Jump to content

legend

Members
  • Posts

    30,120
  • Joined

  • Last visited

  • Days Won

    1

Posts posted by legend

  1. 14 hours ago, Ghost_MH said:

    Someone explain to me that I'm just exhausted, because this is either a great trailer or this actually doesn't look like trash.

     

     

     

    Never watched one piece. Animes that go on forever never really interest me much. (Side thought: maybe if I got high regularly these kinds of animes would be more interesting).

     

    That said, while this trailer suggests they're giving it the old college try, I'm still unconvinced they can pull it off. Good animes are often good because they're animes. Even the more grounded ones tend to lean into the art direction to their advantage. Animation allows you to exaggerate emotions and aspects of the scene without it breaking immersion. You might still get a good story without that, but I don't see anything being gained from it being live action either.

     

    Maybe it being a live action means they have to make it more concise though, in which case maybe that is a reason for me to watch the live action version since I otherwise don't want to engage with the never ending random stories. That might be the best reason for it I can give!

  2. 25 minutes ago, BloodyHell said:

    I think i misunderstood. Racial segregation is disgusting. But I thought you were saying living in the country was “the most boring existence possible”.

     

    I see. No I did not mean that. There are lots of virtues about living the country. I own a house and even though I am *far* from an outdoors person I love the sounds of the wind in the trees, birds, etc. I absolutely get enjoying that.

     

    25 minutes ago, BloodyHell said:

    Cities are concrete jungles though. You can add parks and trees, and rooftop gardens, but it’s still incredibly unnatural.  Im glad you guys enjoy it, more cheap land for the rest of us out here, but I’ll never understand the attraction of crowds of people, little space or privacy except inside your own walls, and no real nature. That to me would be far more boring than living on my acreage with room to do whatever I want. No HOA’s, no landlords, no bylaws, no neighbors being nosy. One day in a city, even on vacation, im ready to leave.

     

    But they're not. I live in New England and a lot of the buildings in various cities are wood. There are lots of different ways to build cities. I do think many cities fall far short of what cities could be though. Home/rent pricing in particular is often completely fucked, but this usually has to do with auxiliary issues and various forms of corruption rather than being an intrinsic property of cities. My understanding is home pricing in Toyko is far better than in many US cities despite Tokyo being one of, if not the, most dense cities.

     

    25 minutes ago, BloodyHell said:

    but if you meant living with only one race and shunning any kind of diversity, then I agree, that’s gotta be a miserable existence. 

     

    :thumbup:

    • Halal 1
  3. 16 minutes ago, BloodyHell said:

    I mean, these people are reprehensible, but im not sure what a ‘boring existence” means to you? 

     

    Lack of cultural exchange limits diversity. Limited diversity is boring in itself and stagnates progress, which in turn is boring.

     

    16 minutes ago, BloodyHell said:

    Cities are the most unnatural thing we’ve ever devised.

     

    So is medicine and airplanes and cars and rockets and computers and robots and boats and video games and basically anything humans of today use and enjoy regularly. Who gives a shit what's natural?

     

    16 minutes ago, BloodyHell said:

    I personally can’t imagine living in high rise apartments in a concrete jungle, or even suburbia. That to me would be the most awful and boring way to exist. I can’t imagine leaving my cement and brick box to exist in more cement and brick outside, in a space with no semblance of nature or wilderness, wildlife, rivers, secluded beaches. That to me is beyond boring, it’s torture. Yhe next thing to prison. So boring is relative. I’ve never been bored. I can hunt deer behind my house, I can fish trout/bass/salmon in the river in back acreage. I can walk on my land without neighbours calling the cops.

     

    Putting aside the weird limitation that cities can only be "concrete jungles," what the hell does race segregation have to do with living in a city?  I don't live in a city myself.

     

    • Halal 3
  4. 1 hour ago, Signifyin(g)Monkey said:

     

    Yeah, I think he's talking about the concrete implementations.  I'd wager he'd be skeptical that you can effectively apply AI to any system involving exchanges between human beings without continuous input from human beings, even if it utilizes unsupervised learning.  Although I've heard him concede that it's possible if you're dealing with a closed system in a controlled environment, like a game with a limited move-set.  (But most of the important systems that operate within society are open, evolving ones, in uncontrolled environments.)

