Natural Sin Final and last of epiCRealism
Since SDXL is right around the corner, let's say it is the final version for now since I put a lot effort into it and probably cannot do much more.
I tried to refine the understanding of the Prompts, Hands and of course the Realism.
Let's see what you guys can do with it.
Thanks to @drawaline for the in-depth review, so i'd like to give some advices to use this model.
[expand to see Advices]
Advices
use simple prompts
no need to use keywords like "masterpiece, best quality, 8k, intricate, high detail" or "(extremely detailed face), (extremely detailed hands), (extremely detailed hair)" since it doesn't produce appreciable change
use simple negatives or small negative embeddings. gives most realistic look (check samples to get an idea of negatives i used)
add "asian, chinese" to negative if you're looking for ethnicities other than Asian
Light, shadows, and details are excellent without extra keywords
If you're looking for a natural effect, avoid "cinematic"
avoid using "1girl" since it pushes things to render/anime style
to much description of the face will turn out bad mostly
for a more fantasy like output use 2M Karras Sampler
no extra noise-offset needed, but u can if you like to š
How to use?
Prompt: simple explanation of the image (try first without extra keywords)
Negative: "cartoon, painting, illustration, (worst quality, low quality, normal quality:2)"
Steps: >20 (if image has errors or artefacts use higher Steps)
CFG Scale: 5 (higher config scale can lose realism, depends on prompt, sampler and Steps)
Sampler: Any Sampler (SDE, DPM-Sampler will result in more realism)
Size: 512x768 or 768x512
Hires upscaler: 4x_NMKD-Superscale-SP_178000_G (Denoising: 0.35, Upscale: 2x)
Useful Extensions
!After Detailer | ControlNet | Agent Scheduler | Ultimate SD Upscale
ā No VAE needed but it is better to use one for more vibrant colors
ā Feel free to leave Reviews and Samples - and always have fun creating ā¤
Description
Easier Details, more realistic skin, cinematic look, better hands
FAQ
Comments (14)
i'm trying out the model with the promps from the example images, but get a lot of double faces and double bodies, what can i do to prevent this?
Did u use resolution 512x768 with Upscaler 2x?
New era uses the same v1 inpainting and pix2pix?
U can try, I didnāt make a new one since I donāt really use them
trying to replicate the image of the girl in the forest but it always crashes in the middle with something related with the upscaler (4x_NMKD-Superscale-SP_178000_G). Any idea why?
U donāt have the Upscaler. Get it from here https://upscale.wiki/wiki/Model_Database and put it in the models - ESGRAN folder
WOW ! Such realistic results dude... Kudos to you... I am facing issue getting my trained model to give more realistic images, any suggestions on what parameters(lr, steps, num class images) might be of help to me, or anything about the dataset... Thanks in advance šš
Hey is there a place to contact you? Would love to commission you in creating a custom model for me š
What are the main differences between newAge and newEra and V1?
Check the version Descriptions
You have to click the "About this version" to see it.
@epinikionĀ Was hoping for a bit more detail than that .... obviously, I can see what is here on the Civitai page, but thank you for making the models.
@EricRollei21Ā itās a process of refining and slightly adding stuff, to get better results. Not so much to say, sorry š
I have tried several models to represent a market scene in ancient Rome, and I have to say that this is the one that best captures the idea of the prompt.
Naturally, without careful inpainting of hands and faces, no stable diffusion 1.5 model can achieve error-free results at such low resolutions, but this model is the one I disliked the least in terms of faces and limbs of the 4 or 5 I have tried for a prompt that I consider complicated for the current level of image generation models without adding loras, inpainting or any external retouching.