| |
Sander
Registered: Jan 2002
Posts: 493 |
An update on ‘Pixel art in the C64 demoscene’
Dear fellow sceners,
In response to the comments we received, we’ve made some changes to the document.
Our goal was always to find common ground to maintain the fun and integrity of our hobby.
What this is not:
- It’s not written to limit anyone (only to encourage openness)
- It’s not aimed at specific individuals (it’s a scene wide practice)
Read the document here
We’d really love to hear your thoughts on this update.
Please post them in this thread, be kind and keep it constructive and on-topic please. |
|
... 127 posts hidden. Click here to view all posts....
|
| |
Sander
Registered: Jan 2002
Posts: 493 |
Quoting KrillI may be a bit naive (with an entirely wrong idea about how pixel graphics are made), but isn't this a bit of a false dichotomy?
As in, what about a third scenario, where you use AI to give you some ideas for a certain background object, or composition, or the like?
Or to give you that particular background object, which you'd take and modify (after AI involvement) to suit your needs and integrate it via the rest of your toolchain?
I believe The Sarge sketched the two extremes, as there are many scenario’s in-between.
Like you said, one example could be incorporating elements of AI in your work, in a supporting way.
In the end it boils down to:
- Doing it all yourself: Full creative authorship
- Outsourcing parts of the process to e.g. AI: Less creative authorship (to any extent)
Here’s a graph.
It doesn’t automatically mean the results have no or little creative value. It’s debatable in each case, depending to what extent AI has been applied.
As for your example earlier in this thread
It is creative use of AI images, tied together by a (assumingly, a human made) creative concept.
- Without the concept, how would you value each image seperately? I'd say, rather worthless.
- What if the concept was also AI generated? How would you value the demo and the author? I wouldn't value the 'author'.
- What if it was all kept a secret? Then the author would suggestively claim it all.
(As a side-note; As long as the prompt ‘cool bird’ (example) generates beautiful images, one can not claim artistic ownership over style, composition, etc. AI will give you 99+% of the image for free, so without a great human idea, it is generally just covering up creative poverty). |
| |
Burglar
Registered: Dec 2004
Posts: 1050 |
Quoting SanderHere’s a graph.
Exactly. One could ask where to draw the line in relation to graphics competitions :)
For X'2024 I will make the necessary changes in votox to support transparency. This includes workstages.zip support and a better option to share the creative process in text. All will be shared automatically during voting.
I will not mandate transparency, but rely on proper scener behaviors: respect your fellow artists, please don't wire or use AI :) |
| |
Jetboy
Registered: Jul 2006
Posts: 228 |
Quoting Sander
I believe The Sarge sketched the two extremes, as there are many scenario’s in-between.
Like you said, one example could be incorporating elements of AI in your work, in a supporting way.
In the end it boils down to:
- Doing it all yourself: Full creative authorship
- Outsourcing parts of the process to e.g. AI: Less creative authorship (to any extent)
Here’s a graph.
I can pixell all the pixels by hand, yet image might not be creative at all. Creativity is much more complicated than that. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11148 |
Indeed, and you can be very creative just by "remixing" what others created. "Doing it all yourself" is mostly about craft, not about art. |
| |
Krill
Registered: Apr 2002
Posts: 2854 |
Quoting SanderHere’s a graph. Haha, that ELI5 graph was entirely superfluous. But funny nonetheless. =)
Quoting SanderAs for your example earlier in this thread
It is creative use of AI images, tied together by a (assumingly, a human made) creative concept.
- Without the concept, how would you value each image seperately? I'd say, rather worthless.
- What if the concept was also AI generated? How would you value the demo and the author? I wouldn't value the 'author'.
- What if it was all kept a secret? Then the author would suggestively claim it all. It's part of a graphic novel that was written by a human and painted by a machine.
There was obviously no attempt to hide the fact that the imagery is AI-generated.
Whether parts of that story are machine-written, no idea (i don't think so), but at some point there'd be a supervising human mind directing the machines to produce somewhat coherent content.
It's not a demo at this point, but i consider it a piece of art that undoubtedly sprang from some human creativity (and i don't mean the countless unnamed artists who made the original content to train the AI).
Quoting SanderAs a side-note; As long as the prompt ‘cool bird’ (example) generates beautiful images, one can not claim artistic ownership over style, composition, etc. AI will give you 99+% of the image for free, so without a great human idea, it is generally just covering up creative poverty. Obviously. |
| |
4gentE
Registered: Mar 2021
Posts: 119 |
Quote:I can pixell all the pixels by hand, yet image might not be creative at all. Of course. And how creative is your work is for the audience to decide. It’s only fair to give them full disclosure.
Quote: It's part of a graphic novel that was written by a human and painted by a machine. The “graphic” part of that graphic novel is horrible. I hope we can agree on that. It also fails as a proof of concept. Because it’s superfluous: (1) of course prompt-to-img AI can do this, no need for proving it, that’s its main purpose and (2) of course images can be converted to HAM. So which concept it proves? |
| |
Krill
Registered: Apr 2002
Posts: 2854 |
Quoting 4gentEThe “graphic” part of that graphic novel is horrible. I hope we can agree on that. It also fails as a proof of concept. Because it’s superfluous: (1) of course prompt-to-img AI can do this, no need for proving it, that’s its main purpose and (2) of course images can be converted to HAM. So which concept it proves? Thank you for giving some explanation to your original dismissive post #12. Let's agree to disagree, shall we. |
| |
4gentE
Registered: Mar 2021
Posts: 119 |
Yeah, sorry for that short and pretty dumb first post (#12) on the subject. |
| |
Jetboy
Registered: Jul 2006
Posts: 228 |
Quoting 4gentESo which concept it proves?
Does art need to prove anything? |
| |
Rex
Registered: Sep 2011
Posts: 14 |
With the right spirit and wording you can inspire others to join your efforts. Here is an example that I like a lot https://handmade.network/manifesto
The new version of the document does move somewhat in this direction. However some of the wording still seems more judgemental than inspiring. For instance the title and introduction indicates that the document defines good and bad art for the entire C64 scene.
To me, this smells more like gate keeping than inspiration. When I read this, I want to rebel against the arbiters of good taste - even if they may have valid points.
If the document was renamed to "Human made Pixel Art Manifesto for the C64 Demoscene" and reworded to inspire others to join, I believe you might have a better shot at starting a movement.
I am a coder, and not an artist. So take my opinions on art with a grain of salt.
“The old guard in any society resents new methods, for old guards wear the decorations and medals won by waging battle in the accepted manner.”
― Martin Luther King Jr., Why We Can't Wait |
| Previous - 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 - Next |
