legend

1 hour ago, Commissar SFLUFAN said:

 

She was Adam's first wife and she was cast out of Eden because she wanted to be in the dominant position during sexual intercourse.

Christians are always threatening us with a good time.

24 minutes ago, stepee said:

Also apparently the new path tracing runs at native 4k at 15fps on a 4090 lets go boys 

The DLSS3 sounds like good performance though! Unfortunately, I'm still using my 3080 and I don't know if it will translate to decent performance on that.

It's always fucking CDPR that makes me want to upgrade!

1 hour ago, stepee said:

 

I played mostly with blades, running around all crazy, going high on speed and agility, and slashing everyone up. That was the most fun for me but I don’t remember the specifics of how I specced! Maybe @legend remembers those details better as he played similarly!

I do not remember either unfortunately! It's been a while and I've been waiting for the expansion before I play more again. But it feels like the expansion is always just out of reach!

Alright, they got me with this one!

25 minutes ago, Commissar SFLUFAN said:

 

I really, really wanted her to use the name "Skywalker" as a description of a particular type of Jedi who uses Luke's philosophy.

 

If she had to adopt a surname, it really should've been "Solo".

Yeah "Skywalker" as a new Jedi philosophy would have been more interesting. I suppose they could still force it into that direction. All indications in the movie is that it's a superficial name choice, but it's the last word in the series, so they could always re-justify it later. Maybe give JJ a bit of his own medicine of other directors changing the interpretations in the next movie  

I like Daisy even though I found Rey's arc shallow and find the whole "Skywalker" name choice stupid. Hopefully they can do better by her.

Yeah, that works for me. Good trailer.

3 hours ago, Keyser_Soze said:

@legend How come my RL agent isn't learning?

Fortunately no one at my work does that. But I have had people I don't really know on Discord DM me because they know of me from an AI server and want to ask me something but all they send his "Hi" and wait for me to respond before sending anything more. It feels like a trap so that once I respond they feel they'll have my attention to dive into some really long and ill-defined question that I won't really want to respond to. Like "how come my RL agent isn't learning?" I dunno, because you did it wrong? Debug it or do some data analysis for Christ sake. I don't know anything about you, nor what you're doing and I'm not going to download your code and debug it myself for you. Ask a specific fucking question or don't bother. Or, if one more person wants to tell me about this really "novel" idea they have to apply RL to market trading but they just need some one to do all the work for them even though they're being told it's a terrible idea for a million reasons, I'm going to gouge someone's eyes out.

I guess this thread stirred something in me.

While there are obvious advantages to the Windows handhelds, I'm kind of hoping they all fail and that Steam Deck is a wild success to further encourage Linux gaming support. If I could remove Windows from my life, that's a win, and gaming is the only reason I use it.

2 hours ago, stepee said:

 

Sounds like they just want to be the fantasy version of freemasons and using a total control of all AI technology to accomplish that goal. Almost just openly saying that they want to make sure they are part of an exclusive club that controls the world. Ambitious, I respect the hustle.

That's probably more true than I'm comfortable with. They will always tell you they're doing what they do for the benefit of all mankind, but the fact that they so specifically benefit from their "altruistic" efforts is extremely suspect. Some of the worst actors in our history would tell you they do what they do for the benefit of mankind, so it's not like this would be new.

4 hours ago, Bacon said:

That's the only service no one has.

FTFY

1 hour ago, stepee said:

 

I figured there was something behind this like that but what is the benefit to them to try to it restraints on AI? Is it to make sure they are the (only) ones with the keys to use the technology for their own purposes?

If we took them as wholly honest actors, these people believe that AIs will transform from low intelligence to all-powerful genies who will destroy mankind in a matter of weeks, simply by thinking really hard. The AIs won't need to do, you know, science that requires interacting with the world, to advance even. It can just think really hard for those few weeks on the data that exists and divine all the missing answers to the universe. (No, I'm not kidding, this is what they believe.)

