This is the second post of three about renderings by Dream AI using some of my images as seeds.

As a quick reminder, Dream AI (well, all the AI models) has the option to provide an image to serve as a seed for what it creates. The Pro version of Dream AI has the option to output a video of the creation.

I don’t know how accurately the video represents the creation process because it’s too quick relative to the amount of time it takes to render something. Then again, it might be speeded up.

Ah, crap . . . basically, I don’t know what the video represents, but it purports to be the AI in action as it renders the image. For instance, this is the video generated for the above image.

FYI, the prompt was steampunk submarine in an underwater city, and the seed photo was a leaf on some rocks (you can see it in the video).

So, whenever I need to pass some time (waiting for stuff, waiting in the car, etc.), I play with Dream AI by WOMBO. The above is a rendering from the prompt “steampunk spaceship”. Each prompt gets me four renderings. I may save all four or none, depending on how they look.

Here’s another from the same prompt . . .

Dream AI by WOMBO

Most of my days are pretty good.

Not fantastic, mind you . . . no one showers me with accolades, money, and songs, but also no one is calling me names, spitting on me, or kicking me.

I get up, have a coffee, feed the birds, do a few things, go to the gym, come back and have a snack, do a few more things, eat dinner, watch a few shows with Melisa, do a few more things, then head to bed.

All in all, a pretty good life. Occasionally, something brings a bit of a downer, and this was such a day. Don’t worry; nothing personally terrible, but still a downer.

About two weeks ago, I loaded the Dream AI App on my Samsung Note 20. The app is free, by the way, but you can pay to upgrade it and get a few extra capabilities.

I don’t even remember the prompt for the above. Unlike the graphics I generate on the PC, I usually don’t save those to the account . . . I just download them to the phone. I assume the prompt had something to do with Halloween, but I don’t know for sure.

Anyway, because it’s on my phone, and because I’m sometimes waiting for stuff, I played around with the app a whole lot . . . the gallery I’ll be linking has 120 images on it, but I won’t be sharing all of them here.

Readers might wonder about the title. Well, let me mansplain.

For them who don’t often read my posts (the overwhelming majority of the Known Universe), I posted a few AI Generated graphics back at the end of last month. Appropriately enough, they were all about Halloween (HERE).

But I left out the Dream by WOMBO offerings. Not intentionally, mind you. I just hadn’t run them yet.

“So why show them now? And you still haven’t told us about the title.”

Impatient, a bit, Bob? Well, let me finish mansplaining.

This post has been scheduled ahead of time since my internet presence will be sporadic over the next ten days or so . . . and, yes, more AI Art Generators creations . . . of spaceships! Maybe even one we might recognize.

Here are some of the terms I used in these prompts, although the prompts were not exactly the same for the different AI Art Generators (DALL-E, Dream, MidJourney, and NightCafe):

Steampunk, spaceship, in space, in city, 3d shading, hyperrealistic, 4K uhd, hyperdetailed, photorealism, copper, crystal.

So, let’s begin with DALL-E. I asked for spaceships in space, because, apparently, if you don’t specify space, there’s nothing to indicate these ships are in space. Heck, even when you ask, you can’t always determine we’re in space . . . we might be underwater, for all I know.

As near as I can tell, this is an early Borg design, before they figured out cubes — green cubes — are all the rage. The DALL-E generator has a definite look to its creations . . . but then, so do the other AI Generators. Anyway, to continue . . .

This post has been scheduled ahead of time since my internet presence will be sporadic over the next ten days or so . . . and, yes, more AI Art Generators creations . . . dinosaur creations, no less!

Here are some of the terms I used in these prompts, although the prompts were not exactly the same for the different AI Art Generators (DALL-E, Dream, MidJourney, and NightCafe):

Steampunk, Mechanical, T-Rex, in city, 3d shading, hyperrealistic, 4K uhd, hyperdetailed, photorealism, copper, Michelangelo.

So, let’s begin with DALL-E. I asked for a Michelangelo rendering of a mechanical T-Rex . . . obviously, Michelangelo put very little effort toward making these striking . . .

I think I might be able to conjure up something like that just by doodling. The next one is a little better . . .

