an object. Value derives from what people are willing to pay. If people pay more for a rock with an Apple logo on it, then the rock with the Apple logo is more valuable. It's a quality other rocks don't have.
And what does "non-economic reasons" even mean? Should we all only drink tap water because it's cheaper and keeps us alive just as well? Or are we allowed to have some pleasure in life as well?
The time you need to find out competitor pricing (quote) is also inefficiency. Ideally (unachievable of course) all the options and prices (and fair comparison based on product utility only) is immediately available for any customer upon demand, with zero time spent on research.
Irrational behavior is when customers choose a product not for its utility divided by price (but note that pleasure is a type of utility).
Now, exactly how to calculate aforementioned utility is a big pandora box, the whole schools of economists grew up on that question.
Free markets for housing are likely to settle into an "optimum" where some percentage of people cannot afford housing at all, because while construction/rental of housing for them would net a return, it's not worth the opportunity cost. Plus you should not wait for the market to respond to a lack of housing: people will be homeless in the meantime. Supply/demand is mostly reactive, because building for anticipated future demand years down the line is very risky, so most investors don't like that.
Markets don't optimize for "everyone gets some", yet that's precisely what you need for housing. You'll always need the government to come in at some point to provide for those left behind by the free market.
The thing about arguments like this is they're usually used in service of blocking housing. As in we shouldn't do what Austin did because it won't fully solve the problem. We should instead stick with the status quo, which gets much worse than Austin.
Isn’t the initial response to a lack of housing that people consume less housing than they would like, rather than homelessness, eg families with children sharing rooms more than they might like, adults living with roommates, or just people having to live further away from where they would like to be (or moving out of a city altogether)?
I don’t dispute that there are levels of affordability that are bad enough that they start to lead to various forms of homelessness, but it doesn’t seem to me like a fundamental rule that, if some people can’t afford to live alone in a large amount of housing, they also can’t afford to live with roommates sharing a smaller amount of housing, and that the right level of housing prices should also price some people out of those arrangements (ie it demands a pretty high level of inequality if you assume that the market allows typical people to afford to live alone and that sharing can typically reduce per-person rents by half or more)
Individual landlords also dislike those and may not allow it, because the ability of the household to make rent now depends on all adults in that household. 3-4 broke adults who may only loosely know each other and with their own individual plans are a lot less stable than, for example, married couples. It's basically 3x the risk and hassle. Chances are you'd have to evict them after just a couple months.
They'd rather just leave the apartment empty and hope to find a better tenant.
Is that not typically happening only for more egregious situations? E.g. the ban is on more than 3 or 4 unrelated adults living together. There are plausible reasons why that should be regulated, it is not clearly a conspiracy by homeowners to prop up the value of their own homes.
> Free markets for housing are likely to settle into an "optimum" where some percentage of people cannot afford housing at all, because while construction/rental of housing for them would net a return, it's not worth the opportunity cost
You provide no basis for the idea that the returns on housing have to drop below the point at which its financially viable to build before housing becomes affordable. You just say "because" then restate your premise.
> Markets don't optimize for "everyone gets some", yet that's precisely what you need for housing.
Just because markets don't optimize for it, doesn't mean it doesn't achieve it.
> You'll always need the government to come in at some point to provide for those left behind by the free market.
"Not worth the opportunity cost" is not the same as "not financially viable". It means that there are lower risk/higher return investment opportunities than creating that housing. More simply stated "Why would I want to turn my $100 into $110 +/- 20% if I could turn them into $130 +/- 20% instead?"
"I lost a foot" is not the same as "diabetes caused me to need my left leg amputated below the knee", but if you're talking about whether or not you can make it to the store a few blocks away without assistance the distinction isn't really important.
The trick is to acknowledge the market as the main mechanism at play, and have the sense to work around the margins. When the tinkerers get too enthusiastic, they tend to do more harm than good.
Yes, but no. Node itself merely needs a standardized, pluggable layer of indirection in its file APIs. If someone wants to implement a VFS using that, that's cool.
Basically an "fs-core" that everything ultimately goes through, and which can be switched out/layered with another implementation. Think express-style routing but for the filesystem.
That'll keep things simple in node's codebase while handing more power to users.
Few people would have had their own TVs in their room 30 years ago. That wasn't common. They were huge, expensive, and not remotely interesting enough to capture the attention of most people for prolonged periods. It was common to have family rituals where there was about 2-3 hours of watching TV during/after dinner together. That was when they aired a movie after some news.
Even game consoles, if you could afford them, really wouldn't capture your attention that much. Nobody plays Super Mario every day for hours weeks on end. And at least to us that was just another social activity anyways. We didn't play these by ourselves.
But I think all that misses the point. You would be doing pretty much none of these in place of another social activity. They either were a social activity, or they filled in otherwise dead time.
When you're having dinner with your friends or family and everyone is looking at their phone, that is replacing something. I remember getting playing cards and chatting at the dinner table when I was young. Nowadays people just get out their phone or disappear to other personal devices as soon as they are done eating if there's any dinner ritual left at all.
reply