How did you get started drawing / painting?

[Slime_Punk]

I pulled out my tablet and tried to sketch some stuff out and realized I have the drawing talent of a 3-year-old.

So, uhh, how did you learn to make things that actually look pretty plausible? I can use any tips and examples I can get. And I'd also love to hear stories about how you got started and what inspired you along the way!

 

[Slime_Punk]

This is where I'm at, it's atrocious. Send help

(I didn't give these proper renders because they're trash, but you get the idea)

 

[mysterionz]

Even after almost ten and a half years, that still applies to me been drawing ever since second grade myself

 

[Misery]

I've begun truly diving into this only recently. Right now, this is the thing I'm currently working on:

Her name is Amalga! She's from one of my favorite games. This drawing is far from finished, there's a lot to add yet, though Amalga herself is mostly done.

It's taken me a long time to get to the point of being able to make something like that.

The problem I personally ran into is that I had no proper teaching or anything, and just trying to figure it out on my own meant I got stuck for quite awhile.

I could make these:

Those may look nice, but they arent quite what they seem. The one thing I *really* know how to do is to replicate stuff I'm looking at. I simply looked at already existing images of these characters and put them onto paper. It is a slooooowwwww process. Very slow. And not particularly satisfying. However, it was very useful for getting used to the pencils and whatnot. Learning to draw more smooth and less shaky.

I also know how to do whatever the heck these are:

I just refer to these as "shape art". They're very... random. The shapes used and such have no particular meaning, I just mash things together until I get something interesting. They're more satisfying to do than the mimic drawings at least.


The other major thing I know how to do, and the thing I actually specialize in, is brush lettering. It involves making stuff like this:

I have books about this, and practice paper (that's what that was made on, thus the lines). It's complicated and doesnt work like it may look. It looks a bit like cursive... it isnt. Cursive is one constant flowing line, when being written. Brush lettering is actually a series of many individual markings, with the brush leaving the paper between each one. This was an interesting lesson, in that the process of making something often doesnt quite work in the way that it may at first seem (and that's something to keep in mind). But also this involves a very specific type of tool, which is brush pens. They're what I use the most often, and I've got this huge case full of them. If you look closely, you can see that individual strokes vary in width, that's what brush pens are for. Very much designed for lettering (but they can be used for other things too). I think they're also useful though because they encourage more fine control when drawing something, since careful pressure is important (the more pressure, the wider the line) as is very smooth movements.

Seriously I learned a LOT about making art just from that alone.

As for how I got started with all of this, well, that's actually pretty important to me in multiple ways.

Look at this horrible guy:

This character, named the Pumpkin Rabbit, is designed by a guy named Martin Walls. He specifically is THE inspiration for me doing ANY of this. But also, there's this:

Another of his characters, and also where my name Sophie comes from. She's from Martin's series called The Walten Files. Aside from being a favorite series of mine, what truly drew me to Martin's work was the screwball style he uses. I cant think of another artist that does things like he does (seriously, look up his stuff). It's very funky, and his stuff drives home a very important point: I dont have to mimic someone else's style to make appealing things. I need to find my own style, even if that style is kinda weird and wobbly. Martin proves that "weird and wobbly" can seriously work, when done with the obvious enthusiasm that he puts into his stuff.

As it is, Sophie there was the first time I DIDNT just mimic the exacts of what I was looking at (outside of shape art, I mean). Her overall design is his, of course, but I applied *my* style to her, creating a different look than how Martin draws her. The Pumpkin Rabbit, on the other hand, is another mimic drawing (and it's VERY accurate). Also the line "safety in pills, Sophie" is directly from the series she's in. My success in drawing her then led to enough confidence to draw Amalga, which again applies my style (whatever that is exactly) to a character someone else designed.

Martin's work though and his approach to things, while considering also the amazingly positive reception his stuff gets and its popularity, taught me one other very, very important thing: we shouldnt criticize our own work, because we're bloody bad at doing that. Why did his stuff teach me that? Because, again, his style is all sorts of screwy, but instead of looking at what he made and going "this isnt good enough", he just EMBRACES that screwiness and runs with it, and it WORKS. He has a whole Youtube series and everything, which gets like a bazillion views and is popular enough to get a ton of theory videos made about it (yeah, it's one of those super cryptic sorts).

And I think that's the primary lesson here: you may start out thinking your work isnt good... but that doesnt mean it's bad. It means you personally PERCIEVE it as bad. Instead of doing that, just... roll with it. Experiment and improvise and just DO it. It doesnt have to be some perfect painting. What's important is that it is YOURS.

