Legit!
Hence why I try to create community for specific things (like selfhosted, or /anime and /manga (Thank you to @xdd for taking that one ๐). There is literally no need for /anime2 for example. Makes things cleaner and we have a much better way to post/discuss etc.
Besides, i rather go to /anime than having to open /animeX (multiple tabs) to browse the same communities.
Heck yeah, never tried doing it this way. will have to see what i can do with this :D Thanks
Correct! you want to do it in txt2img first (render really good image). only THEN do you wanna upscale it (not via img2img as it will take a shit ton of time and change the image slightly). u want to do it via extras tab then, and then choosing a method to upscale. i can tell you exactly what you need, how to do it in future if u want. go with you step by step
first 2 strike me as emma watson for some odd reason.
Going with what pinecone mentioned here, this is what i do most of the time. like 9.99/10 of the time.
Basically, hires fix > base resolution of 512x768 (by upscaling with 2, which will generate an end image of 1024x1536).
then u can move that gorgeous (but tiny resolution) image to your extras, to then get 5120 x 2880 or whatever resolution
PS: a cool method to use for horizontal images (like wallpapers) instead of vertical like phone wallpapers, is to switch the values. so instead of 512x768 u can do 768x512
If i recall you are talking about denoise strengh
You can always try upscaling without re-rendering the whole image. It will stay the same, but upscaled to a higher resolution. This is the method i use to upscale to big resolution.
example:
The above image has a resolution of: 5120 x 2880
Initial render was 512 x 768. So it keeps the whole image intact, no changes, but does the upscale on it. Mind you, any imperfections will stay in this method.
If you truely want to upscale with a re-render you can expect the image to slightly change. I never had luck keeping an image to stay the exact same when upscaling using img2img
I got you!!!
spoiler
Model: Rev (as you expected, using fantasy embeding)
Denoising: 0.55
Clip: 2
Hires upscaling: 2
upscaler: latent
size: 512x768
seed: 1070493286
CFG: 8.5
Sampler: Euler a
Steps: 40
~~Upscaled 4x (so 1024x1536 x 4)~~ See below [1]
Prompt: Beautiful, best quality, extremely detailed face, perfect lighting, fullbody, from above, bare feet, horror, horror fantasy.
Negatives: worst quality, low quality, monochrome, zombie, animal ears, tail, pointy ears, rabbit ears, dog ears, cat ears, watermark, username, patreon username, patreon logo, bad artist, bad hands
[1] Damn, i realized I didn't upscale it right. Here is a better version:
Original size: 2MB
New upscaled: 34MB
I just downloaded the source and changed variables around. Working on adding an image now, which is giving me a pain in the ass to position xD coz i have to rebuild the project over and over to test.
Highly recommend building outside of docker if u want a quick test. I chose to stick with docker, which means every time i gotta create a new docker image just to see my changes, try to fix, repeat, etc.
Lemme know if u need help.
Nice :3 I use the same server, just changed a few things and different theme etc.
Dude... Dude wtf