That's a style thing and entirely dependent on your prompt, not really a render quality issue. I intentionally ask for bloom and glow on my pictures to give them a softer look, since it's more elegant and less like it was taken with a phone camera. But the process I use can do that just the same.
Compare the images I posted and your own of the same concept... My images are a big improvement over them if I might say so. Also I'm only using the scenario and concept you had posted earlier. If you keep moving the goalposts, none of us can help you. If this is the look you are going for, then find out which checkpoint they used to create the images. This is a completely different style and concept all together. You achieve sharpness by using as high resolution as possible from the beginning, depending on hardware and SD. Then do a very good upscale, with further refinement. I suggest that you try out SDXL checkpoints, Fenrisxl for example. This is the ckp I used for my recent Mista Fox, you can find by using the link below.
It is but I wasn't talking about the style, but the clarity and sharpness of details. Two artworks must not share the same style in order to to be comparable in several subterms. I wasn't trying to offend any of you, I just think those pictures are on another level in terms of those two factors.. and as you said yourself earlier, we're all here to learn!
I wanted to try higher but actually SD is starting to shutdown my computer again and again.. as soon as the upscaler kicks in at 3x settings (even though that worked fine earlier with the pics I posted here).
Alas, you are actually talking about style, because the stuff you're talking about doesn't have anything to do with upscaling, which is what lead us to this discussion in the first place. The type of visual you shown in your example pictures are the result of prompts with a focus on sharpness, checkpoints like
No, what lead us to this discussion was this question from me that you yourself quoted earlier:
Clearly I was asking what else I can do to get sharper results, besides upscaling. Your post now has answered this question partially, so still thank you
By the way @ all regarding this issue here,
In all my years (okay, year singular) using Stable Diffusion I have never heard of such a thing. And there is no such thing as generating without a starting seed either. You may not have used the same seed as whatever source prompt you got, but there is always a seed, it's what allows the generation of the initial noise field from which the image is created.
Yeah it seems that was just a lucky shot. Trying further it still crashes with fresh "projects" and also with other upscalers. Hmmm, very annoying.
I blame A1111. I could barely ever make more than a few images with it without it crashing or cease responding for some unknown reason or another. It's just a bad piece of software. ComfyUI has surely a much steeper initial learning curve, but it's rock solid, never freezes, never crashes, and it's faster, and it handles images much, much larger.
Yeah I actually thought of just switchting and trying comfyUI instead.. I like the node-approach too! Since SD is the only thing that causes issues on my computer I also blame A1111 rn. It usually otherwise never crashes, not when I play games or work with heavy editing programs.
There's quite a few things that would cause RAM to get used some of them can easily use over 32gb of ram.
Is there a way to check any of this in the settings? Resource monitor didn't show that my ram went full (I think 5 out of 16gb were in usage iirc). But it definitely happens after the upscaler kicks in and progress slows down. I just don't know why it worked so well earlier but not anymore. I didn't switch between checkpoints or VAE's since earlier.
Only times i've had a1111 freeze is when running it on a 1050 with 2gb of vram and half of that gets used by the OS.
Keep your task manager open and look at the GPU output, It should show you graphs both for the VRAM and shared memory. IF usage for the shared memory goes up when the upscaling starts or towards the end, it's likely what's causing you issues.
If you are limited to 16Gb of ram I will perhaps suggest the obvious thing here. As soon as you can afford it, invest in 2 pair more sticks. I did and it has been tremendous. It will help the rest of your hardware have a longer reach until you can invest in a better GPU. Perhaps this is the reason I'm able to use SDXL checkpoints, not sure of this though. SD 1.5 doesn't like to go as high as 1024px but SDXL does. So you have more detail to work with right from the start. This makes the next step bring way more sharpness and detail than if you started out with 512px. But if SDXL is out of reach try 768x768 first, next is 960x960. Use the one that your pc can deal with. Then use hiresfix and adjust the multiplier until you get out of memory error. Ok now you know your limit within txt2img. Next go to img2img and use SD upscale and see how much you can get from this.