So, rather than writing my story for the challenge, working on my photos, or doing anything productive, I occasionally return to playing with AI Art Generators.

I’m old, you see, so I don’t feel the push (or pull) to achieve anything in particular. After all, it’s all I can do just to put on my pants in the morning without falling over . . . most mornings, at least.

So, when the mood strikes, I do what the mood wants, and today, I wanted to share balls; crystal balls.

For these AI Art Generators renderings, the prompts contained words like crystal ball (occasionally misspelled as balll), dystopian, realistic, copper, detailed, and so on. I ran those with both cyberpunk and steampunk qualifiers, and I’ll begin with the cyberpunk renderings.

As I’ve been doing in these posts, I open with the renderings from Dream by WOMBO.

Yes, more AI-Generated art. For the few who might not know what is meant by Steampunk . . . well, you can click on THIS LINK. Basically, imagine no electricity and everything powered by steam. So, for instance, you could still have computers (HERE) but not run by electricity.

I ran a number of prompts on all four AI Art Generators (DALL-E, Dream by WOMBO, MidJourney, and NightCafe). Understand, there may be more AI Art Generators, but these are the ones I’m currently playing with.

The prompts ran from simple to more complex, and some incorporated other stuff besides ‘Steampunk Computers’ such as: drawn by Michelangelo, in the desert, made of copper, and some I currently don’t remember.

For instance, this was Dream’s concept of a steampunk computer as drawn by Michelangelo.

Dream By WOMBO

As we’ll see later, quite a different interpretation from MidJourney.

Just a quick post because it got late as I was playing with more AI Art Generators.

Yes, I wasted time I should have spent writing, but I had a good reason. I had credits that were expiring in a few days, and I didn’t want to waste them, so, first up, MidJourney.

But what prompt to give? Obviously, “robots playing chess in a desolate landscape“.

Well, that didn’t go so well, now, did it?

Yesterday I sort of dissed NightCafe . . . and, it’s true that trying to make something that looks realistic is a near-impossible task . . . so the answer is to not worry about realism and just go for broke, which is what I did . . .

but first, three offerings by Dream by WOMBO. I’m posting them because I find the engine fun to work with, and there are still many options I’ve not explored.

As mentioned before, the same prompt and style will not generate the same graphic because there’s also a random number generator that picks a seed photo. Of course, you can provide your own, but I’ve yet to try doing that.

I’ve not tried it because there are still many styles to try.

I wish they would be more consistent with their looks . . . the fact that two identical prompts and styles can give you drastically different results makes it difficult to plan out something specific.

In that regard, it’s fun to play with, but frustrating to create with.

But, on to NightCafe . . .

Yesterday’s post covered three AI Art Generators, and the easiest site to play with — and it’s free — is Dream by WOMBO. Click on the link, enter a few words, pick a style, and you’re in business. And that’s what my sister AnnMarie and me did this afternoon over the phone.

I ran this on my computer
She ran the same prompt and style on her computer

Obviously, the same prompt and style choice doesn’t generate the same graphic, and that makes sense otherwise, we would have a finite number of possible results.

I mean, there probably is a finite number unless they have a quantum random number generator . . . but I could be wrong.

From here on, my trading cards have a title, whereas hers just say “Dream” for the title.

Anyway, we tried different prompts where we used the same words and styles . . .


Some, like this next pair, were closer than others . . .

Some might remember when I played with Deep Dream Generator a few years ago. If you need a refresher, read THIS post, and, if you are a masochist, links to all related posts are HERE. The SmugMug Gallery collecting all those images is at this LINK. Warning: there are 382 images, and while many are (in my opinion) very good, there are many where I was trying out stuff and learning.

What is Deep Dream Generator? Well, it’s a bit like what I do with Topaz Studio when I artsify my images. Essentially, it blends the content, style, and colors of one image with the style of another to generate a mishmash with elements of both.

For the sake of this post, here are a few images to give you a flavor of what Deep Dream could generate:

Hummingbird photo plus Lava photo.
Buffalo and tree bark.
Gorilla and Singer sewing machine

Some of the photos are small because you ‘earned’ abilities and options the more you used the app. By the way, the site is still active; it’s just that I lost interest in it and moved on; however, my account is still active (HERE).