|
Rutibex posted:human DNA is only 75 million lines of code, 691 megabytes, and most of that isnt brain. not even close to a AAA game. Dall-E is way bigger so it should be even more sentient than a human
|
#
?
Jun 13, 2022 04:39
|
|
|
|
| # ? May 30, 2024 14:53 |
|
|
Rutibex posted:human DNA is only 75 million lines of code, 691 megabytes, and most of that isnt brain. not even close to a AAA game. Dall-E is way bigger so it should be even more sentient than a human Yeah but the code is a procedural generator my dude
|
|
#
?
Jun 13, 2022 05:33
|
|
|
|
|
#
?
Jun 13, 2022 05:41
|
|
|
the nightmare faces are intentional right? if they unfucked it the faces would look normal?
|
|
#
?
Jun 13, 2022 06:30
|
|
|
  "Columbo arresting Jessica Fletcher for her crimes"   "cybernetic dad telling a joke to bored kids"  "portrait of a "Dana Scully" in the style of Basquiat and hannah höch"  "Arrakis, a fusion of art deco and star wars architecture, artstation, volumetric lighting, by james clyne::4 ralph mcquarrie::1, realistic beautiful high detail render" (swiped this prompt from someone else and added 'Arrakis')  "futuristic 70s Soviet Union cigarette ad with pretty woman smoking and cool catch phrase"  "futuristic, 70s West German cigarette ad with pretty woman smoking and cool catch phrase"   "stylized poster, general Sherman, vaporwave"  "stylized poster of futuristic 80s Soviet Union soft drink" "designed by Syd Mead" series:  robot lawyer  cellphone  energy drink  computer mainframe  laser rifle  hightop sneaker  cybernetic dog  cybernetic dad
|
|
#
?
Jun 13, 2022 07:23
|
|
|
Dall-e doesn't have a clue what to do with simplified cartoons but LMAO does it produce some great expressions.  Also I'm not sure why but this result made me laugh for like five solid minutes  Finally have some nightmare fuel, free of charge.  
|
|
#
?
Jun 13, 2022 07:40
|
|
|
KakerMix posted:"designed by Syd Mead" series:
|
|
#
?
Jun 13, 2022 08:23
|
|
|
Rutibex posted:the nightmare faces are intentional right? if they unfucked it the faces would look normal? Afaik Dall-E mini doesn't do any face distortion, that's just the best it can do
|
|
#
?
Jun 13, 2022 09:51
|
|
|
those of you who got approved, how long did the Dall-E waitlist take?
|
|
#
?
Jun 13, 2022 11:05
|
|
|
Cybernetic dog is a good dog.
|
|
#
?
Jun 13, 2022 14:12
|
|
|
Rutibex posted:human DNA is only 75 million lines of code, 691 megabytes, and most of that isnt brain. not even close to a AAA game. Dall-E is way bigger so it should be even more sentient than a human those lines of dna code are for creating the hardware on which human consciousness runs the software, that is what we don’t understand yet for HUU-MONS
|
|
#
?
Jun 13, 2022 14:24
|
|
|
Don't we mostly each have to write our own software? Having to reinvent the wheel each time is why it takes us so very long to reach useful adulthood. At 1 year old we're still total goobers compared to everything else.
|
|
#
?
Jun 13, 2022 14:59
|
|
|
imagine being a massive accumulation of cells that are all doing their own thing, but are dependant on each other, then at a grand distance from that action there's a voice that seems to think they know what's going on inside but it's the most surface level version of an understanding. that's human consciousness.
|
|
#
?
Jun 13, 2022 15:03
|
|
|
LifeSunDeath posted:imagine being a massive accumulation of cells that are all doing their own thing, but are dependant on each other, then at a grand distance from that action there's a voice that seems to think they know what's going on inside but it's the most surface level version of an understanding. that's human consciousness. Each and every individual cell acts according to their structure and the laws that govern them. They aren't "aware" but like if they were I'm sure they wouldn't be aware of the superstructure organism that they're apart of. I wonder if it's possible if there was a greater super-organism we were apart of. We'd never know it and it would operate on scales far outside a single person's view hell yeah i should do some drugs
|
|
#
?
Jun 13, 2022 18:07
|
|
|
The Fattest PI posted:Each and every individual cell acts according to their structure and the laws that govern them. They aren't "aware" but like if they were I'm sure they wouldn't be aware of the superstructure organism that they're apart of. I wonder if it's possible if there was a greater super-organism we were apart of. it's called a corporation
|
|
#
?
Jun 13, 2022 18:21
|
|
|
Dall-E has no idea what Wayne Campbell looks like but loving loves Garth for some reason. wayne and garth from wayne’s world at peenemunde 1944 helping the german war effort and v2 rocket program  wayne and garth from wayne’s world in the cockpit of a boeing b52 stratofortress over the heart of soviet russia
|
|
#
?
Jun 14, 2022 00:44
|
|
|
it has some idea who wayne is, but clearly prefers garth 12 garths 5 waynes lol
|
|
#
?
Jun 14, 2022 00:58
|
|
|
Handen posted:Dall-E has no idea what Wayne Campbell looks like but loving loves Garth for some reason. wayne's world war
|
|
#
?