Dont worry about comparing yourself to other artists, and dont do what a lot of people do and just try to mimic the styles that you see others using. I guess that's my other main advice, because that is what's working for me.

Seriously, what you've shown here really DOES have potential. Again, roll with it, dont fight it.

 

[Rodafina]

I am definitely a work in progress… Any artist will always be evolving, I suppose. But here’s a few things that gave me more confidence in my work and started to make things look more like I wanted them to…

– Highlights and shadows. Study the heck out of the highlights and shadows everywhere you see them and try to incorporate them boldly into your work. They can make a huge difference in making things look finished vs. in progress.

– Tracing is not as evil as it seems. Of course, you can’t really take pride in and certainly not credit for traced work, but tracing existing artwork or photos can really help with confidence in your lines and, again, the placement of highlights and shadows.

– Draw like the dickens… Just keep going and going and going. If you don’t like something, often you can just keep going with it until it becomes what you want it to be. Watching how awful the first few stages of peoples’ art looks is revelatory. There is so much pre-planning and erasing that goes into a finished product.

– Draw the same thing over and over again. Almost like the way people describe cooking – if you can get down a few basic drawings to give you confidence and satisfaction, you can then add to these things over time and build your repertoire.

 

[Atrapa Almas]

Your drawings are not trash if you enjoy doing them. Nowadays drawing is like playing chess in the sense that you cant beat an artificial intelligence, so in my humble oppinion its more about the drawing joy, experience and self progresion than about the result.

I started copying animals in a very autistic mode when I was child. So I had a book with a Tiger picture and just tried to do it. I saw my drawing and the picture, spot mistakes and did again. Spotted new mistakes and did again. And again. And again.

My mum told me that it was incredible that I drawed tigers so well strting by the tail. She did not seem to spot that all them were exactly the same Tiger.

Once I got bored of the tiger I started with a pelican. The same pattern. Then other animals. Then comics. At some point I took a drawing course when they showed me how to actually do sketches, anatomy principles and some other stuff.

I drawed at the bus station, at class, at home... So I eventually became better.

If you enjoy drawing, just do it. If you enjoy learning to draw (which is different) take some YouTube free courses. Just dont be too hard on yourself. In 10 or 20 years there wont be drawing proffesionals that can compete with a IA.

Hugs.

 

[Crossbreed]

When I was 12, a friend and I started making up ideas for amateur comics. I would make up origin stories for his & my characters. He was good at drawing men, but his women looked like men with boobs.

When I started drawing, I made all of the freshman mistakes. I eventually incorporated ovals & rudimentary skeletons. Of the two of us, I was the first to discover how the womanly skeleton was subtly different from that of a man.

I eventually bought some figure drawing books and I have a collection of 1:6 scale action dolls that help me to draw more specific faces.

Recommended Reading...

Doubtless, there are many more great anatomy books out there geared toward the artist, but these are the ones I know and can recommend.

studio-126.blogspot.com

I have two galleries here.

Light Corps

www.autismforums.com

Studio 126

www.autismforums.com

 

[GypsyMoth]

I pulled out my tablet and tried to sketch some stuff out and realized I have the drawing talent of a 3-year-old.

So, uhh, how did you learn to make things that actually look pretty plausible? I can use any tips and examples I can get. And I'd also love to hear stories about how you got started and what inspired you along the way!

My mom was an art student. I was, maybe 5, when she took her first painting class. I watched her as she worked on it. It's a painting with my finches on it and a couple of cats in a basket she clipped from a magazine. I still have that painting--it now hangs over my bed. She went on for a BA in fine art, which meant our household revolved around her part-time endeavor for years and years and ...

There's a way of looking at things that some people have naturally and others can learn. When I'm drawing, it's so intense that I have a hard time not seeing things that way. I eventually took a couple of college-level drawing classes and then a watercolor class. My instructor was most impressed that I was one of two students he had ever had who literally sat there for the longest time just looking and observing before I even started drawing. Everyone finished long before I did, but I had one of the nicer first-tries. (I've attached that painting below. It took me a couple of paintings before I learned how to not get that theatric look with the cloth. If you look closely, you can see where I painted over a perspective line.)

BTW, watercolor is a second cousin to drawing. I actually liked painting landscapes in India ink long before discovering watercolor. It's a lot like sketching.

If I were to teach drawing, I'd recommend two supplies: a spiral bound pad of paper suitable for sketching, and a pencil. Everything else they sell is extra. Get a nice, small size for the paper, maybe a 5x8? so that you can carry it with you wherever you go. When you have a free moment--waiting at a train station, waiting in the doctor's office, on break at work or in study hall at school, pull it out and work on your sketching. It's fine to doodle, to do abstract drawings that don't mean anything, and of course, to sketch what's before you.

