Does that huge negative prompt actually work? I found that I usually get better results if I keep the negative prompt shorter and turn up CFG scale to 8 or 9.
I also found that a high CFG scale is good at showing you which prompts work together and which ones don’t. If you crank CFG scale to 15-20 (which will naturally produce some abominations) but find that it still ignores part of your prompt, change that part specifically because all it does is confuse the model.
It is an interesting question. I am not sure. The style is from a YouTube channel I followed to learn about automatic1111.
I included the expanded version of the style for duplication purposes. I get better results with it on and the sfw style certainly keeps the model behaved.
I have also had very good results without it. It seems that just casting a wide with a few batches will yield at least one or two useful images, regardless of using the styles.
I’ve been running SD on AMD GPU and Linux since more or less the beginning. It’s been smooth sailing all the way. Not nearly as fast as some equally expensive RTX cards. But, it is what is is.
Awesome, I am still finding my way and am happy if I simply don't crash. I don't have a frame of reference to compare to a Nvidia card for this, but it does seem like we have a little more work in getting things smooth with the AMD cards. I can't say that my speed is terrible. Most renders finish in reasonable time. I am simply amazed that we can do this on consumer grade hardware.
Indeed. I’m in complete awe by this technology. It’s an amazing pass-time that tickles the creative side. As for getting an idea of how different cards and system compare, you can check out vladmandic.github.io/…/benchmark.html
I also have an 6800 XT, and the performance on that particular benchmark is around 9 it/s. Something like this looks to be a rough indication.
I run the tests out of interest. I am leaving some performance on the table due to my launch options, but I need those to avoid to many out of memory and other errors.
I managed to find an extra iteration or two without sacrificing to much stability.
6.06 / 7.59 / 9.11
Running with Doggettx selected as the optimiser in the optimiser config inside automatic1111.
I installed the google perftools as suggested in this thread
"sudo apt install libgoogle-perftools-dev"
And then added the following memory management options as suggested in this thread
by exporting: "export PYTORCH_CUDA_ALLOC_CONF=garbage_collection_threshold:0.6,max_split_size_mb:128"
Made the following changes to my web-user.sh
Uncommented the command line options as follows: export COMMANDLINE_ARGS="--medvram --upcast-sampling"
And added the following lines to the end of the file
No problem, backup and be careful. I think AMD still has a lot that can be done in the rocm drivers themselves. There should be gains left to make. Nvidia is not helping with their pricing either. Which should see more users on AMD. Hopefully better support for us. I am pleasantly surprised by what the card can do. I got it for gaming at 1440p, where it was the best bang for buck. The AI stuff is a cool bonus.
I am on Windows, automatic1111 with directML and rendering is pretty fast. 7700x, 6750xt and 32GB at 4800mhz and a couple of m.2 drives. No xformers tho and some problems when upscaling in txt2image, but it renders prompts with default settings in 10 to 15 seconds. Fast enough for me. AMD has updated their Adrenalin drivers lately to have better directML performance.
Some things can take some time or aren't supported on AMD, but it's surely faster then my rtx 1070 and 1080 rigs wich performed adequately, except with training.
For sure. It works, especially if you use Shark with the experimental driver, but the speed difference was an order of magnitude for the rocm compiled driver on linux. I am already needing more card though. The 6800xt 16gb ram is not enough.I am running on medium ram settings. I hear rocm support for windows is coming soon, so that will be interesting as well. There were some rumours earlier this year.
I am working my way there. I am interested in the gaming possibilities. NPC dialogue and so on. But I wanted to get the environment working first. I found more guides for stable diffusion. Now I can venture deeper knowing that rocm is working.
Not yet, but I probably will in future. There is so much going on at the moment in the field of deep learning. Currently I try to focus on stable diffusion, since it's open source and gives very impressive results.
StableDiffusion
Hot