[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/tv/ - Television & Film

Name
Spoiler?[]
Options
Comment
Verification
4chan Pass users can bypass this verification. [Learn More] [Login]
File[]
  • Please read the Rules and FAQ before posting.

08/21/20New boards added: /vrpg/, /vmg/, /vst/ and /vm/
05/04/17New trial board added: /bant/ - International/Random
10/04/16New board for 4chan Pass users: /vip/ - Very Important Posts
[Hide] [Show All]


[Advertise on 4chan]


File: images-34.jpg (51 KB, 678x452)
51 KB
51 KB JPG
I'm STILL waiting for Chrome Lords
>>
>>216797753
Better pic
>>
>>216797778
Why do AI pics look worse than this today?
>>
>>216797793
something about those early models was different, they lacked resolution but it was like they had more dream material in them, they were more mysterious and alien

the new ones are too refined and micro managed
>t. berrymix appreciater
>>
File: Image_fx.png (1.41 MB, 1408x768)
1.41 MB
1.41 MB PNG
>>216797778
I've still been having fun with it. I think the "___ as a dark 80s fantasy" is still possible, but everyone just got tired of doing that.

AI tools are really fun for hobbyist story tellers and I've been having fun making alt-history stuff with it. Pic Related, images from the German-Soviet Victory ceremony in Berlin, 1944.
>>
>>216797793
datasets have all been tainted, it was inevitable. these were all made at the right time
>>
>>216797793
Midjourney (the AI used to gen those images) can still make them.
>>
>>216797793
You still can, just have to give it very specific prompt
Otherwise it would just use the default, generic style
>>
>>216797793
>>216798355
>>216798787
it's because of reinforcement learning. new image models (2024+) are all trained on synthetic slop to follow the prompt with no concern for style or creativity
>>
>>216798811
Skill issue.
>>
File: Image_fx (3).jpg (138 KB, 1365x768)
138 KB
138 KB JPG
just tried to make a similar image and its not working.
>>
proof that plebs don't even like art. they just like arbitrary genres.
>>
>>216797793
>Why do AI pics look worse than this today?

That picture is from years ago!!!FACT!!!
>>
File: smoke.jpg (278 KB, 1344x896)
278 KB
278 KB JPG
>>216798861
Are you prompting for stuff like 70mm panavision, matte painting background, etc.?

>70mm panavision fantasy movie still in the style of jim henson studios, a trenchcoat detective holds a lit match to his mouth cigarette, femme fatale in cocktail dress raises large revolver further down the rainy midnight alleyway, flickering bar sign in background with text "SKILL ISSUE", realistic skin texture, smoke, screengrab, film grain
>>
ok so why? why are the humanoids borged they don't look very happy, what is the deal?
>>
File: 1766411303141791.jpg (87 KB, 881x823)
87 KB
87 KB JPG
Am I crazy or does this image have an early AI vibe?
>>
>>216800622
are you telling me its not
>>
>>216800622
No looks like the guy who did the costume designs for the dark knight worked on this film
>>
>>216799736
tech noir
>>
>>216800622
looks like shaq's steel
>>
>>216797753
mandy is the closest youll ever get to chrome lords

https://www.youtube.com/watch?v=SEjj77XvghI
>>
>>216797793
they are giving us downgraded versions, elon's grok used to be make porn from any picture but now you can't even pay for that
>>
File: ComfyUI_03609.jpg (3.81 MB, 2160x1536)
3.81 MB
3.81 MB JPG
>>216797793
Everyone started leaving out or obfuscating artists/styles and they got worse at mixing and matching looks.
>>
>>216797793
>>216798355
is the closest answer
AI/chatbots etc have a sort of internal slider which allows wilder/.freer 'thought' or a 'safer'/stricter mode.
it used to be a lot wilder, which allows for more creativity. over time they've made it safer and safer, tamer and tamer, and less able to make or say something that might put them in any kind of perceivable legal trouble whatsoever - it's also made the chatbots and image gens act like neurotic jew lawyers which is why it all looks bland and pointless now, and why the chatbots throw a fit if you accidentally say the wrong keyword.
the wild west is over, pretty quickly too.
>>
>>216797778
>told I have to hate AI because it will put people out of their jobs
>the same people who want me murdered for not being a troon-worshiping commiedrone
>the same people who don't care about Americans losing jobs to illegals
>the same people who said "learn to code" when coal miners were laid off
>>
>>216798579
>>216797778
>>216797753
AI can't make that anymore because it keeps referencing itself instead of taking stills from the films that inspired the prompts in the first place.
>>
>>216803567
AI can detect AI images with 99.9% accuracy
>>
>>216803567
AI version of inbreeding/cannibalism
>>
>>216797793
Because they look real
>>
File: 1680829833809594.jpg (1.48 MB, 1792x4096)
1.48 MB
1.48 MB JPG
Christopher Nolan's Evangelion
>>
>>216805299
Of all people, I know Wes Anderson is an Eva fan, and I think he could do something really interesting with it if paired with the right screenwriter.
>>
>>216803086
You can't make images like this btw
>>
File: ComfyUI_03754.jpg (2.81 MB, 1536x2160)
2.81 MB
2.81 MB JPG
>>216806300
Technically, I didn't... I loaded >>216797778 into a local LLM, had it describe the style/aesthetic, loaded in another image (a pic of Sabrina Carpenter before - this pic here was Etna), had the LLM reimagine the second image with the style of the first and then finally distilled the output into a usable prompt.

Gotta say, I've been digging this old slop aesthetic it puts out.



[Advertise on 4chan]

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.