Trike

Does Sox need a pet cat? We have several varieties currently in stock at Trike’s Robo Emporium in Skyway City.

I’m going to have to try this. I’ve accidentally received turnarounds but never asked for them. Last night I was goofing around with keywords like “orthographic view with annotated schematics” and it included “art station style” in the pictures. Bit of a surprise, so I leaned into it. One of my robot characters was created by Percival Perry, so I tried asking for “Percy Perry PTY 1885” and some were actually pretty good.

It’s weird how it has trouble with some characters. I’ve made some female drow warriors in Bing, so it doesn’t have a problem with it. What I *have* discovered is that sometimes the order you ask things makes a difference. For instance “hero flying” returns different results than “flying hero”. No idea why, other than maybe that’s how the training images were tagged. Upload a game screenshot of the toon and we can give describing it a go. Here are some of the ones I’ve made: Prompt: art station style, photo realist, a female drow elf warrior with white hair, pointed ears, and deep black skin tone Prompt: art station style, cel shaded, dynamic pose, full body, a female drow elf archer with white hair, pointed ears, and deep black skin tone

Do you prefer Fantasy or Science Fiction? Tavern or bar?

This was the first thing I saw when I logged in last night. 😂

I recorded how long it takes. This is pretty typical. Added a bit of music for fun. https://www.instagram.com/reel/C28k2oUO2rG/?igsh=azZ4cjlicTNwN3Aw The prompt: “art station style, a steampunk duck in a dungeon, spraying robed cultists with water” The results are AWESOME. Edit: BTW, you can see I have zero tokens. The indicator is at the end of box where you type in the prompt. Having tokens prioritizes your render, I guess, and they seem to refill every day.

Those look great. And I have to compliment you on the name “Draculad”. Brilliant! Wish I’d thought of it. Did you use Bing for these? I’ve found that it has definite blind spots when it comes to some things. I’ve given up trying to recreate many characters because it was just returning totally random stuff that was completely unrelated to what I’d asked for. For instance, see those short aliens in the swamp in the Space Patrol pics? I initially asked for them to look like a platypus. It gave me the most bizarre creatures in response. The most hilarious ones were Kermit the Frog-looking things, and it kept trying to give me alligators — even for the human astronauts! 😂 I eventually asked for purple and teal otters, which it liked better, but still rendered weird stuff. These are all from its idea of “platypus”. 😜

I’ve found it also depends on the day and time. Saturday morning around 10 am Eastern it gave me the message “too busy, try again later”. But then in the middle of the night last night between 1 am and 4 am Sunday I was able to generate a hundreds of those Space Patrol images fairly quickly despite not having any tokens.

I’m given to understand that Adobe’s Photoshop AI plugin Firefly was similarly trained on public domain art in order to avoid legal hassles. I’ve been primarily using Bing’s Image Creator, which often returns things that are suspiciously similar to existing art. I specifically did my Space Patrol series to mimic government art of the 1940s/1950s while using the prompt “1930s pulp” to try and avoid copyright infringement. Not sure I was entirely successful, but I gave it a go.

Join the Space Patrol! See the Galaxy! Meet new friends! Welcome to the Atomic Age of spaceflight! We have all the latest 1954 technology! The farthest frontier has endless wonders and out-of-this-world encounters! Learn technical skills that will stand you in good stead for a lifetime! Multiple career paths available! From Alien Diplomacy all the way to Zoological Exploration! Serve aboard the most technologically advanced spacecraft ever devised! The Space Patrol stands ready to defend the galaxy! Sign up with the Space Patrol! Adventure awaits! See your local recruiter today!

No idea, but it somehow focuses the algorithm to be consistent. Sbloyd used it in a prompt so I tried it. Whenever I leave it out the results really go off the rails. I suspect it’s stealing from the artists on that site. Which is why I’d never use AI-generated art for commercial purposes, but as an entertaining mini-game of “persuade the idiot to do what you want” I’m all for it. As for The Phantom + The Green Hornet, there’s a good name: The Phantom Hornet - the ghost who flies. :)

Sometimes it just works. So random.

