I've long generated images using 3D modeling and rendering. Some are really good, almost photorealistic, others are OK, and some suck. The quality depends on the tool I used and how much effort I put into it. A lot of the early ones made with Daz 3D are just terrible. Newer stuff is better. The better stuff usually came from 3DS Max, which I got quite used to, but its insanely expensive and not available for Linux, upon which I do almost all my work, so for 3D modeling I've switched to Blender, which is free and works anywhere.

Of course now we have a new method: generative AI. That didn't exist when I started this site, and still has a way to go, but it produces some first-class realistic images. And some crap. Here are a few of the images I tried just to see how it goes.

	This my attempt to render High Queen Siptiss from The Black Tower. It meets my specifications, but looks like nothing I had in mind. Still, this is the best one that came out. It looks fairly realistic, but notice that her scepter takes a strange jump as it passes through her hand. That's the sort of glitch you have to watch out for. She gets the rather stupid idea of sending a battle fleet into the Kyattoni galaxy.
	A rendering of Titrinka, the capital city of Kyatton. The AI didn't do too bad. It sort of got the scale of some of the buildings, but at best the tall one looks to be about half the height of the actual ones. The residential buildings reach up some five kilometers. And Titrinka is famous for being a rainbow of color. No color here. And Titrinka doesn't have streets; the Kyattoni have no need of them.
	This one actually works as Jemah Hahrl's research station on Dahrshannah. The climate and terrain are right. I didn't visualize the station itself that precisely, but if I had, it would have been something like this. Of course I had to go through several attempts to get this one.
	Even if you're reading my stuff, you would not have met the twins yet because I haven't written them other than an allusion in the prologue. Kolikki (the girl) and Kolinda (the boy) appear in A Remembrance of Evil. There is an argument that they're half human, but there is also the argument that they're not human at all because their genetic material is not DNA. Also, humans never have boy and girl identical twins. I didn't ask for them to have their hands in each other's pockets, but it's a perfect and serendipitous detail, as they're psychically linked. Don't fuck with them. They're actually much scarier than the Grady twins. No matter how hard I tried, I couldn't get the dang AI to produce T-shirts without the silly images.
	Just another werewolf with horns, not what I was looking for. The twins, along with Stebek Wyrn of Abranda, confront the horrors of Trigdru, known across the galaxy as The Planet of Evil. Nobody, but NOBODY goes there of their own free will. But... You know. There are things on Trigdru corresponding to virtually every monster ever conceived on Earth, and some that haven't been. One in particular, I didn't even try to render.
	Two abominations that came up when I was trying to render a creature native to Frelorala. The AI was clueless because it had never seen anything remotely like what I was trying to describe. It's a creature that starts out from a seed like a plant, but soon grows a heart and circulatory system, lungs, and a brain. It reproduces through a crown of flowers like an earthly plant, but also has a ring of typically eight toothed tentacles with eyes that quickly gobble up non-pollinators that find the flowers.
	A rendering of the Crown of Briannon. Not at all what I pictured, but sometimes that's a good thing. AI often puts in details you never would have thought of, and even if you never use them, they can serve as a springboard for more ideas or a fountain of inspiration.
	An attempt at the Star of Ages from Tetragrammaton. Again not what I envisioned, but it gives me some ideas. When Timothy Saugers learns he must find the Star of Ages, all he knows is that it's hidden somewhere in the universe, protected by an impenetrable barrier, and rotated out of our spacetime. Talk about a task!
	When AI screws up. Well, this image has the elements I asked for, just not in the way I expected them. You often get things mixed up like the poor inventors in both versions of The Fly. It doesn't actually know, for instance, that men don't have rockets for heads.
	I just couldn't resist tossing in this example of misguided AI in Trump's misguided attempt to convince people he actually goes to church. I can see him resorting to AI fakes, but someone on his team should have noticed that Donald only has five fingers on each hand, something AI wouldn't actually know.

So, that leaves me three ways to generate images for my novels, for covers, publicity, and general edification:

1. Licensed images
2. 3D modeling and rendering
3. Generative AI

And all three have issues.

Licensing images can be expensive and usually have restrictions on how you can use them, like on one web site and in a single publication of up to 100 copies. After that, you need another license. My favorite source was CanStockPhoto, but they shut down. Alas! Even if it weren't for the cost, you can spend hours or even days going through gallery after gallery looking just the right degree of visual excitement. Three hundred space stations so far, and not one of them is right.

Likewise, 3D modeling can eat up a lot of time. That's why some of my renderings are crap. I didn't have the hours to spend on one. And the learning curve is pretty steep. You have to learn all about editing meshes, modifiers, coordinate mapping, materials, which include things like anisotropy, ambient occlusion, light sources.... You get the idea. Daz 3D is a notable exception in that you can put together a character without too much ado and do a quick render. If you want your character to actually be in a setting, that's often another piece of software. More hours get what you want.

Generative AI can produce realistic images quite quickly, but tweaking it to what you want can be tricky. You have to get the phrasing just right and there is still no guarantee. The techie solution is to train your own neural network models. That's well within the capability of home users with the right hardware, but again you're looking at hours or days to accomplish that. On of all that, getting the same spaceship twice is less likely than getting struck by lighting. You need 3D modeling for that.

There is a fourth option, but it's realistically closed to me. Except for drawing anime, my artistic skills are nowhere good enough to paint a realistic scene, and even if they were, that's more hours or days.

There you have it. A choice of methods, all of which require a significant investment in your valuable time. Here's hoping that AI improves enough to change the equation here.

At this stage in re-writing my web site, comments are not functional yet. That's too bad because I'd really like to hear from others on this.