Gaming would be a better place if AI was only meant to generate tweets for Timmy.
Community MetaSteam | January 2024 - Let's get this year started!
- Thread starter Mor
- Start date
- Status
UPDATE ?? :: Resident Evil Revelations General Discussions
am i the only one who got update ?!
Wake up babe it's time for Capcom's heel turn.
Looks like they reverted the change now, but they're obviously working on cracking down on mods.
When it comes to games visuals, personally I value a game's art style far more than I do games that push technical boundaries, games that push realism.
Even when it comes to mundane environmental assets, even without AI, there are games that lack a visual identity and games that lack visual cohesion, and I can't help but feel that AI generated assets may end up making these issues worse.
Though I guess this doesn't apply to games that are just pushing realism.
I don't believe the few benefits of AI are worth all the negatives because lets be honest we know why all these companies are racing to use it and it's not because the execs are salivating at the possibility that their artists or programmers' lives will be easier and games will be better. They're salivating at the thought of not having to pay salaries for half of their company. That insomniac hack really showed just how expensive employee salaries are, the massive bulk of SM2's $300m budget went to salary not marketing or other things.
We always choose the most dystopian route of virtually anything new we come up with. It's one thing when it's something smaller scale like choosing to put mtx in a game but this decision is going to lead to millions being laid off and also it's going to lead to most of our art and culture becoming so stale and soulless all so the people at the top can save money.
It can be fun but best coop experiences i had were when game was designed around coop and not when coop was simply there.
Also
XD
I genuinely don't see the fundamental difference between what is happening now and e.g. what people said when recording music first became possible. Before that, every time you wanted to hear music, you had to get an artist (or a group of artists) to perform it for you.
Photography is similar: you want a portrait? Before, you paid an accomplished artist for many hours of their work. Then basically everyone could do something with the click of a button.
To me those seem like extremely similar situations in the type of interactions between technology and art that result from a new invention.
I personally think that, just like with live music performances vs. recordings and paintings vs. photography, there will still be room for both manually and AI generated text/image/sound/whatever, based on both individual preferences and different use cases.
My biggest concern with generative AI is really in politics, or specifically in how it will be used to manipulate opinions. Closer to home, I expect that it will eventually reduce the quality of codebases in some aspects, but it's also already an extremely useful tool for some development tasks (and I do expect that it will eventually reduce the number of developers required for a given output).
I agree, but I'd rather also have the good coop experiences and not just the best
Last edited:
If I'm not wrong this should have the same functionality a DS R4 flashcard. If you want to deal with homebrew maybe should be better to install a chip IC and CFW. Let's see what happens.
Why let community mod your SF6 game when you can charge 100 euros for few cosutmes.
UPDATE ?? :: Resident Evil Revelations General Discussions
am i the only one who got update ?!steamcommunity.com
Wake up babe it's time for Capcom's heel turn.
Looks like they reverted the change now, but they're obviously working on cracking down on mods.
The difference is that previous advancements made the creative process easier, while generative AI completely bypasses the creative process.
Including photography, professional/high quality photography has a creative process to it as well, even if it's a completely different process from drawing, it's not just a click of a button.
Be it photography, digital drawing tools, digital audio tools, and so on, the creator has control of both the creative process and the end result and there's a progress to the creation.
The same is not true for AI generation, that control is gone, the creative process is gone, and instead it's just hoping the algorithm spews out an end result close to what the person wants.
Well, it's hard to explain, since after all, art (not just drawings but all forms of creative creations) is completely subjective.
Given that the most prominent voices in AI are libertarians or outright Nazis, I'd be very worried!
I feel like this is an extremely close comparison.
Everyone can take a photo, but there is a substantial appreciation for good, artistic, professional photography -- even though it doesn't require any of the traditional art skills that would be necessary to create a similar image without the invention of photography.
To me, this mirrors the situation with current image generation very closely. Everyone can get an AI tool to generate an image, but the vast majority of these images won't be anything special. (Just like e.g. random cell phone photos hold little interest from anyone other than the person who took them [and maye not even from them])
Well, as I said, I am worried. But I'm also worried that some leftists (which I consider myself to be too) act like it is possible, at this point, to stop generative AI being used, rather than thinking about ways to deal with it, or even leverage it.
That’s nowhere close to being true. Sure maybe in online spaces like X but in professional IT world AI is huge in very different branches of it.
I see your point, but there's one big difference between photography and AI generation, a professional/high quality photo takes skill to create, the photographer has control of how the photo is taken and has control of the end result, while getting a good result from AI generation depends on luck and persistence.
