I used to visit McDonald’s regularly back in high school, but then the local McDonald’s had “euro deals” which was either the regular hamburger or the cheeseburger (switched every week) for €1. And they were open until like 5 in the morning so it was cheap and available if I was on the way home and a bit hungry after a night out.

But now they already close in the evening, their prices have risen a ton, and the portion sizes seem to have gotten ridiculously small. A “large” menu with one of their biggest burgers costs like €20, and for that I get a mushy, bland industrial burger that I can eat in about 3 bites, and like 10 or so fries. It’s more like a snack than a meal. I rarely eat fast food, because every time I try it I end up still hungry afterwards and disappointed with the quality and cost.

And just down the street from the McDonald’s here, there’s a “frituur” (a typical Belgian takeaway place that sells fries, burgers and fried snacks), and their “small” portion is at least 5x more than McDonald’s “large” portion, more than I can eat, and much better in flavour too. And their burgers are also bigger and better than McDonald’s offering. Best of all, a full meal there is like €12, and after that I’m completely stuffed and still have leftovers to feed to my dogs or chickens.

McDonald’s, and the other big fast food chains like Burger king and KFC are all just overpriced, undersized garbage. I really don’t understand why people go to them often enough to keep them in business while there are so many alternatives available which are both better and cheaper. At least in this area.

Do you think an installation with a CoP of >1 gives you free energy or something? It does not. An installation with a CoP of 2 still uses energy, and twice as much for the same heat demand as an installation with a CoP of 4. A CoP of 1 would be equivalent to a resistive electric heater, which is generally considered to be the worst way to heat a house. Cost wise with the energy prices in my area you need a CoP of between 2.5 and 3 to break even in energy costs, and with the grid here still being largely dependent on coal and gas in the winter you’d need a CoP of around 2 to be positive in CO2 footprint. Most heat pumps can easily reach that CoP, when used sensibly.

Tanks with thermal gradients also do not matter for CoP. What matters is the difference between the hot side and the cold side. If you have to heat something up to a higher temperature, that efficiency is lost, regardless of what happens with the heat afterwards.

The video also shows the exact opposite of what you claim. You can clearly see that the power required to run the heater goes up as the temperature increases. And yes, topping up requires relatively low amount of power compared to a cold start, but that’s because it doesn’t have to add much heat energy to the tank, not because of a difference in efficiency during top-up.

I’m also not saying you need a heat pump for every single tap, I’m just saying that relying on one single heat pump for all your heating and cooling needs in your house is a stupid idea. It of course makes sense to have one heater for all hot water taps in your house. But it does not make sense to run your heating and dryer on the same heat pump, unless you like wasting a lot of electricity.

You also mentioned per-device inefficiencies, but it’s actually the opposite. Heat pumps are at their most efficient when properly sized for the installation. If you have only a single unit then it has to be sized for the peak demands you might need, and it will not run at peak efficiency in the summer when it just has to power your dryer. Having a dedicated heat pump inside the dryer that is properly sized for exactly the heat output the dryer needs is much more efficient. And that is even ignoring inefficiencies behind heat distribution you’d have with a centralised setup.

A lot of startups are using at least partially consumer grade hardware, which will likely get offloaded to the used market when the bubble bursts as they try to recoup some of their losses. This will put some pressure on pricing especially for used parts.

I don’t think it’ll put pricing back at pre-AI levels immediately, but it will likely cause prices to start to drop pretty soon after the bubble bursts, even before manufacturing is able to change focus.

You clearly did not read the full comment chain, nor my full comment.

The different temperatures are not a problem because they’re difficult to achieve or anything. Most devices that need heat would have a pretty wide tolerance. The main problem is efficiency, when that heat is provided by a heat pump.

If you have underfloor heating, that can typically work with water temperatures as low as just 30°C. If you have a heat pump that needs to heat up water to 30°C, with outside temperature being at about 0°C (realistic winter temperature here, and it makes the calculation easy), then a good heat pump typically reaches a CoP of 4 (sometimes even slightly more).
If you instead would use an internal heat grid of 65°C (a pretty typical practical maximum of conventional heat pumps, and a pretty typical temperature for a heat pump powered clothes dryer to run at), then that CoP typically drops to only just about 2.
So, if you let your heat pump run at 65°C and use a heat exchanger to run your heating off of it at 30°C, you use nearly twice as much electricity for every watt of heat you add to your house compared to just running the heat pump at 30°C and using the output directly.

Typically the heating in a home is the application that demands the most amount of heat by far. But it’s also the application that is usually the most tolerant to low temperatures (especially if you have underfloor heating or forced airflow radiators). So you save a huge amount of electricity by having the circulating water temperature as low as possible, but that also makes that heat not very useful for most other devices (a clothes dryer operating at only 30°C would be highly ineffective).

Heat grids work great if they are supplied by industrial waste heat. But if the heat has to come from a heat pump they’re a terrible idea because of how much heat pump efficiency decreases with increasing temperature delta.