I'd also recommend going to an art gallery and go to a museum. Start trying to understand what makes art work. Drawing is the most fundamental skill any artist can have, but what's more fundamental than that is intrinsic to who you are, and that's the ability to observe. Learning how to draw helps us learn how to see.

One of the best pieces of advice I've run into concerning sketching is that a sketch is a quick impression of what's before you--it's not meant to be photographically detailed. Try not to spend more than 5-10 minutes sketching out a scene. What this does is it forces you to just see the major shapes and their relationships. You're just looking for an impression.

Once you've made a number of fast sketches of a subject--let's say you've spent some time sketching a bridge over water and now to make a more detailed drawing of it--now you are ready for a more detailed study. Since you can't take the bridge with you, some of your sketches might be a close-up of the stonework the bridge is made out of, or the way the grasses and the meadow flowers overhang the river bank, or maybe the way the reflection of the bridge on the water caught your eye and you have made a separate sketch of that. Take all of your sketches and lay them out before you, and from them, you can now start drawing a more realistic interpretation of the bridge you so enjoyed studying and observing.

Don't worry if you can't get the perspective right. That's an advanced skill. Right now, as you're learning, study your objects intently. How many changes in surface textures can you find? What do the transitions between these surfaces look like? How do light and shadow add depth to what you're looking at? Do the colors change in intensity? Is the shape of the object pleasing to the eye, or does it elicit an emotional response and how might you preserve that feeling in what you draw? These are just samples of the sorts of questions you might ask yourself to help you observe more details more fully.

I've been sharing a fairly intellectual approach up to this point, framed by questions and answers, but that's not how I draw. Think of that as warming up to draw. I really can't explain how I draw. It's rather on a different, nonverbal level of consciousness. As I'm drawing, I tend to remember things I've forgotten. It's very emotional and it's highly visual. I tune everything out around me--not on purpose, it just happens. I have little to no sense of the passage of time and I generally feel happy, even if the memories that arise are not happy memories. I feel focused and calm and if you speak to me, it takes a lot longer for me to be able to understand what in the world you're saying because I'm just preoccupied with something more important at that moment. It also takes me a moment longer to respond.

The most important thing to know about art is that you should enjoy it. It's a pure aesthetic. It's sublime, it communicates across time, and it is supposed to move and transforms its viewer. Don't be afraid to be moved or transformed by the experience of drawing or of admiring art. In doing so, that enjoyment will gradually transform you, and it will transform the way you see the world.

I don't draw anymore. Though it's on my to-do list, I never seem to be able to find the time.

 

[GypsyMoth]

In 10 or 20 years there wont be drawing proffesionals that can compete with a IA.

I disagree. Artificial Intelligence may get the brush strokes right, may render the form and the lighting and the textures photo-realistic. But emotion and meaning come from somewhere within. I am not a big fan of Picasso, but wow, every single drawing of his I've seen is wound so tight with emotion it drips from the page. This has carried over into some of his more geometric work, too. There will never be a Picasso AI. Sure, AI will be duplicate his work, or make work like his work, but it will never be able to interpret that world like Pablo did himself. I think we will, at best, connect only superficially with AI work but art that truly speaks to us will be subtle and filled with the best--and the worst--of humanity.

My favorite painting is an oil done in monochromatic red. There are no lines, there are no shapes, there is no form, there are no pixels or dots or anything concrete with which the mind can latch onto and say definitively, this is a ___. I told its owner, "I love this painting. If it were mine, I'd call it, mother and child in a red room. Turns out, the title was something similar, mother and child rocking in a sunbeam. Or something like that. It was the rendition of the image of what-it-is-like when the sun catches your eye. It was the experience of the moment that the artist had captured that fascinated me so greatly. There is no AI that can ever do. No AI can do that because they are not human; they cannot share in the all-too-human experiences we experience and share among ourselves every day.

 

[Luca]

I have been drawing since I was probably 4 or 5 years old. I'm still not entirely satisfied with my work and on a few occasions some people have been so harsh about it that I sometimes quit for years at a time. I originally went to art school when I was starting college and I dropped out.
However, I have been able to sell some of my pieces (and some of my photography) in local galleries and craft shows, and I was recently invited to join an association for disabled and chronically ill artists, who are talented enough to have their work displayed and sold. So I think we are often our own worst critics but a lot of it is also in the beholder. Not everyone has the same standards or the same idea of what art should or shouldn't look like.

