Is AI the future of Indie RPGs?

[Bohrain]

Comics:
https://www.artexmachina.org/magazine/
https://www.evilcorpgame.com/blog/2022/9/29/arc-a-tv-show-based-on-the-evil-corp-boardgame
https://www.reddit.com/r/StableDiff...ew_pages_from_my_midjourney_produced_printed/
https://www.reddit.com/r/StableDiffusion/comments/yjqxhw/the_lesson_a_free_comic_download_made_with/
https://www.reddit.com/r/StableDiffusion/comments/y6dhl6/consistent_comic_character_test_with_sd/
https://www.reddit.com/r/StableDiffusion/comments/x4cqsp/working_on_a_comic_book_inspired_by_2000ad/

Examples of various professional Artists already using it as a Tool in their workflow:
https://www.reddit.com/r/StableDiff...w_to_make_money_as_an_artist_with_a_personal/
https://www.reddit.com/r/StableDiffusion/comments/ybuw1h/thoughts_of_sd_from_an_artist_in_the/
https://www.reddit.com/r/StableDiff...rt_is_popular_and_makes_money_confessions_of/
https://www.reddit.com/r/StableDiff...a_3d_artist_and_im_using_stable_diffusion_to/
https://www.reddit.com/r/StableDiff...thesis_of_a_professional_illustrator_in_love/

Various Proompting Magazines and Guides:
https://www.promptia.art/
https://openart.ai/promptbook
https://docs.qq.com/doc/DWHl3am5Zb05QbGVs

For games we have a respective thread in General Gaming with various examples and where people, including many from the Codex, have already posted AI-generated art they're going to use in their games, and I'm pretty sure people here already know that.

You can flail around and be butthurt all you like, you're not going to stop technological progress.

Most of those seem like small novelty things though. Especially the comic book ones, none of those examples had any dynamic scenes. I don't really doubt that AI won't be useful in the comic book art pipeline, since lot of manga artists use edited or traced photographs and this is something that could easily be replaced.
The artist section didn't really have much people elaborating what they are doing with the AI exactly. One guys does make 250$ a month from pixiv doing AI outputs and editing them which is a nice but nothing groundbreaking. The 3D artist using them for texture generation is interesting, but I don't really know enough of the subject whether it's much different from procedural texture generation methods that already existed.

 

[Maxwell's Demon]

When did the general conversation about AI in games suddenly become about Stable Diffusion alone?

Oh right, when I posted proof that you don't know what you're talking about and you had to move the goalposts.

No, you claimed that the video was outdated.

You implied it couldn't be used for indie games.

being used in the entertainment industry.

sure buddy

Comics:
https://www.artexmachina.org/magazine/
https://www.evilcorpgame.com/blog/2022/9/29/arc-a-tv-show-based-on-the-evil-corp-boardgame
https://www.reddit.com/r/StableDiff...ew_pages_from_my_midjourney_produced_printed/
https://www.reddit.com/r/StableDiffusion/comments/yjqxhw/the_lesson_a_free_comic_download_made_with/
https://www.reddit.com/r/StableDiffusion/comments/y6dhl6/consistent_comic_character_test_with_sd/
https://www.reddit.com/r/StableDiffusion/comments/x4cqsp/working_on_a_comic_book_inspired_by_2000ad/

Glad to see that gary's mod is making a comeback.

That boat has sailed. You haven't been keeping track of shit for nearly five years.

So upscaling is now defined as "artpiece" ? Whatever.

Oh, so now it has to be used in the entertainment industry and featured as an artpiece too?

 

[The_Sloth_Sleeps]

Comics:
https://www.artexmachina.org/magazine/
https://www.evilcorpgame.com/blog/2022/9/29/arc-a-tv-show-based-on-the-evil-corp-boardgame
https://www.reddit.com/r/StableDiff...ew_pages_from_my_midjourney_produced_printed/
https://www.reddit.com/r/StableDiffusion/comments/yjqxhw/the_lesson_a_free_comic_download_made_with/
https://www.reddit.com/r/StableDiffusion/comments/y6dhl6/consistent_comic_character_test_with_sd/
https://www.reddit.com/r/StableDiffusion/comments/x4cqsp/working_on_a_comic_book_inspired_by_2000ad/

Examples of various professional Artists already using it as a Tool in their workflow:
https://www.reddit.com/r/StableDiff...w_to_make_money_as_an_artist_with_a_personal/
https://www.reddit.com/r/StableDiffusion/comments/ybuw1h/thoughts_of_sd_from_an_artist_in_the/
https://www.reddit.com/r/StableDiff...rt_is_popular_and_makes_money_confessions_of/
https://www.reddit.com/r/StableDiff...a_3d_artist_and_im_using_stable_diffusion_to/
https://www.reddit.com/r/StableDiff...thesis_of_a_professional_illustrator_in_love/

Various Proompting Magazines and Guides:
https://www.promptia.art/
https://openart.ai/promptbook
https://docs.qq.com/doc/DWHl3am5Zb05QbGVs

For games we have a respective thread in General Gaming with various examples and where people, including many from the Codex, have already posted AI-generated art they're going to use in their games, and I'm pretty sure people here already know that.