Jun 14, 2022 02:23
|
|
|
KakerMix posted:
how come yours are better
|
|
#
?
Jun 14, 2022 02:26
|
|
|
Dang It Bhabhi! posted:how come yours are better There are three major AI art things right now. Dall-E mini, which is the one you see all over social media and in this thread with the hosed up faces. Then there is Dall-E 2, the one with the news articles and bears doing trigonometry and all that. If you see someone saying 'Dall-E' and it looks choice, it's the invite-only Dall-E 2. I'm using the third wheel sorta one, MidJourney, that is also invite-only but not as strong or capable as Dall-E 2. It's extremely strong with mixing ideas, doing portraits, and creating whole new art in the styles of whatever you ask it to, like Syd Mead. Not so much with doing goofy context stuff like even Dall-E mini is decent at.
|
|
#
?
Jun 14, 2022 02:34
|
|
|
Dang It Bhabhi! posted:how come yours are better Dalle2 > Dalle mini
|
|
#
?
Jun 14, 2022 02:35
|
|
|
What kind of prompt babble (trending on artstation, etc) are you adding for midjourney?
|
|
#
?
Jun 14, 2022 02:36
|
|
|
KakerMix posted:There are three major AI art things right now. There's also Disco Diffusion which you can run in a Google colab at that link. I haven't played with that one myself but I think it's popped up a few times in here. It can create some pretty great results, not sure what it's limitations are or what it's best at but here are some examples I grabbed from an article: "A beautiful wilderness in which a lonely monolith glowing, by Asher Brown Durand"  "A scenic view of the pleasure dome of Kubla Khan in the tropical jungle of Xanadu, by Asher Brown Durand"  "A close-up view of a man meeting the devil, by Caravaggio"  "A lady in a long white dress lying amongst many large snakes, tropical jungle, by Prateep Kochabua"  I don't think I've seen any examples of this one doing pop culture stuff.
|
|
#
?
Jun 14, 2022 02:56
|
|
|
TIP posted:There's also Disco Diffusion which you can run in a Google colab at that link.  the artist is dying
|
|
#
?
Jun 14, 2022 03:20
|
|
|
Alan Smithee posted:
it just means the new art will be finely crafted posts to put into the machine
|
|
#
?
Jun 14, 2022 03:36
|
|
|
What if Jethro Tull was formed in the 80s, when saxophone solos ruled the land? 
|
|
#
?
Jun 14, 2022 03:39
|
|
|
what if jethro tull started in the ??s when sousaphones were all the rage
|
|
#
?
Jun 14, 2022 03:46
|
|
|
 
fuctifino fucked around with this message at 04:24 on Jun 14, 2022 |
|
#
?
Jun 14, 2022 03:57
|
|
|
|
|
#
?
Jun 14, 2022 04:05
|
|
|
Tunicate posted:What kind of prompt babble (trending on artstation, etc) are you adding for midjourney? If you're in the discord (discord being how you interact with MidJourney at all) I'll just see what other people are generating in the various channels and swipe their prompts. It's a nice way to see what the AI does with certain prompts. "strong" artists like Syd Mead or Zdzisław Beksiński are so strong that you can get away with saying "thing by Zdzisław Beksiński" and it will kick out stellar results every time. You can do the same thing with anime or 'rendered in octane' to get the AI to attempt to output something in a realistic style. There is a ton of weighting you can do with the options and things to really tweak what the AI will spit out. The other part is I'm cherry picking things, Syd Mead stuff is usually good right from the start, but most things require perhaps 3-4 passes to get something that works, then maybe 3-4 more passes on that specific out put. For every good picture you see there might be 20~ others you don't see that didn't look right. TIP posted:There's also Disco Diffusion which you can run in a Google colab at that link. Yeah this one as well, I messed with it briefly before I got my invite. Let's see what MidJourney does with these exact same prompts. I'll run it till I get something presentable and will tell how many revisions it took to get there. quote:"A beautiful wilderness in which a lonely monolith glowing, by Asher Brown Durand"  First output out of the four you get at once. Looking at Asher Brown Durand's art I'd say that Midjourney gets the artistic style better, though both of the AIs "get" the artist since they both output essentially the same composition. Here is the first output, this one was #1:  quote:"A scenic view of the pleasure dome of Kubla Khan in the tropical jungle of Xanadu, by Asher Brown Durand"  Similar situation here, but with the dome itself being from a poem so there are different interpretations of it. Picked #4 out of the first output since I feel it matches closer to the one done by Disco Diffusion. Here is the first output for this one:  quote:"A close-up view of a man meeting the devil, by Caravaggio"  Again MidJourney seems to get really close to the artist's style. Took a bit to get this prompt to spit out something I'd be happy with. Here are the first two:   Poor results, imho. I added "face to face" to the prompt and got this:  and I went with #4 which is the upsized one you see above. quote:"A lady in a long white dress lying amongst many large snakes, tropical jungle, by Prateep Kochabua" Context like this is where MidJourney just isn't very good and gets showed up by even Dall-E mini. The AI INSISTS on making the woman in the dress a part of the snakes and the snakes a part of the woman. Everything looked like this, even when I changed the prompt around trying to get it to generate something decent.  Changed the prompt to "woman in white dress sleeping on a bed of snakes, tropical jungle, by Prateep Kochabua" and got better results, but still bad:  11 passes with the original and variations of the original prompt with nothing to show for it. Changed the prompt to "woman with snake, jungle, by Prateep Kochabua" and got better results, but not what Disco was able to pull off. 7 prompts with a couple dead-ends. MidJourney still insisted on blending her arm with the snake, just completely unable to keep it all separated. Like this:  The one from Disco isn't 100% correct, but those snakes are not the woman, nor is she the snakes. I gave up.
|
|
#
?