I don't see anything wrong with any of the artwork in this thread and I actually do like it! If you guys aren't satisfied with your work, only practice and learning and improving techniques will solve that.
@Silhouette Mirage You draw like an adult. Don't worry. What I would consider "drawing like a 3 year old" would be using stick figures or drawing a "girl in a dress" as a circle and a triangle lol
And the fact that you understand symmetry means you understand more than a preschooler...
I'm not an overly harsh critic of art and there isn't a lot of artwork that I would genuinely consider "bad," and I try not to be too judgmental towards hobby artists (such as myself) and beginner/novice artists, but even I do find some things unappealing.
And with that said, there is some very famous and very professional artwork that I can't stand.

The thing that has helped me improve the most is practicing, and learning and applying new skills. My drawings from high school are terrible and embarrassing compared to what I'm able to do now. And I'm still not always satisfied with my art and sometimes I just erase everything or throw it out, even if it took me six hours.
And I will give advice from my own personal experience... don't let judgmental, elitist or overly harsh critiques prevent you from doing something you are passionate about. And remember that everyone draws in a different style and you shouldn't put all your energy into trying to replicate art that you think is "better" or more marketable than yours. You would be surprised at what people genuinely enjoy or are willing to purchase or commission.
If you don't practice, you don't improve. Take criticism for what it is (good or bad) and move on.

Here is my most recent project, which I'm going to get as a tattoo at some point in the near future... I know it isn't perfect, but I'm proud of it and I like it enough to use it as a tattoo design.

 

[Misery]

I disagree. Artificial Intelligence may get the brush strokes right, may render the form and the lighting and the textures photo-realistic. But emotion and meaning come from somewhere within. I am not a big fan of Picasso, but wow, every single drawing of his I've seen is wound so tight with emotion it drips from the page. This has carried over into some of his more geometric work, too. There will never be a Picasso AI. Sure, AI will be duplicate his work, or make work like his work, but it will never be able to interpret that world like Pablo did himself. I think we will, at best, connect only superficially with AI work but art that truly speaks to us will be subtle and filled with the best--and the worst--of humanity.

My favorite painting is an oil done in monochromatic red. There are no lines, there are no shapes, there is no form, there are no pixels or dots or anything concrete with which the mind can latch onto and say definitively, this is a ___. I told its owner, "I love this painting. If it were mine, I'd call it, mother and child in a red room. Turns out, the title was something similar, mother and child rocking in a sunbeam. Or something like that. It was the rendition of the image of what-it-is-like when the sun catches your eye. It was the experience of the moment that the artist had captured that fascinated me so greatly. There is no AI that can ever do. No AI can do that because they are not human; they cannot share in the all-too-human experiences we experience and share among ourselves every day.

I think the problem that artists are running into with AI though is that it is indeed possible to take the thing past the "looks like AI made it stage".

Like, if you let Midjourney just work entirely on its own? It's gonna have that "Midjourney look". Anyone can get it to do that.

But despite what a lot of people think there is actual skill involved in using MJ (or any other), and the process (when actually done in full, with some freaking time and effort) sorta... well, the best way I'd put it is that it kinda "adds" that emotion to it, with someone steering it. That, to me, is how it's meant to be used, but it also brings up problems in that one way or another, the AI still creates the final image, even if you were at the steering wheel the whole time.

But then there's other things that are more... ambiguous.

For instance, I did this with it:

Just another AI image... right? Not quite. See, this image is the result of an experiment of mine. When MJ had its recent giant upgrade, it received some new capabilities, and I decided to test one of them and see how far I could push it. The idea was to merge the concepts of one image with another, to get something that has the properties of both. And I wasnt content to use just any image.

I make fractal art... it's my primary thing when it comes to digital art (I've even got an entire Deviantart page for my fractal stuff). It's hard, it takes forever, but results in a lot of things I'm quite proud of. But what happens if I give them to MJ here?

So, I took two of my best ones, and simply fed them to it. I gave it no direction beyond the feeding, no description of HOW I wanted them combined. Just gave it the images and allowed it to do what it wanted... just to see what'd happen.

That up there is the result (after iterating and upscaling). Indeed, it absolutely has properties of both of them. But more to the point it just LOOKS like exactly the sort of thing I'd be trying to make with my fractal programs. That's my sort of style that I aim for (when making the 3D fractals, to be specific).

Question though... is it? Is it mine? The raw materials were indeed mine... each of the two fractals I fed it took hours to make, using my own skill and knowledge in the use of fractal tools. But... what is this? Is this mine? It's purely a combination of "mine", it literally has no other properties whatsoever, but... what is it now?