You can flail around and be butthurt all you like, you're not going to stop technological progress.

I had a look at the first one, the comic. I enjoy comics. Ok being fair....its pretty bad...2/10 standard here.
I'd say you'd get better (and maybe quicker) results using alternate automated techniques, such as rigs, etc.
You could also do a google image search, and trace over since its virtually all static faces with different zooms.
Not seeing any advantage there. It has a long way to go to even compete with a terrible comic artist.

 

[The_Sloth_Sleeps]

I think most people will look at this and think its shit - is my assessment.
Comic book fans even more so. Its going to get a bad reputation - fast.
Will there be a place for it? Yes probably, but as I say before, its going to be a niche, more so in the shovelware markets.

 

[Non-Edgy Gamer]

One guys does make 250$ a month from pixiv doing AI outputs and editing them which is a nice but nothing groundbreaking.

When a project takes you 30 minutes instead of 8 hours, it's pretty groundbreaking.

It makes non-artists into artists and shortens the workflow of existing artists considerably.

Most of those seem like small novelty things though.

Most of these people don't have jobs working in these industries. You don't suddenly become a professional comics artist working for a company or earning commissions just because you can draw. And if you're already in that position, you don't change your actual workflow as soon as a new technology comes out. Non-artists do, however, have a much shorter path to creating products, which is why these people can do so on a whim.

As users who have the time and inclination adopt AI for projects, we will see even more. Especially now that AI tech has moved further from the initial releases of SD and MJ.

I expect whiners like Sarathiour and The_Sloth_Sleeps to seethe even more as a deluge of "shovelware" games (i.e., games that actually exist and aren't a lazy pipedream by a guy literally named "sloth") start being churned out by independent creators who didn't have the money to hire artists for their projects before.

 

[Non-Edgy Gamer]

I think most people will look at this and think its shit - is my assessment.

Reminder that this is the art that Sloth has posted so far:

 

[Dexter]

This dumb faggot will still seethe and throw out his empty platitudes even when we have fully Animated movies and AAA games produced using AI tools in 5-10 years instead of Indie games, free Webcomics and Commercial Mangos.

Reminder that it's barely over 2 months since the Stable Diffusion Open Source release.

 

[Non-Edgy Gamer]

Guy designs a mock visual novel game (including interface) using AI.

 

[Dexter]

Also maybe of interest for Pixel artfags:

https://codepen.io/tmmvb/pen/ExRPOLa

 

[Non-Edgy Gamer]

2/10

still can't generate tilesets for me.

check out this sword I drew:

 

[The_Sloth_Sleeps]

This dumb faggot will still seethe and throw out his empty platitudes even when we have fully Animated movies and AAA games produced using AI tools in 5-10 years instead of Indie games, free Webcomics and Commercial Mangos.

Reminder that it's barely over 2 months since the Stable Diffusion Open Source release.

Again projecting into the distant future, assuming linear improvements in the future based off what you perceive as past performance.
We have a pretty good idea of the state of the art of automated image generation - its fair to say its a far cry from being what you are imagining it to be.
The comic strip example serves as a reasonable demonstration of the limitations.
-Static portraits, with manual lighting touchups to add variation. No expressions or poses.
-Extremely poor scene composition. Probably very difficult to compose a scene with a prompt.
-Unrecognizable characters when pose is even slightly changed, some cheap manual tricks used to add variation.
-Stylistic inconsistency between outputs.
-One mess of a comic.

The kicker is, all this could be achieved using alternate techniques. I can imagine that the effort involved would be working around the above limitations. No such problems would need to be tackled using other techniques, so whilst you may spend less time actually generating an image, working the result into a finished product may even be less effort, and you'd get a better end product.

 

[Non-Edgy Gamer]

The comic strip example serves as a reasonable demonstration of the limitations.

Which one? Where is the one you posted even from? When was it made? How?

If it's just some dude changing the prompt using the same seed over and over like it looks like, sure. But that's something people knew wouldn't work before SD was even released.

-Unrecognizable characters when pose is even slightly changed, some cheap manual tricks used to add variation.