     

    Actually, that idea is one reason why I find the history of AI-generated language translation really interesting.  According to the accounts I've read, the early attempts to achieve it involved trying to build the most accurate data-model of human language as possible, and it was assumed once that was done you could translate text/speech between one language and another without using large data sets.  But it never really worked until researchers shifted to using large data-sets of actual translations run through correlative statistical algorithms, LLM-style.

     

    Do you find, working within the field, that the original idea of finding the original 'golden key' of the pristine Chomskyan data model for language still alive?  Are there those who think Big Data techniques will eventually be made obsolete in this respect?  

     

    Yeah, early AI attempts were very much oriented around rule-based system and logical deduction. We usually refer to that as "GOFAI" which stands for "good old fashioned AI." Pure GOFAI is effectively dead and very few people believe in it. The world is too complex for humans to encode all the necessary rules and it doesn't have especially good mechanisms for learning from new information, a hallmark of animals.

     

    However, there is a need to give current methods more of an ability to reason. The trick is how to endow systems with reasoning that is compatible with learned models that are more murky. There's actually a bunch of good work in this direction. In my field it would fall under "model-based reinforcement learning" where agents statistically learn models of how the world works and then plans within them (typically still using statistical planning methods, but plan all the same!)

     

     

    1 hour ago, Signifyin(g)Monkey said:

    Not coincidentally, he has a fairly iconoclastic reading of the Turing Test.  He has argued that there's no way to know whether a machine passes the Turing test because it has become smart enough to convince the observer that it is a human being, or whether it is because the observer has become dumber--dumb enough to believe the machine is human, when it clearly isn't.  He frames the text that articulated the test itself as partially a reprimand or cry for help from Turing towards the institutions and people he worked for, driven by the way he was treated for his homosexuality, i.e., "if you're dumb enough to forcibly drug me with hormones for being gay after I basically helped you win WWII maybe you're dumb enough to believe machines and people could be the same".

     

    It's a fascinating interpretation, but I'm guessing not a popular one.

     

     

    You might have already known, but just to be clear, the Turing test is not the same as the Church Turing thesis. The former was Turing's rhetorical argument on intelligence, the latter is a computational theory claim (the foundations of which were simultaneously developed by Church and Turing, hence the name) which is far broader and is effectively the foundation for all computer science. Overturning the Church Turing thesis would be to computer science what overturning evolution would be for biology. The ramifications for what it would mean about reality would be really weird and extreme.

     

    On the topic of the Turing test though, it test succeeds as rhetorical argument, but fails as a literal test because it is very easy to dupe people with mechanisms that are pure smoke an mirrors. So I agree with him on that much. I don't think his framing of Turing's argument for the Turing test matches the historical context or content of the work :p Primarily, Turing wanted to object to the fuzzy and kind of mystical descriptions people had toward intelligence (or "thinking" more specifically) and push them towards adopting a functional perspective. If a machine can do all the cognitive things people do, it doesn't really matter what your fuzzy mystical thoughts on intelligence are. All we should care about is the function and we can evaluate that directly without untethered philosophy.

     

    There's some debate over whether Turing did think the imitation game would also be a good test or if it was purely for the rhetorical argument, but either way his drive was moving towards functional ways of thinking about intelligence.

     

    Even if the Turing test as a literal test was better and less prone to humans being duped, it's also kind of unnecessary for most analysis of AI.  We know what we're building and what it lacks. It is necessary to empirically measure how good any AI system is at something, but we don't really need to empirically test what kinds of cognitive faculties it has. I don't need to "talk" to Chat-GPT to know about the seriously lacking cognitive faculties it has, for example. I know many things it lacks by its construction. People get overly swept up by AI scientists pointing out that these models are black boxes we don't understand. We actually understand a lot. What is usually meant by that is we can't tell you precisely how the process it carries out works because it's an incredibly complex process. If we knew how it worked precisely, we wouldn't have needed machine learning in the first place. We would have just programmed the process ourselves! But while we lack that precise knowledge of how it reaches its answers, we very much know how it learns to do that and what limits that imposes.

