A New day, a new beginning

[MadMax359]

really like the last one!

[cerberus]

really like the last one!

I had a feeling you might.

Couldn't post any last night, got the GoDaddy lockout message. Still, never mind, today's another day. It's a bit of a mixed bag again.

This one was pretty much an accident, leftover prompt and silly slider setting. However, sometimes images speak to you and this one is quite insistent.

00205-2667504740.png

Another accident leftover prompt on an Img2Img generate, but I like it and it's quite different in feel from the source image.

00047-400983336.png

Img2Img taken from an old drawing, nothing like the source, even the artist would have trouble recognising it.

00172-400983336.png

Another based on one of my photomanips, bit closer to the original than the last one in my previous set.

00285-3525890497.png

This is based on an old pen and ink drawing, but again you wouldn't be able to link the two. I've done a bit of text to accompany this one (see below picture)

Passage.jpg

Julie had been told by the natives that she had to enter the tunnel naked, to do otherwise would be an affront to the gods and would have brought a terrible retribution. Though uncomfortable with the idea, she'd reluctantly agreed; but was now beginning the regret the decision. The entrance to the tunnel was down a steep bank of wet sinewy organic material, that strangly seemed to wrap itself around her body and cling in a peculiar way. It was almost as if it was alive! “God”, she thought, “I hope not”; however, the more rational side of her psyche dismissed the idea as preposterous. Reaching out she unconsciously squeezed some strands from the side of the tunnel, oddly they felt warm. Standing, as she, was up to her lower thighs in a mixture of water, earth and more of the sinewy material, she contemplated the tunnel ahead with some trepidation. It should have been dark beyond the entrance but a faint glow emanated from the walls to dimly illuminate the tunnel ahead. Further down, the tunnel seemed to open into a larger chamber but nothing was visible beyond that point. She moved forward half a step to see how easily she could progress. The clingy sinewy threads didn’t immediately fall away, but seemed to reluctantly slide off her body in a way that was surprisingly erotic. Deciding they weren’t really an impediment to motion she continued; however, she found that, as some slipped off, her motion forward resulted in others rising to take their place. This was distinctly disturbing and she hoped that mess in the tunnel floor didn’t get any deeper or she felt she wouldn’t be able to control herself ...

[Theo]

Just now saw all these after a months hiatus. The good ones are too numerous to mention; it's like walking into a buffet — a gourmet buffet!

[cerberus]

Just now saw all these after a months hiatus. The good ones are too numerous to mention; it's like walking into a buffet — a gourmet buffet!

Thanks. It's surprising what can happen in a month. That's about how long I've been using SD, seems like ages now!

[Theo]

Thanks. It's surprising what can happen in a month. That's about how long I've been using SD, seems like ages now!

Forgive me if I (most likely did) missed something, but what is this "SD" you speak of?

[Viridian]

Thanks. It's surprising what can happen in a month. That's about how long I've been using SD, seems like ages now!

Forgive me if I (most likely did) missed something, but what is this "SD" you speak of?

Stable Diffusion.

[cerberus]

Viridian is correct, all the images in this thread are generated using the Stable Diffusion AI and a browser interface called Automatic1111. Stable Diffusion is an Opensource application (i.e. it is free to download and use); however, it is fairly resource intensive and needs a reasonably up-to-date and powerful PC with a good graphics card to run it (it uses graphics card memory for working memory as that's faster than normal memory). Several of the online AI generators (e.g. Perchance) also use SD as their background generator. I started this thread when I got my first images after installing SD (hence the title) and use it as a way of documenting my development in using SD and exploring it's options, features and various extensions. You may notice me mention things like "Txt2Img" and "Img2Img", these are the basic image generation source features, the first generates images from text prompts (like Perchance) and the second uses a supplied image as a source. There are various other extensions that can be loaded and used to further tune results (e.g. ControlNet, by which you can supply an image to constrain the pose of the generated results, InPaint which allows you to regenerate selected parts of images to correct faults or change things you don't like etc). With SD the resources available are extensive and wide ranging, unlike say Perchance, where you get what they make available to you and are subject to any restriction on generated output they choose to impose.

Hope that helps make more sense of some of my cryptic comments in the preamble to image sets.

[Theo]

Stable Diffusion.

Viridian is correct, all the images in this thread are generated using the Stable Diffusion AI...

Hope that helps make more sense of some of my cryptic comments in the preamble to image sets.

Hmmm... I may be trying this Stable Diffusion AI. The price is right, and I've got a GTX 4080, so that might work for me. Thanks to you and Viridian for the intel!

P.S. It's probably not your cryptic comments, I probably just missed it. I just like to look at the pictures lol

[cerberus]

Hmmm... I may be trying this Stable Diffusion AI. The price is right, and I've got a GTX 4080, so that might work for me. Thanks to you and Viridian for the intel!

P.S. It's probably not your cryptic comments, I probably just missed it. I just like to look at the pictures lol

Nothing wrong with looking at the pictures, that's what they're there for. If you only look at the pictures though you will miss some of my crummy jokey comments but you may consider that a godsend!

The GTX4080 with 16Gb should work well, my card has 12Gb of memory and, on occasions, I've used all that; if SD starts using "shared memory" (i.e. base memory) it slows down quite a lot. The other things to bear in mind are disk space and internet connection speed, this stuff is big. Excluding image files (they take quite a lot of space) my current install is about 54GB. You don't start with all that but to get up and running you'll need about 30GB of it, that's takes a while on a slow internet connection (I'm getting about 27Mb).

The link below is to the YouTube video that I used as a guide to doing the install, it is worth having a look to see how it's done and whether it is a can or worms you want to open. I followed this and got a working install at the first attempt, so I think it's a good guide.

https://www.youtube.com/watch?v=kqXpAKVQDNU

One of the useful extensions you may (will) want to add later (ReActor) requires Visual Studio, you'll have to decide whether you are comfortable with installing that (note: it's free for personal use). If you have any questions I'll help if I can.

[Jinn]

Love the ropes. Bondage works well with quicksand; they’re like kissing cousins.