I think Toyota’s cruise control handles that situation nicely. It detects that situation (shows it on the display) and smoothly adjusts your distance and speed so you’re at a safe distance for the merger. I have it turned on probably about 90% of the time when driving.

But that the best cruise control I have experienced so far. I’ve used other cars that would be very slow to react and adjust the distance after the merging, and also one that would just automatically slam the brakes whenever someone merged in front of you. Both situations are not something you’d want to happen in heavy traffic, when there might be someone driving close behind you.

Central heating systems use recirculating hot water to spread meat from a central heater (which can be a heat pump) to multiple radiators. In theory there’s nothing preventing you from connecting other devices to that loop of recirculating hot water to use the heat from your central heating system for those too.

It would however add quite a bit of complexity to the devices and plumbing, which would also make both more expensive.

And it also wouldn’t really be worth it. Heat pumps get less efficient when the temperature difference between the hot and cold side gets larger. So while your air conditioning does indeed produce heat, dumping that heat into the outside air at ambient temperature is much more efficient (in terms of watts of heat moved per watt of electricity used) than trying to use it to heat up water to 50° or so to be used as heat source for a dryer. Then there’s also the issue of balancing demand and supply of heat. Your dryer could only use excess heat from your air conditioning if both are running at the same time, so the devices need to be synchronised or you’d need heat storage, which adds additional complexity, expense and inefficiencies.

Devices that need heat also do not all need the same temperature. Your radiators, clothes dryer, dishwasher and so on all have different ideal temperatures, and because heating pump efficiency decreases as the temperature difference between the hot and cold sides increases it’s more efficient to run it with the lowest temperature delta that’s feasible. So it’s more efficient to just give each device that needs either heating or cooling its own heat pump (unless you have sets of identical devices which all need the same temperature and that can all be integrated in a single system, such as central heating radiators or floor heating zones). And the complexity involved in distributing heat between unrelated devices in your house makes giving them all their own heat pump likely the cheaper option too.

When playing trough a game the first time I never care about achievements. I do not disable them, but I just ignore them and don’t check which ones the game has. And if a game does not have achievements then I indeed do not miss them.

But if I have finished a game I ended up really enjoying and want more, then I do check the achievement list for inspiration for what to do. And sometimes I end up discovering parts of the game I initially missed that way.

I have never bothered with RetroAchievements, but I also don’t get the hate about them. They are achievements that are not a part of the game, that you need to deliberately install to be able to get them. So if you don’t like them, just don’t install them? Some people clearly enjoy them, so I don’t think it’s a bad thing that they exist even though I don’t care for them.

People are also way worse at distinguishing them than they think they are. And it’s the same story with images.

If you can use the page code instead then you have the information of which blocks of text belong together, and you know with 100% accuracy what character the text contains. And you also have additional information such as which elements are interactive, and, if semantic tags are used correctly, which part is the header, main content, or such.

Besides, scraping can be done with OCR too. Both softwares can use the same tricks, so if screen readers can work using a workaround, then scraping can work with that same workaround too.

A lot of people parrot that, but it’s not really true, at least on PlayStation. The vast majority of PlayStation games are playable from disc without any internet connection. Of course, you won’t have the latest updates and the game may have more bugs and worse performance if you aren’t connected to the internet, but you can still play games to their completion.

I know that from a legal standpoint you still only buy a licence and don’t legally own the data on a disc, but that doesn’t really matter for the purposes of playing a game. You can play when the servers are down. You can play when your account gets hacked or deactivated. They don’t have a way to take the game away when you aren’t connected to the internet (even if they can legally revoke it).

And even if it would be “just a download key”, it’s still a key that’s transferrable and that you can share or sell. So in terms of ownership it’s still much better than a fully digital license even if you completely ignore the fact that it’s still storage media that apart from a few exceptions still contains a fully functional copy of the game.

For me personally all my games I play on Playstation are on disc, and I bought them all used. For me the discs are the reason to get a console. I live in an area where one ISP has a monopoly so I have to live with a datacap too small to be able to download even a single modern game, and datacap expansions frequently being more expensive than the game I’d download with them. And on the used market the games can be pretty cheap. So all my games on PlayStation are on disc, and I bought most of them used.
If discs are not available for a game, then I pretty much always can get the game way cheaper on PC than on the PlayStation store, and on my PC it’ll run better too, so then I have no reason to buy it on PlayStation instead.
But I do realise I’m a minority in that regard, at least when you look at global scale.

Modifying an existing chip is pretty much impossible. Chips contain structures on multiple layers, but you can only really work on the topmost layer directly. There also usually aren’t really unused gaps, components are placed as tightly packed as possible because if you can make the chip smaller then you can get more chips out of each wafer (because manufacturing costs depends mostly on the amount of wafers and amount of layers on each wafer, not on the amount of chips per wafer).

I guess in theory you could grind the old circuitry off the chip to then build up new circuits on the old chip, but that’d be much more challenging than starting from a new wafer and also not really offer up any benefits.

