Turn off the RT and the performance would be more than fine. Ultra settings often deliver very little uplift over high as is, but ray tracing is so little of an upgrade for the performance cost that the only reason to run it at all is to justify paying $2000+ for a GPU. I think rat tracing is generally pointless for most games as well. The performance cost is not worth the marginally better visuals and most games don't want hyper realistic lighting because it gets in the way of design and game play. No game can truly be designed for it until the technology creeps down into the mainstream and low-end of the market and we're several hardware generations away from that still.

Ryanair wouldn't care about third parties selling tickets for them if they weren't involved in some kind of fuckery of their own. If they want to subsidize certain routes, etc. to draw in customers that's their own business, but Turing around and getting upset at someone else for trying to take advantage of that subsidy isn't any of their business either. While I don't think they're at all obligated to make it easy or possible for any third party to do that, they could solve the problem completely by not engaging in the sort of behavior that attracts these kind of middlemen in the first place. They can have their cake, but that shouldn't act surprised when someone else tries to eat it too.

The CCP doesn't want anyone else to utilize a system that isn't part of their social credit system. Don't mistake an even bigger control freak flexing its own muscles as a sign of actual benevolence.

Java also has double and Double though (and outside of char and Character all of the other classes have the same name as the primitive with the exception of the capital letter), so I don't think that's it, or at least not entirely. For the most part (there are some exceptions) it doesn't matter since the compiler will implicitly handle conversions between the two. Syntax highlighting or other IDE features are likely more helpful at distinguishing between the two anyway.

I haven’t dug in to Rust enough or all that recently to know if there's a deeper reasoning behind the difference. Unless you can see the declaration it's the same problem as Java where you're probably relying on the IDE to inform you what type you're dealing with if you happened to forget.

I was similarly underwhelmed by the first. It was technically impressive, particularly considering it came out nearly 15 years ago and most computer graphics are lucky to hold up for a decade. However, narratively it was bland and from what I've heard the sequels are worse. It seems like Cameron has aimed the films at a younger audience, which of course is going to limit how complex they can be, but there are plenty of Pixar films that do a better job with their storytelling even though they're animated films designed for a family audience.

The film has already made $130 million globally for the first night release according to Box Office Mojo. It will probably do just fine. Everyone else in Hollywood would kill to be in Cameron's shoes right now. His film will likely end up subsidizing a lot of other crap that lost the studio money this year.

What are you on about? Go on Amazon or any other retailer and tell me there's a lack of competition in the alarm clock market. No one is forced to buy this and alarm clocks are about as far down as a person can get on the need vs. want scale of products before crossing over into the realm of wall-mounted talking bass.

The free market is its own referee. Businesses that don't offer or stop providing value to customers tend not to have them and go out of business. No one is forced to buy this product or service. Using it as some reason to shove idiot government interventionism is as stupid as this alarm clock and subscription model.

The $20 basic alarm clock I've had for at least two decades now continues to work just fine. A lot of people I know don't even have one as they just use their phone to set an alarm to wake up. Anyone spending $170 on an alarm clock has more money than sense to begin with so it's little surprise that a product prices to attract stupid people will nickel and dime them after the purchase as well.

Someone will always sell a basic $20 alarm clock and there's little need for anything beyond that. I refuse to feel bad for anyone who wastes their money on something like this even if they're being taken advantage of.

Doing all of this is completely unnecessary. First design your game so that the server never trusts the client. Don't give it more information than the human player could themselves see and never rely on any calculations from the client. That's still insufficient though, so it's necessary for the server to collect and analyze the data it receives from the client. Anything that frequently operates outside of the thresholds of human ability is cheating. Cheat programs are still programs and operate algorithmically and can be identifiable in that way.

There's also the matter of what to do with the cheaters. You can ban them out right, but that's just information to the people selling the cheats. They can do A/B testing to detect the detection methods. I think that a better solution is to quarantine them so they only ever play other cheaters. Anyone falsely labeled will lose horribly in this environment and will be washed out of it. Everyone else will only be inconveniencing people as awful as they themselves are. None of it requires users to install or run invasive code on their own machines.

It's actually bad at this sort of thing. You can ask an LLM to write a 5 paragraph essay about the Gettysburg Address and to give you the program code to perform a word count and it will give you responses that are experts would agree are correct or otherwise good. However the same AI that delivered each if asked to count the number of words in the five paragraph essay it generated will be wrong. The current AIs posses that kind of advanced technology giving the appearance of magic, but it's just a very clever trick. There's very clearly something missing and even if we can't define what that is, once you understand what to look for it becomes apparent that the tools are limited in much the same way an Eliza bot is despite how it once managed to convince many people that it was human despite being a few hundred lines of code.