Trying The Phantom again and this time it kinda worked. Go figure. The pic with all the accoutrements is really interesting. i wonder what it was pulling from to get that? I never changed anything in the prompt, just kept re-entering it. Prompt: "Art Station style, oil painting style, 1930s pulp magazine, a pulp hero similar to The Phantom except his outfit is green and black, full cowl, blank eyes, green and black pinstripe leggings, black boots"

Kaiju attacks Steel Canyon! Simple prompt: "Art Station style, several heroes blasting a giant plant-like kaiju".

The second one looks like you could replicate it in game almost exactly. That’s clearly the Diablo top.

Night hunters. Another perfect rendition first time out that caused me to keep pushing the button to see more iterations. Ended up with a couple dozen, and amazingly only a couple had Batman or Nightwing clones. i don’t recall the exact prompt, but something like “night city, full moon [later crescent moon], low camera angle, male hero wearing blue and silver tights looking at the camera, standing next to him is a female hero wearing green and gold tights looking at the camera, both standing on building ledge”. Later added “robot owl” and “cyborg owl” which didn’t really add anything. i could show them all, but here’s a decent sample. i made a lot of ‘em. 😛

A nice soothing bath after the battle. This one took forever. It kept returning bizarre versions. The original pics had spot-on Captain Marvel suits, I guess because she’s the only superhero associated with a house cat. So I used various other color combos to avoid that. Prompt: ”late afternoon sun, a freestanding bathtub full of steaming water, green and silver superhero outfit folded neatly on a wooden stool, a golden sword leans against wall, a cat sleeps on the windowsill, a few potted plants nearby”

I tried to do a version of The Phantom too and the results were bizarre. Some were interesting, but worlds away from what I was going for. I don’t know why it has that blank spot. I get a ton of Superman, Batman, Iron Man and Spider-Man clones (see selfie post above), but no Phantom. I also ran into the “no” issue, and trying to modify an existing picture by trying variations of “remove” and “delete” weren’t being parsed. One time it removed a leg. Hilarious, but I was trying to delete a cape.

The Leonardo algorithm I linked to upthread for the Purple Fairy allows you to upload a picture or pictures as reference, then choose how closely you want it to mimic it. Sometimes it works fine but other times it feels random. I tried a couple modified pics the other day and it kept returning zoomed-in pictures where you could only see torsos. Ran out of free tokens trying to get it to work, but it might be something for you to try,

Many of these look like cool book covers. I’ve been going the opposite way, trying for minimalist backgrounds. What prompts did you use for these?

These are really complex. What sort of prompts are you using? Bing is uniformly hilarious in how it scrambles words. I’ve had some weird insertions in mine, too. Of these I particularly like “City of Heroees” - sounds like a town full of rescued folks. Briickstown!

Bing AI prompt: “a group of teenage superheroes taking a team selfie after defeating some villains, wide angle” For some reason a young Kenny Loggins showed up in one. Maybe they rescued him? Modified ask: “a group of teenage superheroes taking a team selfie after defeating some villains, wide angle, outdoors city street, debris and wreckage all around” Second mod: “cel shaded style, a group of teenage superheroes taking a team selfie after defeating some villains, wide angle, outdoors city street, debris and wreckage all around”

This guy, though... oof. With Kung Fu Panda and WoW's Pandaren, I had a hard time convincing the machine to give me something that was "not those". Boom-Boom the Panda was one of my more popular characters back on Live. He won several costume contests despite being incredibly simple. It makes sense that his son Bamboom would follow in dad's footsteps.

Yet another instance where AI got it right the first time. These are the only two images I generated for Steelstriker.

I have variations of every type of fox, but Desert Fox has never really come out to my satisfaction, mostly because I envision her as a Fennec. In this case, the AI is way better.