There's been other weird complications too, with this. These AIs arent ONLY able to make full images like this. There's also the ability to add to or manipulate already existing stuff. Deep Dream Generator (which I'm betting isnt getting as much attention right now) specializes in this. Take something you already have... say, a photo, perhaps... and feed it both that, and another image. It will analyze the style of the second image, and attempt to recreate the first image in that style. And it is really, really good at this. There is a "painting" that hangs in my house currently... it started out as a photo of my stepmother and her father together, a slightly blurry older image. When he died, I wanted something to give to her, but I cannot produce that sort of thing on my own... but I know how to get tech to do what I wanted. So, I worked with DDG until I got what I wanted (and this took awhile, and many iterations), and my father had the final result professionally printed on canvas to be given as a gift to my stepmother when grandpa died (which was slightly over a year ago).

Nobody could tell it was AI made. Nobody. Not until I told them. Everyone's face lights up when they see it, usually followed by mild confusion when it's explained just HOW I did it. Of course, it WASNT just AI made. It was literally a recreation of an already existing photo... with all of the emotion and love that photo contained... redone as what I think is an oil painting. It IS that photo, yet it also isnt.

So what does that make all of those things? Do they have no emotion and love in them because I used AI as a tool, despite using images NOT created by the AI? Despite that the things fed to it DID indeed have emotion and love in them? Questions like that are where things get extra screwy.

Of course those are only my examples, and I'm a bit of an outlier. Most people who use a given AI will just feed it text blerbs and use the first result given. To me, that's boring, but what do I know?

I will say one thing though, and possibly this just complicates this even further, which is that the things I've gotten out of MJ (seriously I've done a LOT of stuff with it) actually also can then inspire my physical art. In particular there's one character I got through the use of MJ that I ended up *really* liking, and my current goal in gaining drawing skill is to be able to draw that character myself with my pens and pencils, so I can use her in images of my own. So that adds another element to that.

Just my thoughts on it. No particular reason, I just find the topic fascinating. Sorry, I'll stop rambling now.

 

[Luca]

I think the problem that artists are running into with AI though is that it is indeed possible to take the thing past the "looks like AI made it stage".

Like, if you let Midjourney just work entirely on its own? It's gonna have that "Midjourney look". Anyone can get it to do that.

But despite what a lot of people think there is actual skill involved in using MJ (or any other), and the process (when actually done in full, with some freaking time and effort) sorta... well, the best way I'd put it is that it kinda "adds" that emotion to it, with someone steering it. That, to me, is how it's meant to be used, but it also brings up problems in that one way or another, the AI still creates the final image, even if you were at the steering wheel the whole time.

But then there's other things that are more... ambiguous.

For instance, I did this with it:

View attachment 94076

Just another AI image... right? Not quite. See, this image is the result of an experiment of mine. When MJ had its recent giant upgrade, it received some new capabilities, and I decided to test one of them and see how far I could push it. The idea was to merge the concepts of one image with another, to get something that has the properties of both. And I wasnt content to use just any image.

I make fractal art... it's my primary thing when it comes to digital art (I've even got an entire Deviantart page for my fractal stuff). It's hard, it takes forever, but results in a lot of things I'm quite proud of. But what happens if I give them to MJ here?

So, I took two of my best ones, and simply fed them to it. I gave it no direction beyond the feeding, no description of HOW I wanted them combined. Just gave it the images and allowed it to do what it wanted... just to see what'd happen.

That up there is the result (after iterating and upscaling). Indeed, it absolutely has properties of both of them. But more to the point it just LOOKS like exactly the sort of thing I'd be trying to make with my fractal programs. That's my sort of style that I aim for (when making the 3D fractals, to be specific).

Question though... is it? Is it mine? The raw materials were indeed mine... each of the two fractals I fed it took hours to make, using my own skill and knowledge in the use of fractal tools. But... what is this? Is this mine? It's purely a combination of "mine", it literally has no other properties whatsoever, but... what is it now?

There's been other weird complications too, with this. These AIs arent ONLY able to make full images like this. There's also the ability to add to or manipulate already existing stuff. Deep Dream Generator (which I'm betting isnt getting as much attention right now) specializes in this. Take something you already have... say, a photo, perhaps... and feed it both that, and another image. It will analyze the style of the second image, and attempt to recreate the first image in that style. And it is really, really good at this. There is a "painting" that hangs in my house currently... it started out as a photo of my stepmother and her father together, a slightly blurry older image. When he died, I wanted something to give to her, but I cannot produce that sort of thing on my own... but I know how to get tech to do what I wanted. So, I worked with DDG until I got what I wanted (and this took awhile, and many iterations), and my father had the final result professionally printed on canvas to be given as a gift to my stepmother when grandpa died (which was slightly over a year ago).