You don't test something by walking down the golden path where everything works out perfectly. Instead you consider ways to break it and show that it doesn't work. If you want to convince me something is intelligent it had better damned we'll be able to take two pieces of knowledge it has and make logical connections between them. If you're not testing well or hard enough you're only pulling the wool over your own eyes.

Over what time period? If I told you that if you give me $100 now, I'll give you $300 in 10 years, that's a fairly good deal if you think I have a good chance of delivering on that or enough other income sources to squeeze if that investment goes sideways. It's the same idea here only in the billions of dollars as opposed to the hundreds.

Microsoft investors either put up with this same shit the last several dozen times the company stuck a lot of money into some dubious but potentially profitable proposition or they sold their shares to someone else who would. Some of it pans out and a lot of it doesn't. The tech industry is the sort of place where one winner more than covers dozens of losers.

All of these companies get filtered through a further layer of a market that treats them in the same way that they treat AI. Microsoft, Oracle, and any other company you can buy shares of are just an investment that may or may not pay off. Staring directly into it will drive you mad, but somehow it works better on average than anything else we've come up with. It's not just companies that believe that AI will pay off, but larger humanity as well, at least to a certain degree best measured by willingness to invest in the companies investing in AI.

If you think all of that is utterly wrong and stupid then you should start shorting the stocks of the companies investing in AI. If you're right, you'll make money from doing so. If you're wrong you'll lose money. Whether you care about making money or not, this is a useful exercise for you to reflect upon how much you really believe your own position. As a wise man once said, "Put your money where your mouth is."

There are a few possibilities that stick out to me. One is that lead is more available and is being uptaken to replace something else that isn't available in the necessary quantities. The human body is good at making some substitutions when some molecules aren't available, so it would not surprise me if other organisms did the same. For example, if humans restrict carbohydrates in their diet, the brain will switch to using ketones in place of glucose. Plants could be using lead in place of something else in a similar fashion.

Another is that it's just random noise. It's an interesting observation that should be studied in greater detail but if it wasn't a hypothesis they tested for then it's not something that was properly controlled for. If you've ever done a controlled study before you've no doubt found some interesting results that you weren't even looking for. Sometimes you've inadvertently stumbled onto something more interesting than what you were looking for in the first place and other times it's coincidence and not reproducible.

Until someone takes the time to investigate and study it, no one may know why we got this result.

It's beautiful to see a whole new generation realizing that Microsoft Works is an oxymoron.

They've been in control of government about as much as Republicans over the last 20 years. If they were going to do anything shouldn't those evil companies already have been ripped to shreds? The evil companies contribute to Democrats and Republicans alike, employ their friends and relatives, and take other steps to ensure that neither party will make any serious efforts to root them out. The truly insidious ones even get legislation passed that gives the appearance of reining them in while actually achieving the opposite.

No political party has much incentive to deliver you results, because at that point they have nothing to promise you or anyone else. For whatever reason the promise of action is valued by the electorate as much or slightly more than any actual accomplishment. The best you can do for yourself is to avoid doing business with evil companies. While not everyone will be as personally responsible as you, it's still your personal resort choose who you do business with.

I'm surprised Europe has t figured out a way to regulate the flood of cheap Chinese crap out of existence. While they have tariffs of their own in some cases they usually just find other ways to enact protectionism to keep foreign competitors out of local industries.

The unfortunate truth is that most people value cheap crap in the now more than more expensive quality products that will last much longer. Unless and until you or anyone else can change human nature the cheap crap will flow, whether from China or whichever country eventually replaces them.

AI can spit out a lot of boilerplate type code (some of it close enough to correct) that it can replace the sort of code monkeys that were slapping together bits of code sourced from across the internet all day and hoping it would work before seeing if adding another snippet from elsewhere might solve their problem. The LLMs just do this more efficiently.

A skilled developer could use an AI to generate most of this and fix the mistakes well enough that they're more productive than the aforementioned doorknob licker. Better developers would figure out a way to write an abstraction layer to simplify that process even more, but often times that's a lot of work and requires someone who really know what the hell they're doing and for most companies another half dozen monkeys are just as good in their estimates.

Anyone who's ever had a boss or customer that doesn't really know what they want to begin with no longer cares if an AI craps out some slop that might break in a few edge cases when the requirements will change a week from now and all of that code will get ripped out and replaced with the next sacrificial offering. If what you're doing is really just rapid development / prototyping why not use an AI to speed up the process of eliminating all other alternatives before settling on something that will last for more than a few months?