T w e n t y S i d e d
When you want an image to use for your character portrait, you have two choices:
- Spend years mastering the art of sci-fi / fantasy illustration. Perhaps attend art school. Hone your craft until you can fully realize your character on paper just as you imagine them.
- Use Google Image Search and swipe something that looks roughly like how you want your character to look.
For whatever reason, most players take the lazy way and opt for #2, despite the fact that there is a 90% chance they are going to look like a brooding androgynous goth / punk elf holding the wrong weapon.
Go figure.
– Shamus, Monday Sep 25, 2006
That’s funny, I was just thinking the other day how all those AI art programs have taken over D&D character portrait making. I haven’t personally partaken since I have opted for option #1 and spent the last decade and a half I think I just took 4 psychic damage. The idea that I’m old enough to have been doing anything of substance for a decade and a half. That can’t be right. honing a little-used art ability. However, many of the D&D groups I’m in have been overrun by very customized character portraits which suffer from new problems; having the wrong number of fingers and all looking too human. I’m not really here to argue the many ins-and-outs of morality and computer generated art, but I will say the tech is cool, and of all its uses, portraits for D&D tables is likely pretty harmless.
Edit: The words I uttered mere moments ago to my partner: Man, I gotta go check comments. It makes me nervous when I have a post go up AFTER I’ve gone to sleep.
Yup, nailed it. I set up the whole post and managed to forget to actually upload the comic itself. My bad guys, it should be fixed now.
This weeks French comic can be read here.
Footnotes:
[1] I think I just took 4 psychic damage. The idea that I’m old enough to have been doing anything of substance for a decade and a half. That can’t be right.
Gamers Aren’t Toxic
This is a horrible narrative that undermines the hobby through crass stereotypes. The hobby is vast, gamers come from all walks of life, and you shouldn't judge ANY group by its worst members.
This Scene Breaks a Character
Small changes to the animations can have a huge impact on how the audience interprets a scene.
What is Piracy?
It seems like a simple question, but it turns out everyone has a different idea of right and wrong in the digital world.
How I Plan To Rule This Dumb Industry
Here is how I'd conquer the game-publishing business. (Hint: NOT by copying EA, 2K, Activision, Take-Two, or Ubisoft.)
Seven Springs
The true story of three strange days in 1989, when the last months of my adolescence ran out and the first few sparks of adulthood appeared.
Never bothered with character portraits, personally. Don’t really have the artistic skills for #1, #2 usually doesn’t get close enough for my tastes. Haven’t tried AI art, but i did try Hero Forge; it just felt tedious and boring and I always quit halfway through
Also, the commentary on this one posted, but the comic didn’t
You get used to it over time. Tell you what, the psychic damage is a lot worse if you feel like what you’ve been doing for the last decade or so isn’t of substance…
(Bah, this should have been a seperate comment. Curse You, WordPress)
Seriously, just wait until you tell someone in your group “I’ve been gaming/arting/etc, since before you were born.” That’s when the Old Age Psychic Damage really piles on.
I remember, I think it was Mumbles, saying something about how long ago a band from her childhood or something was and how it made her feel old. Shamus, ever the resident “Old Man” of Spoiler Warning said its going to really hit you when its someone you still think of as new and gave Lady Gaga as the example.
And now, one of Lady Gaga’s early hits “Poker Face” is 15 years old. There are kids in high school who were just being born when that song hit the charts. I was almost 30. Shamus was even older, if such a thing is possible.
As Vernal said the comic itself is missing.
I have to say, NONE of the article being on the front page is a refreshingly new problem.
Only the end of the article on the front page, boss!
Like Vernal I also generally didn’t bother with character art. I have no ability to draw whatsoever and I sometimes don’t even have a perfectly clear visual of my character in my head to be honest. Another player (spontaneously) sketched one of my characters once and it was cool but didn’t look “right”. If I wanted to do a visualisation of a character I’d probably go for something like Hero Forge or some other kind of composite character builder. Toyed a little bit with some AI a while ago when it was way less advanced, it reacted to prompts in a very… approximate way and most things came out as very blurry shadowfigures, neat if I wanted to create dreammonsters, not so neat if I’d want to make an actual character portrait. And thinking of doing it through AI now feels kinda dirty with all the ethical concerns surrounding it…
(and yeah, the page is missing)
Does the comic not show up for anyone else?
There’s a third option, which is to pay someone to do it for you.
Way back in the day, before Google Image search was much of a thing, I used to play an online text-based game where you could “fight” others characters using what we’re effectively automated dierolls, and a pre-FarmVille style stat training where you got “energy” every few hours. The gameplay was sort of lame, but there was a rich message board community, which was the “real” game.
There was no character portrait per se, but the de facto was to have a banner forum sig with your name and portrait. If you had one, that’s what your character looked like.
There were a few people with some graphic skills who’d make banners for pay. Kind of a lucrative business- the going rate was 2 disposable items that you bought with cash and gave in-game energy for a period of time. They went for $5 each and a banner cost 2 of them or more.
Thinking back, this would have been a great fit for AI art. The lower end banners were not all that customized and were kinda samey. And really nice ones were expensive. There was a clear have/have not divide that people took way too seriously.
Yeah, that’s what I was going to say. AI image generation seems like a perfect way to get some interesting “original” character portraits without needing any art skills.
I’ve been going that way, but it’s surprisingly hard to get what you imagine sometimes.
I’ve tended to use picrews in the past for character portraits, when I haven’t been able to commission custom art.