Nobody could tell it was AI made. Nobody. Not until I told them. Everyone's face lights up when they see it, usually followed by mild confusion when it's explained just HOW I did it. Of course, it WASNT just AI made. It was literally a recreation of an already existing photo... with all of the emotion and love that photo contained... redone as what I think is an oil painting. It IS that photo, yet it also isnt.

So what does that make all of those things? Do they have no emotion and love in them because I used AI as a tool, despite using images NOT created by the AI? Despite that the things fed to it DID indeed have emotion and love in them? Questions like that are where things get extra screwy.

Of course those are only my examples, and I'm a bit of an outlier. Most people who use a given AI will just feed it text blerbs and use the first result given. To me, that's boring, but what do I know?

I will say one thing though, and possibly this just complicates this even further, which is that the things I've gotten out of MJ (seriously I've done a LOT of stuff with it) actually also can then inspire my physical art. In particular there's one character I got through the use of MJ that I ended up *really* liking, and my current goal in gaining drawing skill is to be able to draw that character myself with my pens and pencils, so I can use her in images of my own. So that adds another element to that.

Just my thoughts on it. No particular reason, I just find the topic fascinating. Sorry, I'll stop rambling now.

That's actually really fascinating although it's almost creepy. I have heard a lot of valid arguments both for and against AI art.

I have one particular irl friend-of-a-friend who flips out on anyone who mentions using AI art but I think she's virtue signaling, for a lack of better words, to be honest. She tends to "learn" things on tiktok and then turn into a misinformed activist for subjects she doesn't have a very good grasp on. I think there are valid reasons that artists don't like AI but I don't think she understands the subject well enough to make bold statements like that.
I don't even personally understand it well enough to be telling people "Yeah, you should absolutely not be using this."

The things you create with AI are really beautiful though, and the story about your stepmother's painting is very touching. I would love if someone made a portrait like that from a picture of me with my grandparents.
For me, personally speaking, as an artist, I prefer to draw or paint things myself, and I prefer purchasing art that was made by actual human artists. But I can see the appeal of having someone (or something) else do it on a bigger scale for free.
Also, I think I mentioned recently that some artist friends have experimented with AI programs (not sure which ones) to help them draw pictures of me. They had mixed results though, some of them looked like a completely different person and didn't resemble me at all. Kind of creepy lol, like you wonder whose face that actually is, and where the AI found it...

I don't understand the subject of AI well enough to get into a debate on whether it is or isn't art, or is or isn't ethical. But like I said, I have heard valid discussions both ways.
I think the main concern that a lot of people have is whether it will take opportunities away from human artists and designers.
But if you enjoy it, and you're passionate about it, there's no reason you shouldn't be allowed to use it either. I hope that all made sense.

 

[Shevek]

I started drawing before kindergarten. I've always done more practical drawing than pure art, but sketching skills are the same for both. I started with pencil and paper, and have never felt nearly as comfortable with the eye-hand coordination using ordinary electronics. I rest my hand on the paper, and mostly make lines at right angles to my forearm, rotating the paper to suit. Others draw in books. A good compromise might be to keep drawings in binders. I make short strokes, extending lines or shading areas, and then adjust some line positions by making them wider. Erasing and re-drawing may be needed. A draftsman's erasing shield is very handy for selecting out errors. For long, Aubrey Beardsley type single lines, I would use a flexible ruler.

The work of renaissance artists such as Canaletto are good for learning perspective, and the use of a vanishing point to get a realistic appearance. This can also be studied using photographs, or just looking with one eye.

I recently tried out a set of instructions on cartooning, drawing some very simple shapes arranged together, and was astonished to have a face pop out, without even the outline of a head.

 

[Atrapa Almas]

No AI can do that because they are not human; they cannot share in the all-too-human experiences we experience and share among ourselves every day.

You seem to give high credit to your thoughts, rather than knowledge or facts. Maybe in your mental world "No AI can do that because they are not human" and being human is something very special. But current knowledge on AI clearly shows that they can do it. Not in 10 or 20 years but now.

Paint by algorithm: Can AI make art, or is it all just derivative?

Drawing on advances in machine learning, engineers are pushing the boundaries of creativity, creating a new medium and raising questions about what art can be

www.timesofisrael.com

Most people can’t distinguish between AI and human art, says a new study

Inspired by the $432,500 sale of an AI-created portrait, the study also explores what types of art are most associated with artificially intelligent creators

www.dazeddigital.com

Most People Can’t Tell the Difference Between Art Made by Humans and by AI, a Rather Concerning New Study Says

A new study suggests that humans largely can't tell the difference between human-made and machine-made art.

news.artnet.com

Elephants and monkeys can also create art. There is nothing mistic on the human brain, its just an Ape brain.