It’s the renewables on the grid that have to make a difference. If you’re powering an electric car purely out of electricity generated with fossil fuels you’re effectively building a series hybrid with extra steps, with the combustion engine being outside of the vehicle. Or I guess you could also compare it to a diesel-electric locomotive with the generator outside the vehicle, which he also explained in the video would not be good for efficiency. And yes at the scale of power plants you can do some things more efficient, but it’s not actually that much as the efficiency is mostly limited by similar thermodynamic processes than that happen inside an engine. A typical coal power plant also has an efficiency of 30-40% (so effectively a mediocre coal plant is similar in efficiency to a very efficient engine, and yes, some new ones can reach higher efficiencies, but the vast majority still use an old design, at least around here). But when you get the energy from a power plant you still have distribution losses on the grid, conversion losses when charging the battery and again when discharging the battery, and the efficiency losses in the electric motor, while you do not have grid losses and charging/discharging losses and losses in an electric motor with a combination engine is directly driving the wheels.

That’s why it’s important to keep investing in renewables. You indeed don’t need a whole lot of renewables to offset the efficiency losses of the battery and distribution, but you do need at least some of it. And you also should not just look at the grid as a whole, but at how the additional load of charging the vehicle gets handled. If all renewables on the grid are operating at maximum capacity all the time, and there is a coal plant that is used to burn extra coal when extra capacity is needed, then any additional load can be considered to be running on pure coal power even if the grid contains many renewable sources. But if you live in a neighbourhood where there is a constant overproduction of solar power while the grid and can charge your car at times of overproduction, then you effectively charge your car with fully green energy even if the grid contains a lot of fossil fuel plants.

You can consider electric cars to be the infrastructure that enables the transport sector to become more green, but we do also need to actively increase energy generations via green sources to actually make use of that “infrastructure”. Though I guess there is also some good news: Powering an electric car from exclusively fossil fuel energy sources puts the efficiency of the full chain somewhere between regular cars and hybrids. So it should at least never be worse than driving a regular car. Worst case it’s just equivalent. But it does need investments in green energy sources as the demand on electricity increases to become significantly better.

I think the “80% after 1000 cycles” isn’t ambitious enough. 1000 cycles is still very easy to reach, and some phones already barely last a full day when new so even that 20% lost capacity can really degrade the experience. I’d argue that it’s not even an ambitious target at all, as even bog standard lithium ion batteries are frequently rated for that lifespan or more.

I could understand it if they had an exception for potential future technologies that would be pretty much guaranteed to outlast the useful lifespan of the device (such as solid-state batteries which could in theory last for tens of thousands of cycles and thus last for decades, if it is ever able to leave the lab). But as it is now with the unambitious goal of 1000 cycles so many batteries will be exempt by default that I don’t think it’ll change much.

Blue light is visible, so emitting more or less blue light Alters how colours are perceived on the display. That’s also why enabling the blue light filter makes the screen look yellow.
Screens calibrated for the same colour temperature and wilth equal brightness should emit the same amount of blue light regardless of which display technology they use.

The total voltage or amperage of the battery pack does not mean anything for the battery cells. You can put more cells in series and get a higher voltage at lower current, or more in parallel and get a higher current at a lower voltage. But all individual cells will run at the same voltage in either configuration (iirc between 3 and 4V), and the current per cell will also be the same for a given load regardless of the situation.

The main thing a higher battery pack voltage accomplishes is that the cables connected to the battery don’t need to be as thick, as the required thickness of a cable depends only on current, not voltage.

I’ve heard a teacher using that as a test to see which students are using AI: If the student turn in a report full of em-dashes, then the teacher would put them in front of a laptop running Word and asks “can you please show me how you type those long dashes that you used all over your report?”

If they can’t do it, then their report is considered AI-generated or plagiarism (which are considered equivalent by the school). If they could do it they would get the benefit of the doubt, but when I heard it he hadn’t had a single student pass that test yet.

It’s a better and likely far more accurate test than those complete bullshit “AI detectors”.

It’s not easy to scale up chip production, because it relies on extremely precise machines, which take a long time to build, and many different steps on the way from raw materials to a finished chip.

If you’d want to set up a new RAM chip factory with competitive performance, it’d be an investment of over a billion USD at the bare minimum, and it’d take a few years to set up all the processes because the first chips can roll off the assembly line.

If the bubble popped by then, then your new factory would probably run at a loss because it’s nearly impossible to complete with companies who have had decades to optimise their production processes.

Even if the bubble didn’t pop by then, then the next problem will likely be the wafer supply. Because just like how there are only a few companies with the infrastructure to build modern, high-performance computer chips, there are only a few companies with the infrastructure to build silicon wafers of a high enough quality to build those chips with. And they have only just enough capacity to supply their current customers.

So to then solve the wafer problem, someone needs to be willing to invest at least a few hundreds of millions of USD to build a new factory for those, which again would struggle to complete in a post-scarcity market. And wafers are far from being the only resource with that issue.

TL;DR: It’s be a huge investment and a huge gamble, and would likely end up just moving the problem anyway.

Unfortunately not likely going to happen since AI is used more and more to write software, and AI doesn’t tend to write very efficient software.

Alternatieve method tot know thing from future: wait.