Sign In

StareWare - Screen Glow Light and Hypno/Mind Control Software

180
1.3k
7.6k
38
Verified:
SafeTensor
Type
LoRA
Stats
1,277
7,638
Reviews
Published
Nov 13, 2023
Base Model
SD 1.5
Training
Steps: 1,200
Epochs: 16
Trigger Words
StareWare
looking at screen
<<see description>>
Hash
AutoV2
0E7A180339
default creator card background decoration
iamxenos's Avatar
iamxenos

This LoRa solves two things I've been trying to do for a while. The first is something I've managed to do before by pure luck, but never under controlled circumstances. The second is, of course, feeding back into my twisted Hypno/MC Fetish.

First: You can show people looking at device screens (of many types) and then get it so they are being lit by the screen on the device.

Second: You know... Mind Control/Hypno Stuffs.

USAGE:

Works well on most models I've tried - 0.7 is a good starting point, but it does vary a lot. And, depending upon what all is going on in the scene and combinations with other LoRas, it might need to be adjusted up or down a bit.

HINT: If certain elements are warping/distorting (especially like hands or fingers - but computer screens and other stuff can get it too) then turning it UP tends to help. BUT... it can also burn a bit if you move it up too far. I've left it like this because certain trained concepts within this don't work without the extra training - which then pushes the whole thing closer to that burn point. (I'll be working on improving the "safe range" a bit in future updates.) For now - this works really well in most cases, but it can sometimes be tricky to dial in the proper weight.

Primary Triggers: StareWare and looking at screen. Though it doesn't always work - looking at screen tries to and sometimes successfully helps keep the subject's eyes on the screen. It's inconsistent though. I like to stick it in there regardless.
Cool Side effect: If you do "looking at viewer" and you don't call the type of device - especially if you put your subject on a sofa - it quite often skips drawing the nearby device and makes it as if our POV is from a wall television (not in sight). I have to experiment more with this - it's not something it's trained on - just something I noticed happening during testing.

People: 1girl, 2girls, 3girls (and also 1 woman, 2 women, 3 women if you model skews young) work right almost every time. I think I only had one or two "multiple person" test runs that got it wrong. It CAN take males in the mix, too - but that adds confusion. And with two girls - you usually end up with one "victim" and one "surrender your mind and become like me" person - the corruptor and corruptee so to speak. With three subjects - it can still happen, but it's less reliable.

Screens: Desktop Computer, Laptop computer, Tablet, Cellphone, Holographic Display. Though it doesn't work consistently, "Television" might work too - but I didn't have many source images of that - and the look of TVs is so varied that it just didn't really have enough to latch onto.

Lighting Controls: dark room, and lit by screen Lit by screen can work by itself, though the closer you get to broad daylight, the less of an effect it has. Dark room serves to darken up the background and make that glow from the screen more pronounced.

Places to sit/interact: Sofa, Bedroom, Table, Desk, and coffee table (but a lot of other work, too). Tip: If you call for a sofa, and a laptop, it'll tend to be in someone's lap. If you add "coffee table" it tends to put it out in front of them on a coffee table instead. (Not perfect, but... enough that I can say it's "working" - your checkpoint may vary, though).

If you look in the meta (hit I icon on the LoRa in your list) you can see a bunch of other things in there which either don't really work as intended or that I haven't tested enough to make a call yet. Feel free to try them.

NOTE: Glowing is a risky one - and there are several in there. De-Emphasising something like "[green glowing eyes]" as I did in one or two of the example images does make it better. I'm still not ready to call that a "win" yet. Feel free to play with that - and the rest, but I make no promises at all.


Like it? Buy me a beer.

I'm using Civitai's trainer - so Buzz Point tips are helpful too - even just a few add up.

Want Tips, Tricks, Pix of Things We Can't Show on Civitai?
Join the XenoEngine Discord to discuss tips and tricks, complimentary Lora's and Embeds, and to post (and see) images not allowed here on Civitai.



The Origin of the Name StareWare

There's nothing important in this section, but if you're curious... please read on....

Going on 20 years ago, I wrote a series of stories at MCStories.com about a bunch of girls using self help software on their phones and monitors - the only problem is that the software had no safety protocols so, while it could easily use the flashing lights and binaurals to help you break bad habits, rewire conscious and subconscious control over things like metabolism and inhibitions, and all that - it also had some hangups. These primarily started when our lead heroine decided to lower her inhibitions a bit so she didn't hate her job as a bartender at the local strip club... and from there it gets weird.

The first story in that series, and the name of the software that does it all is: Stareware. And all but one of the other stories by me on there all take place in that same "PinkLink" universe. (Each story from Stareware on tells you where to go next if you decide to bother reading it one day).

I've always wanted to come back and finish it - and possibly update it. But I've wanted to try to do it in visual novel or some kind of game format - but never really figured out exactly how to do that myself. I'm a coder, not an artist. The AI revolution has (hopefully) changed that - and this LoRa is one key part I needed to be able to accomplish that. :)

So... there ya go.