    • Halal 1
  5. 1 hour ago, Signifyin(g)Monkey said:

     

    On the last point, I always thought it was more of a reframing rather than a dismissal.  The claim is that intelligent systems with human-like cognitive abilities are the product of processing input taken from human sources, and thinking of them as freestanding, autonomous minds (or 'intelligences') hides the role these human sources (who are freestanding, autonomous minds) play.

     

    Much of the practical implementations of modern AI use human sources for the data, but that is by no means the limit of the field nor the end goal. My primary sub field is reinforcement learning and by default there is no human information in reinforcement learning. The agent learns autonomously by interaction with its environment. Back during my postdoc, a lot of the work I did was to *add* the ability for RL agents to learn from additional information from people (e.g., through human delivered feedback or commands) because teachers can accelerate learning and allow for user-tailored agents.

     

    The aspirations of AI are very much to move passed supervised learning. It's also why it's hilarious that people think LLMs are near "AGI" when LLMs are still just doing limiting dataset fitting of very well structured human data.

     

     

    1 hour ago, Signifyin(g)Monkey said:

     

    I think it comes out of his 'humanist' view of intelligence and the mind in general.  While he has a fairly nuanced view on the nature of human consciousness (which I think he kind of implies is a necessary predicate for true 'human intelligence' in his arguments), he strikes me ultimately as someone who views it as something unique to humans that a digital machine probably isn't capable of replicating. (particularly the aspect of subjective experience or 'interiority')

     

    That's kind of my interpretation, though, he never says that explicitly.  He touches on his views here--and notes the irony of being a big fan of Daniel Dennet, who holds very different opinions from him:

     

     

     

    Well if he does, that seems very unlikely to be true :p  Human-like intelligence is only limited to humans if either the Church Turing thesis is wrong or the "hardware" of brains is the only kind of hardware that can manage the computational complexity of intelligent processes in practice. The former is very unlikely to be wrong and if it's wrong it has *vastly* weirder implications about reality than just what can be "intelligent." That latter is also very unlikely because evolution doesn't select for intelligence and populations evolved to it only through a series of local hacks. (Although I do think it's very plausible that we will need a hardware revolution for more intelligent machines -- but probably not one that needs to look just like brains)

     

  6. 1 hour ago, Signifyin(g)Monkey said:

    Meant to post this awhile back in this thread.  When it comes to the AI debate, Jaron Lanier is one of the most formidable minds on the subject and whether you agree with his positions or not, you'd be intellectually lazy not to reckon with his arguments.  Formerly they have been relegated to his books (Who Owns the Future and Dawn of the New Everything are good ones to start with) but he published a New Yorker article touching on some of his main themes that is a must-read if you're interested in the subject:

     

    lanier_ai.jpg?mbid=social_retweet
    WWW.NEWYORKER.COM

    There are ways of controlling the new technology—but first we have to stop mythologizing it.

     

     

     

     

    I both like and dislike this article :p 

     

    Things I like

    • Demystifying LLMs instead of embracing the complete an utter nonsense capabilities OpenAI wants you to believe it possess.
    • Focusing on more tangible risks with AI technology
    • Promoting more transparent AI that can be reasoned about

     

    Things I dislike

    • Reinforcing the incorrect view that AI researchers buy into catastrophic risk scenario by pointing to poorly conducted pools and a handful of legitimate scientists buying into the unscientific and unhinged speculation. Myself and so many of my colleges are regularly exacerbated by this narrative primarily fueled by PR focused tech bros with a savior complex.
    • Dismissing "AI" as concept merely because our current technology is far from anything like human intelligence. It's still the aspiration of the field to work to intelligent systems with human-like cognitive abilities. But it's important to note that there is a huge difference between human-like cognitive abilities and human-like. We're building the former, not the latter.
  7. 8 hours ago, Bacon said:

    It's fun overall, but my major complaint is the loot system. Ever since Nioh, the Diablo Looter Shooter Loot Pinata gearing sucks. Like, I'm out here spending what feels like hours dismantling gear. You fill up the 600 inventory slots hella fast. There is a loot filter system that automatically dismantles gear you don't want, but this early on the only gear I don't want is shit without Job Affinity, and even then grey gear still matters if it won't put me under the required amount of affinity.

