I remember there is a bug with the hang glider that if you move down and up you gain momentum
Back in the day far cry 1 was ultra realistic graphics.
The game was fun to play, the most annoying thing is if you throw a rock or grenade where they can see if your stealth meter in instantly goes to zero and they start shooting you.
The game is riddled with small bugs, like you can knock the phone off the rock in the first level bunker then the cut scene will have him picking it up.
Having the trigens fight the people throughout the game was fun, and stealth aspect as well.
It's funny how the game emphasis seemed to be on the realistic mode, but it's like it was only tested on medium - the trigens would die very quickly and easily on realistic if the guards shot them, if I had to guess probably the bullet damage is turned up on that mode.
It's an interesting and legitimate question. But just as it takes business time to pick up on new technologies and adapt, it might take institutions as well. I've seen reddit posts of people getting writing flagged as AI written when it wasn't. The question is, how long until they start embracing it as a tool to use to write with?
Many of the institutions have billions of dollars put back. Harvard for example google tells me 53 billion. So I don't think they are just going to go away, they will adapt somehow.
It has a complex history but a lot of it started with the Prohibition historically, when alcohol sales ended some restaurants started allowing tips and it became more commonplace.
It always seemed to me the basics of a messenger app would be the easy part to be accomplished. It's the marketing that would be hard seeing as there are already existing solutions such as Slack and Discord.
If you're doing inference on neural networks, each weight has to be read at least once per token. This means you're going to read at least the size of the entire model, per token, at least once during inference.
If your model is 60GB, and you're reading it from the hard drive, then your bare minimum time of inference per token will be limited by your hard drive read throughput. Macbooks have ~4GB/s sequential read speed. Which means your inference time per token will be strictly more than 15 seconds.
If your model is in RAM, then (according to Apple's advertising) your memory speed is 400GB/s, which is 100x your hard drive speed, and just the memory throughput will not be as much of a bottleneck here.
There will be LLM specific chips coming to market soon which will be specialized to the task.
Tesla already has already been creating AI chips for their FSD features in their vehicles. Over the next years, everyone will be racing to be the first to put out LLM specific chips, with AI specific hardware devices following.
What exactly is the ideal sort of hardware to be able to run and train large models? Do you basically just need a high end version of basically everything?
It can make mistakes and hallucinate. I've found GPT-4 to be slightly better than earlier versions.
Ideally tell it one step at a time you need done, I've found that it struggles with multiple overlapping steps for some things but not others. It can make tic-tac-toe in one prompt in HTML5 and Javascript. But maybe if you ask it to make it rain in the background or something like that in the same prompt it might mess everything up.
I've also found it's good at common problems and sometimes spotting the cause for syntax errors, but less common problems it might not know.
It's such an obvious answer perhaps is why nobody has commented it. But depending on the use, you might try web speech API synthesis. For example a Windows user might see a Cortana option whereas a Mac user might see Siri.
Counter opinion: You're going to be one of many applicant because everyone else is embracing WFH trend too, so you are competing with everyone, especially with some much experience to show, or a portfolio of trending technologies.
It might be that a dev that has a money cushion and is willing and able to relocate can compete on a level others can't because competition would only be localized.
OP was unemployed a year. Time adds up, that's a lot of money.
> OP was unemployed a year. Time adds up, that's a lot of money.
Luckily my government gives me a full year of unemployment benefits at 2/3rds of my salary besides my savings. Obviously, I don't wish to test my luck regardless, but not everywhere is as cutthroat as the US.
More accurately: your government takes your money and then, maybe, gives it back to you in the form of unemployment benefits and the like, minus significant processing fees.
The alternative is to use the money the government doesn't take and save it for your future "unemployment benefit".
The government would give me more than I would be able to save myself in a year. Over multiple years, I would be able to save more efficiently than the government, true. That's assuming everything goes perfectly however, so I view it as a form of insurance: I make less in order to have less anxiety that I will not be able to provide for myself and my dependents if the overlords decide I am no longer worthy of my job. It's a tradeoff, but one that I will gladly make. As a social good, that money also goes into helping those that were not as lucky as me to be able to make good easy money. And perhaps one day I might be like them too, who knows?
Get diet, sleep, and exercise right. If any of these three are off, energy is lower. I wish I could oversell this point a little more, there has been plenty of times over the years I've done something small on days - missed an hours sleep, ate too much, sit around to long without moving -- and my energy level and performance was affected without me realizing it.
I've found the easiest and usually cheapest way to control diet is to cook at home.
Look into the Mediterranean diet, it has consistently come up when researching health. Of course the mere word diet implies temporary, but the right way of eating should perhaps be permanent.
Get a blood test done and have it analysed to determine if anything is off.
Don't get dehydrated and get kidney stones. Drink enough water throughout the day. Sometimes I carry a stainless steel container with me throughout the day so I'm getting enough. Be careful about drinking something else for example two glasses of tea and getting sweaty.
Perhaps find and add some health books to your reading list. One interesting, seemingly irrefutable argument put forth in "The China Study" though the book controversial, was no single ingredient food will give more health as much as focusing on the overall diet for most people. Two good books are "How Not to Die: Discover the Foods Scientifically Proven to Prevent and Reverse Disease" and "On Food and Cooking: The Science and Lore of the Kitchen" are two good ones.
I'm just not sure if it would make a difference even if it was outlawed.
The algorithms are out there, and it will only be a matter of time until more organizations adopt them to make their own programs. The tech savvy will be able to get and use them eventually. The internet is worldwide and it would be out there somewhere where people could get it.
It will cause problems and disruptions of some kind in various industries. Chat to generated target email will be a problem, with the need for verification and a source.. Eventually generated realistic videos too. I don't know what the solution is. It might be more levels of identity are eventually needed for something like email. We're also in a time people don't trust centralized sources.
I just don't think outlawing it would sufficiently work.
Back in the day far cry 1 was ultra realistic graphics.
The game was fun to play, the most annoying thing is if you throw a rock or grenade where they can see if your stealth meter in instantly goes to zero and they start shooting you.
I played it a lot, hehe.