Jun 14, 2022 04:32
|
|
|
 from pixray: 
|
|
#
?
Jun 14, 2022 04:38
|
|
|
Alan Smithee posted:
Nah the artist is just becoming "the cool ideas guy" rather than the highly skilled and experienced technical artist who spent decades honing their craft. Specifically, the cool ideas guy that knows the exact right way to talk to the AI to get the best outcomes especially by referencing styles and artistic concepts, bearing in mind that most of the public will just mash in "boobs big naturals large" to get something that's "good enough". You still need to know how to frame a shot to understand why one is more artistically appealing than the other (out of the 100 options the AI gave you.) Rutibex posted:it just means the new art will be finely crafted posts to put into the machine I wish I'd scrolled down before posting this because yeah this is much more concise.
|
|
#
?
Jun 14, 2022 04:46
|
|
|
KakerMix posted:The one from Disco isn't 100% correct, but those snakes are not the woman, nor is she the snakes. I gave up. if you look up the artist the one from disco diffusion is nothing like their style, but combining the snakes and the ladies is very much a thing the artist would do:  for a third point of comparison here's the results from dall-e 2 for those prompts: "A lady in a long white dress lying amongst many large snakes, tropical jungle, by Prateep Kochabua"  "A close-up view of a man meeting the devil, by Caravaggio"  "A scenic view of the pleasure dome of Kubla Khan in the tropical jungle of Xanadu, by Asher Brown Durand"  "A beautiful wilderness in which a lonely monolith glowing, by Asher Brown Durand"  it seems like dall-e 2 is unfamiliar with asher brown durand, the other results it produced for this were just fairly generic landscapes that featured peaks or trees but nothing like a monolith if you want to see more dall-e 2 and disco diffusion comparisons this is the article I pulled from: https://medium.com/@nin_artificial/dall-e-2-vs-disco-diffusion-c6de6bfbacf9 TIP fucked around with this message at 05:22 on Jun 14, 2022 |
|
#
?
Jun 14, 2022 04:54
|
|
|
 seems like it ran out of paper
|
|
#
?
Jun 14, 2022 05:05
|
|
|
TIP posted:if you look up the artist the one from disco diffusion is nothing like their style, but combining the snakes and the ladies is very much a thing the artist would do: Thanks for the link, and yeah it's clear that Dall-E 2 gets the context so much better and seems to easily avoid the 'multiple eyes and what the hell is this oh well' shapes that these other systems generate, especially with the man and devil meeting, and how the woman there is sitting with snakes, not IS snakes. I can't easily get MidJourney to generate cars for instance, since it doesn't understand where the wheels all go in relation to the others. Haven't seen Dall-E 2 do with those prompts, perhaps it would struggle as well. Also seems pretty clear that MidJourney isn't super familiar with Prateep Kochabua, casting even more doubts that the AI somehow knew more than what we were asking it. MidJourney does know Durand, since it can get a dude that's been dead since the 1800s to paint stuff like Bladerunner's skyline, street view, future Tokyo or a brutalist parking garage, or a water treatment plant:     
KakerMix fucked around with this message at 05:25 on Jun 14, 2022 |
|
#
?
Jun 14, 2022 05:18
|
|
|
https://twitter.com/Dalle2Pics/status/1536424012053680129?s=20&t=8I7OPJiYQMpJHMweMDcUgQ whole thread is amazing and couple of my own  
Tiberius Christ fucked around with this message at 07:36 on Jun 14, 2022 |
|
#
?
Jun 14, 2022 07:29
|
|
|
Tiberius Christ posted:https://twitter.com/Dalle2Pics/status/1536424012053680129?s=20&t=8I7OPJiYQMpJHMweMDcUgQ tried "A still of Homer Simpson in Pirates of the Caribbean” on midjourney  front facing simpsons are cursed
|
|
#
?
Jun 14, 2022 07:44
|
|
|
Tunicate posted:front facing simpsons are cursed  "a still of Homer Simpson in the Blair Witch Project" from that Homer twitter thread
|
|
#
?
Jun 14, 2022 07:50
|
|
|
Tunicate posted:tried "A still of Homer Simpson in Pirates of the Caribbean” on midjourney homer has nostril boobs because of course he does
|
|
#
?
Jun 14, 2022 10:27
|
|
|
|
| # ? May 30, 2024 14:53 |
|
|
Alan Smithee posted:
|
|
#
?
Jun 14, 2022 13:56
|
|