I'm not completely against generative AI though, I think it could maybe have potential for stuff like npc/enemy behaviour in games, for example, and I think it's also good for stuff like enhancing the backgrounds of really old games, like PS1 era games, and of course stuff like DLSS, but I just really don't like the way it's being used for art currently.
I agree that there are differences, but I think they are not so binary: there are also some skills in AI image generation, and some control over how it acts, at least if you e.g. want to create a specific composition you have in mind.
Just to clarify: I'm not saying that it's exactly the same thing, but I do feel like people often overstate the novelty and understate the similarity to previous changes in art prompted by technology.
For example (and I have to note that I'm not an art historian, so correct me if this is way off base!) it seems to me like the overall impact of artists with a photo-realistic style decreased after photography became widespread. I could imagine that we might see something similar with AI generation, in that the set of things AI does particularly well gradually become less impactful in terms of art compared to those aspects that need more of a human touch.
If we want to be positive about technology (maybe I played too much Talos Principle 2), then we could even imagine interesting outcomes from this. For example, let's say generative AI becomes really good at creating specific, well-established styles and compositions. That might then motivate more art being created in novel styles or with unique compositions.
Finally got this off my backlog. It was a fun time.
Like a Dragon: Infinite Wealth... GET!
I liked the demo and I really hope for some fresh content. They're eco-conscious devs for sure (with all that recycling), but seeing Mahjong, Outrun, etc. in every. single. Yakuza. doesn't make me want to play them more... on the contrary.
I liked the demo and I really hope for some fresh content. They're eco-conscious devs for sure (with all that recycling), but seeing Mahjong, Outrun, etc. in every. single. Yakuza. doesn't make me want to play them more... on the contrary.
First non deck handhelds to use Steam os 3
edit: they misslead everybody. its holoiso.
NEXT Return, AYANEO NEXT LITE, the Ultimate Value Choice Coming Soon… - AYANEO
Founded in 2020, AYANEO is a revolutionary brand for portable gaming consoles. Our team consists of passionate gamers and tech-savvy experts who want to provide gamers with the most advanced and up-to-date gaming experiences. We’ve developed a line of powerful and portable gaming handhelds, with...
ayaneo.com
edit: they misslead everybody. its holoiso.
Last edited:
Played a bit of The Outer Wilds and I don't get it. Annapurna continues its streak of not publishing games I like.
I gave this a try. Very atmospheric game, though I couldn't quite get used to the overly pixelated 3D look. Though the game does have plenty of options (which you cannot access in the demo, after having started it), so maybe in the full game I could find a way to make it easier to look at. Not much in terms of gameplay either, but maybe it is meant to be mostly a third person walking sim
Last edited:
Sea of Stars was amazing. Not only does it look great but has such an amazing soundtrack!
I find the whole TF2 Source 2 thing interesting, just because based off past fan projects, it looked like Valve was okay with people using the IP (did t people even ask for permission and stuff). But the main thing is that Valve specifically made the tools available to moders. In the case of TF2 on Source 2, they did not do that. You could use Source 2, (to make a mod for a game), but not to make a stand alone game by itself?
With Half Life and other games they had the SDK available didn't they? The same can be said for a lot of games and one of the complaints that people have had about Source 2 was that it was not available to people outside of Valve and Gary's company. And this is a result of that (both good and bad). It shows the lengths that people will go to, to get Source 2.
If the tools were readily available, then this would not have happened. As for the Portal 64 thing. I guess it's best not to get on Nintendos bad side. Things have gone rather decently, the last thing that you need is some one that sees you as competition.
With Half Life and other games they had the SDK available didn't they? The same can be said for a lot of games and one of the complaints that people have had about Source 2 was that it was not available to people outside of Valve and Gary's company. And this is a result of that (both good and bad). It shows the lengths that people will go to, to get Source 2.
If the tools were readily available, then this would not have happened. As for the Portal 64 thing. I guess it's best not to get on Nintendos bad side. Things have gone rather decently, the last thing that you need is some one that sees you as competition.
And there's plenty of UE4 games that can hit a locked 60FPS if you're willing to do config tweaks to turn down the effects quality an. Entirely depends if the game constantly overwrites cvars during runtime though.
Oh sweet, finally a Batsugun Steam release. Here's hoping they have the option to run the game with the original aspect ratio, because I had some problems with other rereleases not working right on my CRT monitor.