Of course, that's only if AI isn't solely developed by the "right people"--mostly philosophers. If the right people work in secret on AI, they can make a benevolent genie instead that will allow trillions upon trillions of future simulated humans to exist and colonize the universe. All current social issues are irrelevant to this goal because what's millions of people suffering to a potential future of trillions of trillions of happy virtual people? You might be starting to see why this particular branch is called "The Future of Life Institute."

So consequently, they want to be the ones in charge of AI development and they want all the funding for their "safety work" which is typically bad mathematics and decision theory with manufactured problems because they don't actually understand the basics. And those who claim to be on their side and are influential are often the most irresponsible when it comes to AI tech. Elon Musk, for example, likes to sign on to these things and play the savior from the big bad AI genie, while simultaneously beta testing his shitty autonomous driving on the general public.

There's those among them for whom this is probably just a convenient set of beliefs to empower themselves and they get to pretend to be the savior at the same time. I suspect it's a combination of "true believers" and people in it solely for the convenience. Some probably only partially believe it, but believe enough of it to go along. Sam Bankman-Fried is among their cult, and we saw how well that turned out. I think the primary person responsible for this "movement" Eliezer Yudkowsky drinks his own koolaid though and honestly he's starting to sound dangerous. 

13 hours ago, Remarkableriots said:

Musk and Wozniak among 1,100+ signing open letter calling for 6-month ban on creating powerful A.I.

FORTUNE.COM

---

The letter urges technology companies to immediately cease training any A.I. systems that would be "more powerful than GPT-4."

 

2 hours ago, SuperSpreader said:

 

John Carmack is gonna kill us all. @legend

The open letter is immense BS and many of the signatories were suckered by the cult that is The Future of Life Institute. FLI has a history of getting experts to sign onto open letters for some sensible parts they include in their letters along with their more extreme ideas. They will even argue with you and try to guilt you into signing it to support the sensible parts since those parts are important. Then they always focus the extreme parts once they go public and use the signatures to give the illusion that experts agree with their nonsense in its entirety. I know multiple people who got burned by this tactic on this letter. Many are trying to back pedal now with the fallout or at least clarify they don't agree with all of it.

More people need to understand that FLI is a cult.

For some perspective on that, the FLI letter cites the Stochastic Parrots paper as evidence that experts agree with their claims, and the authors of the paper were none too pleased that they were disingenuously co-opted. You can see their response here:

The DAIR Institute

WWW.DAIR-INSTITUTE.ORG

---

We are an interdisciplinary and globally distributed AI research institute.

If the game is as good as BotW, they deserve $70.

In the dog fight I kept expecting to hear "You're good! But I'm better!"

7 minutes ago, SuperSpreader said:

I don't get it. 

 

21 hours ago, SuperSpreader said:

 

I don't think this is true. From what I've seen is they ONLY speak in memes. 

This was prescient

As a point of reference, Nvidia released CuDNN, their deep neural network library, in 2014, and they regularly make big updates for it. All major high-level DL libraries target it for the GPU support, to the point that if you're doing DL you probably need an Nvidia card. (The only real exception is if you're using Google's TPUs on a cloud platform.) So yeah, they've been pushing on AI for quite some time now.

I think complete digital segregation would be bad for society.

I also think the article's claim was much more nuanced than that and was pointing toward particular memes that may be demeaning.

I don't think it's worth worrying about whether a person of the same race is "allowed" to still share such memes. First, if it's perpetuating a bad idea, don't participate. Why would you worry that you "can't" participate in something that's bad? Second, people who are struggling with things often use humor amongst themselves to cope with it. You jumping in when you don't suffer from it isn't the same -- you're not coping, which means it's not the same thing. Human psychology isn't divided into clear neat rules.

Finally, only a bad comedian continues to make a joke that hurts his audience while gaining nothing of value in the process.

I love the dialog in this show. It really coveys what giant tools these people are.

8 hours ago, BloodyHell said:

short films

This is the way.

27 minutes ago, Spork3245 said:

It’s weird that the people defending Mandalorian S3 are complaining about Ted Lasso S3 with similar complaints. “Feels aimless”, “episode x was weak”, “the stuff with character x is boring”, etc.

I’m enjoying both shows but

I'm also enjoying both.