Even at a denoise of 1 the base image will still affect the generation. Notice how the first few generations match the composition of the input image? It doesn't change until you prompt for long shadows. If you swap to an empty latent you will then get more random generations.
Very helpful and beautifully explained. Thanks!
thank you for the easy to follow video. A lot of great advice
Even at a denoise of 1 the base image will still affect the generation. Notice how the first few generations match the composition of the input image? It doesn't change until you prompt for long shadows. If you swap to an empty latent you will then get more random generations.
Yes I realise this. the video is intended of beginners. The image has very little input at 1.