Apparently what caused Valve to DMCA that project was because they included TF2 assets in the project without licensing them?
EDIT:
Last edited:
I'm still of the position that all these companies that are firing people and replacing them with AI are going to have a reckoning when their value immediately drops and continues to drop. AI isn't at the point where it can replace anyone, yet.
AI is an incredible tool, but as an engineer and a AI developer, GPT-4 (top end LLM / code writing) makes grievous errors that an inexperienced developer would struggle with, let alone a non-developer -- and that's when being asked very specifically for something needed. These models know 0.001% of everything, truthfully.
There's a lot of practical tasks that I develop AI for, and it's stuff that a company would never ever pay for -- unless they were employing someone at exploitative / illegal wages. Literally no jobs have been lost within the scope of my work (so far, hopefully it will stay that way), but the value of products has increased. It has increased accessibility, ease of use, ease of finding information, etc.
I think the other big space beyond a helpful tool or doing lots of extremely low value labor is in entertainment. I've had tons of fun for the last few years playing with AI, and at least some of it will merge well directly into games soon. One of the biggest & most gamelike to me is 'chatbots' or otherwise interactive text. Whenever I think about what AI can do now, I think back to the frustrating evil KOTOR1 alignment, where evil just mean being a stupid chaotic dickhead. I feel that AI will be the solution for a lot of the critical limitations in RPGs. I still envision the quests and core writing to be completely human-written, but then the user may choose to do something clever, or stupid, and the game should react accordingly, rather than just giving a few non-branching reactions.
As far as AI & copyright, there's a big difference between an AI that is either overtrained or maliciously built to directly mimic or copy work such that actual money is lost by one or more individuals, and one that uses copywritten material in huge bulks, such that one constituent part, any one contributor, lends only the tiniest fraction to the end result. If the AI is well made, that's the goal. If I infer off a LLM, it will be taking literally billionths off of everything. Does taking a billionth off of a billion things deserve some compensation? I don't know. A lot of things are like this. Most huge corporations benefit hugely from taxes and work and our paltry compensation is a tiny fraction of what we actually contributed. That said, I'd prefer all AI models be free and open source so that everyone can benefit and they aren't used for exploiting profit.
Really, I'd prefer if anything was made for free, it could heavily use copyrighted stuff. I miss when there were cool fan made games that used game assets or movie assets, or cool music videos made with copyrighted stuff. That was all before the Internet became a money-making machine.
That is to say, if the left cannot find some comfortable position in the AI space, the voices of the right will exclusively be amplified by the power of AI. When AI started becoming more of a thing a few years back, I didn't stick my head in the sand out of fear. I directly began learning it so I could be on top of the wave instead of drowning. In terms of influencing the future of AI, there needs to be leftists and the position cannot be refusal to partake.
As far as actual AI developers, on my team there are trans people, leftists, people who are angry with AI, and yeah the libertarians are there, too. I'm sure there are people that are more towards the right (if anyone is far-right, I haven't detected it but that's not to say they don't exist).
AI can give you some control. Control is not gone. Is the range limited? Yes. Does that mean artists are safe? Yes. AI can only generate a fraction of what humans can make.
Can AI art be 'art' or a form of expression. Yes. Can you have enough degree of control to assert authorship over works. Yes. Is it wholly your own? Fully controlled? Fully ethical? No, I'm sure it isn't, I haven't been inclined to sell any AI art nor do I intend to. But I actually can and do express myself with it, as well as 'explore' it, as if it's a video game someone else made.
Really, I think the threat is very overblown. Try making a horse riding an astronaut in the most advanced (compositional) AI -- Dall-E3 (available on Bing Create). It cannot be done. OpenAI claims to have done it (they touted it). I and others in the AI community have tried many times. Can't do it.
Oh, and I've spent probably several hours working on a single AI generated jpeg at times, besides AI models I've trained for my own enjoyment -- such as being able to create photoreal images of WipEout (the hover racing game), or new stills from Star Trek TNG episodes that don't exist). Usually I use Photoshop / digital editing too, because there's still so much that cannot be done (I'd call it the majority). But it's still expression if a person is expressing.
AI is an incredible tool, but as an engineer and a AI developer, GPT-4 (top end LLM / code writing) makes grievous errors that an inexperienced developer would struggle with, let alone a non-developer -- and that's when being asked very specifically for something needed. These models know 0.001% of everything, truthfully.