     

    I generally like loot, but that does sound like a bit much! Perhaps I'll wait for a sale.

  8. 2 hours ago, Xbob42 said:

    I'm bewildered at WHY they sell gangbusters. Every single element of their games is usually bottom tier trash except maybe KINDA world design, and anything they do even kind of good has been completely eclipsed since Skyrim's release. 

     

    They do absolutely abysmal combat, boring and tedious traversal, middling voice acting, bottom of the barrel animation, hugely noninteractive worlds sans like... fucking stacking cheese or whatever, lame and uninteresting stories, generic characters, simplistic character builds, world class bad difficulty modifiers, barebones AI routine scheduling, hilariously bad dialog choices that don't mean anything, every single conversation is framed in the exact same embarrassingly simplistic way, basically nonexistent enemy AI aside from "get in range, spam attack until you or I is dead" and Jesus I could literally go on for another 3 paragraphs. 

     

    Anything that made games like Skyrim special when it came out has been done and done better, and I think a lot of Bethesda fans who actually play other games are probably in for an unpleasant epiphany. I'm sure many (especially those who only play Bethesda games whenever they come out) will be satisfied enough, but I really don't think we're gonna be looking at universal acclaim and cultural phenomenon meme status here unless they actually have accomplished something I just haven't seen yet, which seems unlikely from Low Bar Bethesda.

     

    Some folks keep bringing Zelda comparisons into every thread, so I guess I'll do it here: Go talk to any NPC in Zelda, especially Gorons, and go look at that fucking Nintendo game on a now low-tier tablet run circles over the animations and facial expressions in a $7 billion studio's flagship title with a virtually unlimited budget. Come on. 

     

     

    When Oblivion released, it was wildly impressive and led the way for open world games. Skyrim was less of an achievement and I was deeply annoyed that they removed my favorite feature: being able to make your own spells (which if you speced your character build for mana led to some absolutely awesome custom spells), but they were still unique enough that it was a stand out game.

     

    But I agree, since then, open world RPGs have evolved so fucking much and I don't know if Bethesda has moved with it. I hope they have and I'm sure I'll still enjoy Starfield. But the demos haven't made me think they can rival the best.

  9. For PS5 games, I think Horizon holds the title of best looking game by a fair margin.

     

    But I agree with @Spork3245 that the handling of HDR (or maybe just the art direction around it) in FFXVI is pretty stunning. The blacks and the contrast with the light sources immediately grabbed my attention and the game wow'ed far more visually than I had expected from the (non HDR) trailers.

  10. 2 hours ago, Spawn_of_Apathy said:

    The biggest thing I want from Starfield and why I hoped Mass Effect went a similar route was I want a sense of exploration, wonder, and planet hopping in something that feels more like an action game. 
     

    I don’t want a simulation or crafting and survival mechanics. I don’t mind some … minor resource management, but I’d rather not even need to deal with fuel unless it’s cheap, very plentiful, and lasts a long time. Maybe saving up to buy or upgrade the engines to jump further, sure.
     

    But I want proper quests and stories to take me around the galaxy with minimal hindrance.  Mining for resources, crafting, base building like in No Man’s Sky is fine if it is all optional. 

     

     

     

    I'm not sure I agree with the idea that No Man's Sky is a good alternative for the kind of game Starfield is going to be, but FWIW, in one of the No Man's Sky updates, they let you customize the shit out of things like fuel usage and crafting to be closer to what you would want (e.g., you can make fuel free). Not saying you should go give it try because it will change everything, but it is at least worth knowing about!

     

    I sadly cannot directly embed their little gif that shows all the options, but if you scroll down on the below page, you can see what they are.

     

    nms-waypoint-book-cover.png
    WWW.NOMANSSKY.COM

    Waypoint: Introducing Update 4.0 Access the universe of No Man’s Sky from anywhere in update 4.0, WAYPOINT! Launching for the first time on Nintendo Switch™, and overhauling fundamental elements of gameplay including game modes, inventory size and usability, milestones, journey cataloguing and much more! Buy now on More purchase options » Out now on More... View Article

     

    • Halal 1
×
×
  • Create New...