This is what VAEs, Textual Inversion and Dreambooth attempt to solve. Does that comic use any of them? Doubtful, since they're all just a slightly altered pose.

-Stylistic inconsistency between outputs.

In this case, I suspect that's the fault of the prompter, not the technology. And as I said, VAEs are meant to solve this.

This is like taking a single DeviantArt drawing and saying that it's all that Photoshop is capable of, and saying that the technology will never work.

I look forward to the next cherrypicked rant about AI you'll make instead of working on your vaporware game though.

 

[The_Sloth_Sleeps]

The comic strip example serves as a reasonable demonstration of the limitations.

Which one? Where is the one you posted even from? When was it made? How?

If it's just some dude changing the prompt using the same seed over and over like it looks like, sure. But that's something people knew wouldn't work before SD was even released.

-Unrecognizable characters when pose is even slightly changed, some cheap manual tricks used to add variation.

This is what VAEs, Textual Inversion and Dreambooth attempt to solve. Does that comic use any of them? Doubtful, since they're all just a slightly altered pose.

-Stylistic inconsistency between outputs.

In this case, I suspect that's the fault of the prompter, not the technology. And as I said, VAEs are meant to solve this.

This is like taking a single DeviantArt drawing and saying that it's all that Photoshop is capable of, and saying that the technology will never work.

I look forward to the next cherrypicked rant about AI you'll make instead of working on your vaporware game though.

Well, this was the first example that our good man Dexter posted. I evaluated it.
Please, feel free to post your example of AI art I will give my assessment.

 

[Non-Edgy Gamer]

Well, this was the first example that our good man Dexter posted. I evaluated it.

This? https://www.artexmachina.org/magazine/

Where on here specifically? I've looked and can't find it.

Please, feel free to post your example of AI art I will give my assessment.

I'm sure you would. Complaining is even easier than prompting, after all.

 

[Sarathiour]

The sheer amount of dishonesty

 

[Grauken]

 

[rusty_shackleford]

Again projecting into the distant future

distant future of 2024

 

[The_Sloth_Sleeps]

Again projecting into the distant future

distant future of 2024

The distant future of 2016 when humans no longer need to drive.

 

[Zed Duke of Banville]

 

[The_Sloth_Sleeps]

Microsoft’s GitHub Copilot sued over “software piracy on an unprecedented scale”​

The lawsuit stated that this is the first class-action case in the US challenging the training and output of AI systems​

by: Zach Marzouk
4 Nov 2022

Microsoft’s GitHub Copilot is being sued in a class action lawsuit that claims the artificial intelligence product is committing software piracy on an unprecedented scale.

https://www.itpro.co.uk/software/36...-over-software-piracy-on-unprecendented-scale

 

[The_Sloth_Sleeps]

The lawsuits were bound to be incoming sooner or later. Some pretty silly folks here were naive enough to think there's no legal implications to repackaging data.

Anyway. It will be interesting to follow the case for the outcome. I expect lawsuits incoming for all this other image generation stuff as well.

 

[rusty_shackleford]

Microsoft’s GitHub Copilot sued over “software piracy on an unprecedented scale”​

The lawsuit stated that this is the first class-action case in the US challenging the training and output of AI systems​

by: Zach Marzouk
4 Nov 2022

Microsoft’s GitHub Copilot is being sued in a class action lawsuit that claims the artificial intelligence product is committing software piracy on an unprecedented scale.

https://www.itpro.co.uk/software/36...-over-software-piracy-on-unprecendented-scale

"first"
lol

this was settled years ago
https://en.wikipedia.org/wiki/Authors_Guild,_Inc._v._Google,_Inc.
https://towardsdatascience.com/the-...ata-science-and-machine-learning-44cfc1c1bcaf

 

[Sarathiour]

Fair use law are an absolute quagmire, but Microsoft is big enough to force whatever shit they are trying to make, and will settle financially in the worst case scenario. Even in the highly improbable scenario that this is not a nothingburger, it's still going to amount to nothing, as it would be extremely difficult to prove that they used a copyrighted set of data.

Might lead to some real moment when someone will try to copyright the output of his machine learning software, when the output in question was a blatant copy of someone's else work.

 

[The_Sloth_Sleeps]

Is the autocomplete in VS2022 using copilot? The current autocomplete is really weird with bizarre suggestions which makes me believe there's some trickery going on.

 

[zangomango]

Yeah I browse art on Instagram sometimes and I saw some pretty awesome art and favorited it, then looked and saw it was made by AI, and I do not like that... we always let our technology outpace our discernment about it, that quote in the last episode of BS:G said it better