No way near as bad as I thought it was gonna be.
This video thing needs at least 2 months to be decent, but it's great to get into it to be ready for the video explosion.
They're irrelevant anyway doing those stupid fucking dances. I'd actually like them if they fucking sat still so I could even see their ass properly but instead, they jerk it around like a fucking pinata as if that's fucking arousingWon't anyone think of the tiktok thots!..
| weight 0.8 | weight 0.6 | weight 0.4 |
|
|
|
gif2gif. You can find it in Extensions tab/ Available. Press "load from" .
I think the intricacy is resolution based. Whenever I have a partial shot hips the texture there is out of this world. But the texture disappears the moment I get a full shot of a person. And then I see the texture again in the MrFox's HiRes renders. So, if you want intricate detail down to the pixel, HiRes is where it is at. That's my observation, no idea if shared.
You are probably right, I forgot about resolution, after all this time I still can't understand what is CFG and sampling steps doing to be honest xD
CGF (classifier free guidance) is how much "creative freedom" you give SD and how much you "force" it to adhear to the prompt. If you set it too high the image can get "burnt" but there is a Dynamic Tresholding extension to fix this. There is also an anti burn extension. You can find it in extensions tab/available press "load from". The sample steps is how many attempts SD does to try improve the image. More is not necessarily better. It is better up to a point, then it is changing the composition or add weirdness and artifacs etc. The Hires steps is similar but it has the added function of increasing the quality of details and fidelity etc. Again this is also not something to just crank it to max.
thank you so much, have to try.
For what i've understand it's work like "how much fantasy you need?". let me explain: i work at cfg 11 because is a good value i find work for most prompt i try. but when i do some extreme fantsy promp, especially when you want mix person with animal, i need to lower cfg to 8-7 to get result i want, otherwise i get a person and a animal in image,without mix.
Yes when you use a higher sampling resolution, SD has a problem handling this and you get monstrosities.
That's golden info for me, I never knew what the difference was, and never knew how to generate more than 1 image of the same seed.
It increases the vram usage though. So it's not an option if we are trying to use as high resolution as possible. It's a shame that it is generating 2 or more at the same time when using batch size. But there are good news. You can use xy plot script instead.
I don't use hi res for txt2img, upscale in img2img phase so it's not a problem for me
Xformer doesn't change quality. The rumor is that it makes the image generation less predictive. In my experience, this is not a thing. If there is an increase of variation it's so small that you can't differentiate it from the normal variation. Meaning repeated generation of the same seed and same settings. Each image will be slightly different. But this will happen regardless if you use xformer or not. Imo more variation is not a bad thing regardless. Yes, do use xformer. Any boost in performance we can get is well needed.Today i installed xformers, and oh boy, this shit bussin'. Went from "out of vram when generating 512x768 with x1.5 hires upscale after 7 minutes" to "ez 512x768 x2 hires upscale in 3 minutes". And tbh after some testing, i see no change in quality whatsoever. And finally i can generate some portrait mode pics
Oh man that sucks. I use it all the time, it's great. I run comparisons very often to weed out the best settings, wich sampler to use and many other scenarios. For instance I made several different iterations of my Lora that I could compare with the same prompt and seed thanks to plot script. I have never had any error specific to this. If nothing else works a simple re-install will likely fix it.
