- Jun 19, 2018
- 20
- 22
Have you ever thought about gothifying some KoD characters??
- Mar 30, 2019
- 86
- 100
Definetly not:
You don't have permission to view the spoiler content.
Log in or register now.
Btw: Question to all SD-"Promt Heroes". If you look at then last Image: With both Kira and Sabia in the picture I get a mix between both styles. How do I define a prompt so that I can describe both seperately?
Thank you for explaining how to use AI. I sat at the computer all night and was happy like a little boy.
I hope you enjoy my pathetic attempts...
I hope you enjoy my pathetic attempts...
You don't have permission to view the spoiler content.
Log in or register now.
You should use extensions like this:
You must be registered to see the links
Or this: https://f95zone.to/threads/kingdom-...-future-plans-healthcare.211283/post-14012626
I always use tag "pointy ears" in negative prompt to avoid elf ears for Sabia. Plus this doesn't affect Neve or Ylva ears.
Also i add "mole" in negative, because AI likes to add more than 1 moles)
Here is my current default negative prompt for KOD:
You don't have permission to view the spoiler content.
Log in or register now.
Adding tags like: missing leg, extra limb, e.t.c. works for some images, but also fails another images, so i found that this tags useless in default variant. They really usefull when you try to fix specific image (when image seed is constant).
Last edited:
- Mar 30, 2019
- 86
- 100
Thanks, that works.
But the overall quality is worse than without the plugin.
You don't have permission to view the spoiler content.
Log in or register now.
===============
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
- Jun 19, 2018
- 20
- 22
After making so many different styles, do you have a favourite one(s)? Btw love the goths!
I think Frank Frazetta (realistic colorfull) and Kentaro Miura (grayscale lineart) styles most favorite for me.
I highlighted some of my favorites styles in this two posts:
https://f95zone.to/threads/kingdom-...-future-plans-healthcare.211283/post-14292079
https://f95zone.to/threads/kingdom-...-future-plans-healthcare.211283/post-14292087
You don't have permission to view the spoiler content.
Log in or register now.
P.S. Style for me is not so important, actually) If drawings are not very ugly or kawaii, so it's ok. And it just that AI really great in reprodusing different styles, so it's fun to test them.
Last edited:
Posted 2 new versions of KOD model. Added 1 more character: Jasmine) This time i used higher resolution for training, but result is still crap.
Also i noticed that earlyer epochs produses better results for some characters and bad for another, same true for later epochs, but later better understands character related details. Maply is good example of what i saying)
I think this is my last multi-concept model for KOD. Next models would be dedicated to characters (1 model per character).
Overtrained. Better at details.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
Undertrained. Better at reprodusing NOMO style, except for catgirls.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
No well trained model(
Waiting your feedback)
P.S. Now i can test all styles again
Last edited:
In honor of the new versions of KOD model
. We are rolling out the update:
. We are rolling out the update:You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
maybe solution use both model at variable strength? maybe ~0.2 for one, 0.4 other for best result depending on character?.....
No well trained model(
Waiting your feedback)
P.S. Now i can test all styles again
you know already which character work good for each model?(save me buzz from finding out myself
Am i only one always with not enough buzz? I try to use emote on every picture in kod & makina models so everybody get more but I still feel like beggar everyday claiming free reward
1.maybe solution use both model at variable strength? maybe ~0.2 for one, 0.4 other for best result depending on character?
you know already which character work good for each model?(save me buzz from finding out myself
Am i only one always with not enough buzz? I try to use emote on every picture in kod & makina models so everybody get more but I still feel like beggar everyday claiming free reward
When lower strength you get less Nomo style and 3d or anime, depending on prompt.
2.
v3.5 ep15 decent for all characters at strength 0.7 (less Nomo style);
v3.5 ep5 decent for all characters, except catgirls, at strength 1.0 (more Nomo style).
I can upload another epochs, if someine want to try them, on pixeldrain or another filehosting, i probably delete them in a few weeks.
3.
As i sayed better to create special models dedicated to characters (one per model), because with models you can expect, at least,
that you get right character appearance, without interference from another conceps in model.
4.
I just buy for myself 80k buzz few month ago, and also collect 10k at another (non-lewd) account, by posting few lazy remixes every day, which is entertaining for me.
You don't have permission to view the spoiler content.
Log in or register now.
If we someday start to make fan mod or remake, do we need to keep original (Nomo) artstyle? Just curious about it..
P.S. I was thinking about to request another KOD models from civitai creators, they probably create better models, then me.
In first image Sabia looks like Evangeline Lilly) And now i see orc with tail)In honor of the new versions of KOD model View attachment 3857009
You don't have permission to view the spoiler content. Log in or register now.
You don't have permission to view the spoiler content. Log in or register now.You don't have permission to view the spoiler content. Log in or register now.You don't have permission to view the spoiler content. Log in or register now.You don't have permission to view the spoiler content. Log in or register now.You don't have permission to view the spoiler content. Log in or register now.
Few questions:
1. Is new model better at eyes generation?
2. Is new model better at tusks (for Ylva) generation?
3. Does new model have problems with colors, like oversaturated or pale, or just bad colors?
It's not always possible to work well with small details.
There is very little experience in this area...
1.+
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You wrote above that you need to work with specific models separately. Even though I have little experience. But I noticed that this is true.
Last edited:
Thanks for detail answer.
I think is best to try to stay close to original art style for game mod. Maybe other styles also nice but difference between that and original style maybe look wrong.
Different model for each character would be very nice I think, much easier in prompt aswell so easier to get good picture out with less complicates prompt.
I not try new models yet but I think your old model was good for generating neve/sabia.
Maybe seperate model for only panels and other mod for standing character in conversation would be good. Not sure how difficult to make though..
- Jun 19, 2018
- 20
- 22
Could you by any chance help me and explain how to properly prompt different characters? How to group specific features between for example 2 girls (ex. prompt specifically Sabia to have purple hair and Neve to have white hair in the same image) I've seen a lot of 'methods' like using brackets or 'BREAK' but neither seem to work properly