There's a lot of practical tasks that I develop AI for, and it's stuff that a company would never ever pay for -- unless they were employing someone at exploitative / illegal wages. Literally no jobs have been lost within the scope of my work (so far, hopefully it will stay that way), but the value of products has increased. It has increased accessibility, ease of use, ease of finding information, etc.
I think the other big space beyond a helpful tool or doing lots of extremely low value labor is in entertainment. I've had tons of fun for the last few years playing with AI, and at least some of it will merge well directly into games soon. One of the biggest & most gamelike to me is 'chatbots' or otherwise interactive text. Whenever I think about what AI can do now, I think back to the frustrating evil KOTOR1 alignment, where evil just mean being a stupid chaotic dickhead. I feel that AI will be the solution for a lot of the critical limitations in RPGs. I still envision the quests and core writing to be completely human-written, but then the user may choose to do something clever, or stupid, and the game should react accordingly, rather than just giving a few non-branching reactions.
As far as AI & copyright, there's a big difference between an AI that is either overtrained or maliciously built to directly mimic or copy work such that actual money is lost by one or more individuals, and one that uses copywritten material in huge bulks, such that one constituent part, any one contributor, lends only the tiniest fraction to the end result. If the AI is well made, that's the goal. If I infer off a LLM, it will be taking literally billionths off of everything. Does taking a billionth off of a billion things deserve some compensation? I don't know. A lot of things are like this. Most huge corporations benefit hugely from taxes and work and our paltry compensation is a tiny fraction of what we actually contributed. That said, I'd prefer all AI models be free and open source so that everyone can benefit and they aren't used for exploiting profit.
Really, I'd prefer if anything was made for free, it could heavily use copyrighted stuff. I miss when there were cool fan made games that used game assets or movie assets, or cool music videos made with copyrighted stuff. That was all before the Internet became a money-making machine.
Prominent voices, I'd agree. And I am fucking worried about the discourse online. Because just like any issue these days it has become politically polarized on social media. Noone can think outside of their box. You either tow the line of your followers or you're out. It didn't used to be as big of a problem on the left, but now it is.
That is to say, if the left cannot find some comfortable position in the AI space, the voices of the right will exclusively be amplified by the power of AI. When AI started becoming more of a thing a few years back, I didn't stick my head in the sand out of fear. I directly began learning it so I could be on top of the wave instead of drowning. In terms of influencing the future of AI, there needs to be leftists and the position cannot be refusal to partake.
As far as actual AI developers, on my team there are trans people, leftists, people who are angry with AI, and yeah the libertarians are there, too. I'm sure there are people that are more towards the right (if anyone is far-right, I haven't detected it but that's not to say they don't exist).
I have done traditional art for many years, but never professionally. But now I can say I've done AI art for several years, never professionally.
AI can give you some control. Control is not gone. Is the range limited? Yes. Does that mean artists are safe? Yes. AI can only generate a fraction of what humans can make.
Can AI art be 'art' or a form of expression. Yes. Can you have enough degree of control to assert authorship over works. Yes. Is it wholly your own? Fully controlled? Fully ethical? No, I'm sure it isn't, I haven't been inclined to sell any AI art nor do I intend to. But I actually can and do express myself with it, as well as 'explore' it, as if it's a video game someone else made.
Really, I think the threat is very overblown. Try making a horse riding an astronaut in the most advanced (compositional) AI -- Dall-E3 (available on Bing Create). It cannot be done. OpenAI claims to have done it (they touted it). I and others in the AI community have tried many times. Can't do it.
Oh, and I've spent probably several hours working on a single AI generated jpeg at times, besides AI models I've trained for my own enjoyment -- such as being able to create photoreal images of WipEout (the hover racing game), or new stills from Star Trek TNG episodes that don't exist). Usually I use Photoshop / digital editing too, because there's still so much that cannot be done (I'd call it the majority). But it's still expression if a person is expressing.
Last edited:
Time to leave a negative review
UPDATE ?? :: Resident Evil Revelations General Discussions
am i the only one who got update ?!
Wake up babe it's time for Capcom's heel turn.
Looks like they reverted the change now, but they're obviously working on cracking down on mods.
It's less about the engine, and more what performance targets the game sets. Games that are meant to run at 30 FPS on PS5 or Xbox S X will generally not scale to 60 FPS on the Steam Deck. whereas games that target 30 FPS or higher on PS4 generally do run quite well. It's ultimately about the level of performance we can expect from the Steam Deck. Most of those UE5 games are doing a lot of fairly expensive stuff, like Lumen and Nanite, which you can't exactly "turn off" by my understanding, it's just part and parcel of what games that use these technologies do.