A great benefit of having borderline aphantasia is that I have no concept in my head of what a character I’m building looks like. Once I’ve got their race, background and gear I’ll search for some keywords to get something accurate, choose the first one that doesn’t look awful, and then go back to the character sheet to fill in things like hair colour and distinguishing features. Then in all likelihood I’ll never look at the portrait or the description part of the character sheet again
My group actually has another default they use for making character portraits: HeroForge. They go on, make a minifigure that approximates their character, and then send me a screenshot that we use in our virtual tabletops. You can get quite granular with the character creation — especially considering that it’s free (except for a few premium features).
. . . is it a problem that the first comment I’ve ever submitted on this site looks like an ad? Lol.
I’ve been kinda lazy with coming up with a portrait for the few characters I’ve made and played with, just didn’t seem to important to me. The one time I went out of my way to have one, I never really even thought of using AI to generate a portrait and using google right out the gate just didn’t feel right. I cooked up something approximating the idea I had in my head using the character creation for Lord of the Rings Online and went with that. I wasn’t fully happy with it, but figured it works, since it was more for the benefit of the other players and the DM. For another character I tried to dabble with Fallout 4’s character creator, but gave up after a while and ended up picking another close approximation from what D&D Beyond had to offer.
I used the stock one from the D&D website for Monk and I think modelled my character description around that. And then something similar with my next character, google search. I think I play these things as a choose-your-own-adventure game with combat and adventure mechanics, not a roleplay opportunity. It only clicked recently why (video game) RPGs are so named. I wondered what the words role-playing-game had to do with “game with conversation options and upgrades”.
People actually pick a character idea and then try to pretend that they’re that character?! Instead of just picking something that they like the look of and then doing stuff as they see fit?! No wonder it’s considered a weird hobby! :P
I haven’t played D&D in a long time but I thought I’d try my hand at creating character portraits. For one, I had the AI use a picture of me and told it to turn me into an elvish wizard. So much hair pulling. No matter how many times I insisted my character was supposed to have white hair, it kept it dark. And it sucked at creating a wizardly garb. Plus the pointy elvish ears never looked right.
For another, I didn’t bother using a picture as a starting point. Instead I simply wanted to create a figure that, even to D&D characters, looked like he’d stepped out of some older and even more legendary time (kind of like how in Tolkien’s world, things were more awesome in the past and have slowly been getting less awesome). It created a Conan the Barbarian figure, hugely muscled and with a suitably hardened and weathered face. But it would barely put anything on him. I had to run the image generator several times to finally get it to mostly cover the hero’s body but it still left some areas exposed. I figured close enough.
Personally I think it will be great for internet culture. I saw a picture the other day of Jack and Rose floating in the freezing water, freezing to death themselves as the Titanic sinks in the background, and Jack is taking a selfie of the two of them. I love that someone can have an idea like that now and not have to possess photoshop skills or be an artist themselves to make it real. It’s a force back towards democritizing the internet. I look forward to an era when people who are more logical and practical, like Shamus, can create art and influence culture in the same way that childish and overly emotional artists do.
Hah, unfortunately I’m about to put myself as one of those childish and overly emotional artists you’re worried about. But Id like to toss in my two cents.
I think it’s fair to want to create something with a fun tool or toy like an AI generator. If you’re willing to put up with a similar issue to finding random art online of it not actually looking quite right, it’s just another way to make something fun.
My only real issue, -which I imagine is what has prompted so many artists on the defensive against ai art- is the amount of people selling it. Specifically people who take art directly from artists without permission or payment, claiming that since the AI generated a new image it’s original. The problem being that ai aren’t human, there is no originality. They are always a product of what we give them. just like how ai interviewers have been documented having the same biases as the humans that trained them, such as assuming people are friendlier and smarter because they have a bookshelf or nice office behind them instead of a wall. Scraped ai art being sold is like having someone photoshop your work to resell it (which actually happens, it was apparently worse back when photoshop was newer, but you still see it happen from time to time) except instead of everyone agreeing that the guy shouldnt have done that, everyone wants to make it an argument about art versus inspiration. Ignoring that a robot can’t get inspired in the first place.
It’s genuinely a shame that this is how it’s primarily being used. if I’m honest, Ai could be such a fascinating tool for people to use artistically. The way images break down the closer you look can be incredibly haunting and could make fantastic horror imagery. An artist and someone with experience creating and training datasets working together could make some genuinely interesting products. From an AI breaking down a digital image to generating a surreal reference for physical media. But the amount of people try to use other people’s work for quick cash has driven so many artists and others against the idea of ai at all. It’s probably going to be a while until people feel comfortable enough with the concept we have a chance to see what it can be used for as a medium.
A final note, as interesting as this conversation is I unfortunately do have to do my job as a moderator. Childish and insulting language such name calling, even against a nebulous group of people who just don’t think like you, is unwelcome and unconstructive. This isn’t an argument of the logical versus illogical, it’s a conversation about a very interesting but complex social issue.
I apologize. I was excessive in my description of artistic people.
As for inspiration, the inspiration is coming from the person providing the prompt.
IMO, regarding all these new content generation tools, the entertainment and art industries have always been a bit reactionary about new technology.
I agree that measures should be taken to make sure AI isn’t straight up ripping off art but training an Art AI to draw based on existing images I think is valid as long as the products are sufficiently distinct.
And I think we’re only seeing the beginning. I’m not sure how far down the road it is, but I think someday, AI will be able to generate much more sophisticated content. Movies, television shows. Open World games.
In the short run, AI will be an aid to the existing industry allowing them to generate more content with the guidance of actual creative types. But eventually, I know AI will be able to create TV shows that are plenty satisfying for guys like me.
I don’t care if my entertainment is “true art” that “elevates” or had “important messages”. All I want from entertainment is entertainment and I think eventually AI will be able to supply me what I want without ever having to watch stuff that Hollywood created.