AI can play chess better than us, paint better, and do school essays better than us.

A new AI chatbot might do your homework for you. But it's still not an A+ student

ChatGPT gives users their very own virtual assistant. It could transform academia, experts say — for better and worse.

www.npr.org

Knowledge doesnt come from our inner sacred thougths, grear rethoric or smart logic. We need to learn whats going on the world. What works and what do not work.

No philosopher ever build a plane or created a new mediccine. It was done by scientists who actually tested stuff out of their heads. Those scientists are building AIs that surpass our applied intelligence and capacities in many fields and are about to change the world despite of our inner thinking on their capabilities.

 

[Misery]

You seem to give high credit to your thoughts, rather than knowledge or facts. Maybe in your mental world "No AI can do that because they are not human" and being human is something very special. But current knowledge on AI clearly shows that they can do it. Not in 10 or 20 years but now.

Ya know, this brings up another thought I just had:

Even the idea that AI art just BY ITSELF (as in, not fed anything like I did) doesnt have any love or passion or enthusiasm in it isnt quite right, if you really think about it.

These art AIs learn by example. They must be fed first (as in, their main database, not the prompts) before they can do anything, just like how an actual human artist is likely going to take their own style from other artists they see, even if they arent truly aware that they are doing so. And those examples are images that had all that love/passion/enthusiasm poured in. That's what the AI is working with, images with those elements baked in from the start. Now, what it can do sometimes (a lot more often than you'd think) is completely screw it up (seriously AI art mistakes are absolutely hilarious, it's part of what makes it fun) but it still wont deviate from those "created with love" paintings and such that form its base abilities. That stuff is in there on a deep, fundamental level.

However, when the prompts are very simple, the AI does have a tendency towards being very generic. That "Midjourney look" that anyone who uses it will be familiar with. If you dont drive that particular car yourself, it aint gonna go anywhere particularly interesting. The best AI art images were typically made by someone who sat there engineering the prompt for a couple of hours at least, making iteration after iteration (and the UIs and commands and such that you use with these are designed around that concept). So that's something to keep in mind with it.

Also I cant emphasize enough just how much this has improved since it started.

As an example, at one point I had the idea to do a sort of... surreal pipe organ (you know, the things you see in cathedrals). A surreal one of those with like an epic shape to it and a violet hue to the whole thing.

I spent hours working on that bloody prompt and trying to get it to do what I wanted. This was the best result it gave me at the time, a few months ago:

I do quite like that, actually, it's surreal and cool.

But I decided, just now, to take the exact same prompt that created that final image, and feed it into the now upgraded version of the AI, to see what would happen.

It did these:

THAT is what I wanted it to do the first time. Both of those are from the exact same prompt (MJ always starts you off with four variations when you generate something, and then you iterate and eventually upscale off of those), and there were two others along with them (two is enough here though).

Also I gotta show the owl.

My first attempt at a sort of clockwork steampunk owl sorta kinda:

Again, I spent hours trying to get it to do that. I've got so freaking many melty steampunk owls in my archive. So many. PAGES of them.

I took the final prompt that I got out of that, and again fed it into the thing just now.

This time, it gave me this:

Yeah. It has improved.

They dont have to end there, though. I can iterate them further. However far I want...

Making these is fun and all, but... why stop there? I can Do Stuff with these. Photoshop and Filter Forge and my kaleidoscope & distortion programs, can do all sorts of creative stuff with them. That's where it gets really interesting, at least in my view. Taking it beyond just the AI's output.

I wonder what would happen if I fed that owl into Deep Dream Generator and used one of my fractals as the "style" source? Now I gotta try that.

 

[Atrapa Almas]

But can a picture made by a computer be called art?

This has happened before.

Can a shoe made by a machine sustitude the work of a human artisan?

Can a meal made in a factory taste similar to a meal made with love?

All the industrial revolution was about that, and people reacted just with that kind of "human superiority over machines" logic.

But when you tell them how much the machine made shoes cost vs the human made shoes ...

As machines outperformed humans in manual labors like doing shoes, AIs are outperforming human in mental processes. From medical diagnoses to law decission, engineering calculations, design, art, translating, writting...

All those mental processes (and many more) are to be outperformed by AIs as manual processes was outperformed by machines in the industrial revolution. Then, the owners of the machines displaced the artisans that did all those manual labours. Now, the owners of this AIs will displace all the professionals that do this mental processes. From doctors, to lawyers, to engineers, teachers, psicologists, and also artists.