Perhaps a Steam Deck 2, with a truly generation bump in performance at the same power levels, can do better here, but short of running at 30W or above and basically not having any useful battery life, there isn't much you can do about that on a handheld in 2024.
I meant about control over the creative process and not control over the final result. You have some control on the output, but you have no control over the creative process, you have no control about how a drawing is made, or how a piece of music is made, and so on.
I said in a previous post, personally I like when art is created, and not when art is generated, though, I have no problem with hobbyists doing AI art for themselves, as long as they don't attempt to sell it, don't use it for commercial purposes, or don't pretend that it's not AI art/don't hide that it's AI art.
Last edited:
Yeah this is where I am at too when it comes to using AI that is trained on others art. It seem too common that people pretend to not use AI and sell it and I don't know how to feel about it. In some cases it could be that it is something that helps that person eat that day and I can't hate on that. But I would personally not feel ok buying that and I don't even really like seeing AI art. But I still believe AI is a good tool for artists, like training it on its own art and then using it to speed up and help with menial work that's part of all creative processes that take the most time.
PS5 DualSense V2 Controller with improved battery life seen on the Best Buy website!
- “Exceptional 12-hour battery life, on a full charge, supports long gaming sessions”
Make controller with 4h battery life, sell it for few years, put bigger battery after that and make it v2 version of same controller, profit?
It's so depressing to see Capcom go from having stellar PC ports, great engine, even having Mac ports ! To become complete shits.
I don't agree with what they're doing about mods, but I understand why they're having this reaction, after a Chun-Li nude mod was used in a SF6 tournament, and also after having to fix monster hunter rise save files that were borked because of a mod, I don't know the exact details, but seems like an anomaly investigation (a quest) was modded into the game, users with the mod went online, users without the mod could join the modded sessions, result was borked save files.It's so depressing to see Capcom go from having stellar PC ports, great engine, even having Mac ports ! To become complete shits.
They don’t give a shit about nude mods it’s just they want money and avoid community to release costumes they cannot get money.
Just take a look what the hell they’re doing with street fighter 6 dlc when they charged 20-30 euros for full set DLC costumes on SF4 or 5 and now they are charging 100-120 for them, and community give better costumes for free.
If capcom itself could monetize the nude costumes without going against platforms content policy or audience like resetera forum, be sure that they would release them themselves.
Just take a look what the hell they’re doing with street fighter 6 dlc when they charged 20-30 euros for full set DLC costumes on SF4 or 5 and now they are charging 100-120 for them, and community give better costumes for free.
If capcom itself could monetize the nude costumes without going against platforms content policy or audience like resetera forum, be sure that they would release them themselves.
Last edited:
Ayaneo lite is not using steam os 3.. rather its using holo iso
handheld-hq.com
AYANEO NEXT LITE to Be First Non-Valve SteamOS Handheld - Handheld HQ
AYANEO's NEXT LITE handheld has been confirmed to be the first SteamOS device outside of Valve's Steam Deck.
handheld-hq.com
While I don't think the tournament chun nude mod is the only reason, it feels like it was part of the decision making.
It's not like Capcom games haven't had mods for years and years already, they could have gone this route years ago if they wanted to. So it just feels like something has changed.
It's not like Capcom games haven't had mods for years and years already, they could have gone this route years ago if they wanted to. So it just feels like something has changed.
That's on Valve. Steam Deck has been out for two years now and there is still no clear way for 3rd parties to sell devices with Steam OS3 on it.Ayaneo lite is not using steam os 3.. rather its using holo iso
AYANEO NEXT LITE to Be First Non-Valve SteamOS Handheld - Handheld HQ
AYANEO's NEXT LITE handheld has been confirmed to be the first SteamOS device outside of Valve's Steam Deck.
I know but they specifically said in their announcement that i posted earlier today that
"AYANEO NEXT LITE comes per-installed with the SteamOS gaming system for the first time"
No mention of Holoiso.
They responded and clarified after the fact.
Last edited:
No Granblu Fantasy Relink demo for PC just for PlayStation.
I'm fully convinced they'll release SteamOS 3 for third partis the same day they'll release it for Desktop users, which is to say not anytime soon.
I'd bet they wanna fix any potential issue with the OS so that players have a smooth experience as much as possible on every device, including Nvidia desktop machines, meaning a huge effort since you don't control the hardware like the Steam Deck.
- Status