Ever dreamed about creating your own comic or Manga?

Using AI to Create New Comic Strips without Writing Any Code

A tutorial on how to use GPT-3 and DALL-E to generate original content for the funny pages

towardsdatascience.com

The owners of AI will be able to create content like comics, mangas, animes or movies in some years and humans will no longer be needed. That will hit companies like Netflix like a truck.

So yeah, its a revolution similar to the industrial one. I think Luca has the safest job of us all. Dogs are more attached to humans than to robots.

 

[Forest Cat]

This has happened before.

Can a shoe made by a machine sustitude the work of a human artisan?

Can a meal made in a factory taste similar to a meal made with love?

All the industrial revolution was about that, and people reacted just with that kind of "human superiority over machines" logic.

But art is very different from a shoe or soup. I don't think it can be compared. A shoe is just a shoe, who cares where it came from, I just want it to protect my feet. But art is art. So people can get machines to create a picture but something is taken away from it because it's just a picture a machine made. It loses something important. I can look at a picture a human painted and be impressed by it. But a machine can't impress me, no matter what the picture looks like.

The owners of AI will be able to create content like comics, mangas, animes or movies in some years and humans will no longer be needed. That will hit companies like Netflix like a truck.

I think humans will be needed because I'm not interested in watching for example a movie made by a machine. I would never pay to see that. It's just not interesting for me. I would however happily pay to not have anything to do with AI. And there is no way I'm the only one.

 

[Atrapa Almas]

And there is no way I'm the only one.

Resistance is expected like it was in the industrial revolution.

Before the first photografic camera were created, rich families hired drawing artists to draw them. Do you hire a drawing artist? Or maybe you use your cellphone?

Very rich people still hire human painters, is it because they are the only ones who care about the soul stuff or because they are rich?

Companies that now hire digital artists will hire AI companies instead. Consumers will not be able to spot the difference.

Do you do any research to know if the products you buy everyday were AI assisted? Dont think so. You just buy them checking quality and price, so you probably have already chossen AI asisted products as animal cruelty food, since they are cheaper.

Most artists sell their job to companies that will not care of the soul stuff. You can agree or resist, as many others. But the change is already happening.

 

[Misery]

A shoe is just a shoe, who cares where it came from, I just want it to protect my feet. But art is art.

Some people genuinely do see shoes as being art, though.

I know a couple of people who are like, really into that. Like, big shoe collections and they've gotta know the styles and yada yada yada. It's all about design and aesthetics and more details than I care to know about.

Goes for all sorts of products, not just shoes. Even for things you might not expect.

I think humans will be needed because I'm not interested in watching for example a movie made by a machine. I would never pay to see that. It's just not interesting for me. I would however happily pay to not have anything to do with AI. And there is no way I'm the only one.

This is where things get kinda wonky though, I think.

Technically the first Toy Story was made by a machine. Humans had to DIRECT that machine, using 3D modeling software, but... technically you could say the very same thing about AI creations. Yet the methods and tools are wildly different to the point where the comparison makes no sense, in terms of practical functionality.

And even without AI, there's all sorts of screwball situations where the question pops up.

There's this, for instance:

I spent hours making this accursed thing. It's a 3D render, and there's no AI, so was it made by a machine? Yes? No? But it's also a fractal object... it outright cant be made without a machine, it's literally made of math, and it dives into infinity, as fractals do. But there's no AI here, none whatsoever. There's just me and a really convoluted disaster of a program. A 3D rendering software that works nothing like traditional 3D modelling.

So.... what, then? Who made it? Me, or the app? What is it? Art, or not art?

We havent even gotten into AI with this image, because there is no AI present, and the question is already getting weird. More of a headache than I want to deal with, that's for sure.

It's one of those things where the answer is likely to just be very, very subjective. An endless argument, likely.

Personally though I do think paint and pencils are more interesting, even if I like making fractals. But that's just me, some people are going to be REALLY into the AI stuff.

 

[Shevek]

One of the first machines for mass production was still in use at last report, making gang pulleys for the Royal Navy. Not only are the products handsome and functional, so is the machinery. Things made by craftsmen get embellished with art, because they enjoy it. In AI, the enjoyment goes to the programmer, and as with art prints, we wind up with fewer artists making a living.
Recently, I made an Epoxy weighing scale, and for the first time, I wasn't just in a hurry to get my glue mixed. It took over two weeks, and it is gorgeous as well as practical. The hand/eye coordination is something that evolution made enjoyable, distinguishing us from other creatures with better tools and shelters.