Same laten images in the batch

[krzysztofkuczek]

HI,
I would like to generate the batch, let say with 4 latent images, which would produce exactly the same image (seed is fixed). It should behave exactly the same like adding single image patch 4 times for processing.
I was trying with "Repeat laten batch" but without success.
I just want to generate "movie frames" based on open pose input. So i have 4 open pose frames passed to a control net, and i would like to generate (in one batch) 4 images with different open pose control applied.
The output should be similar looking images but with different posed character. Exactly the opposite way how my workflow works :)

The another idea is to pass from batch to batch (single image batches) something like "frame number", and increment it. IT would be pointer to open pose processor. Is it possible, storing the int in a file, and then reading ?
Thank you in advance.
Kris

[krzysztofkuczek]

I think, that it's because the way how the function below is called

def prepare_noise(latent_image, seed, noise_inds=None):
    """
    creates random noise given a latent image and a seed.
    optional arg skip can be used to skip and discard x number of noise generations for a given seed
    """

and then called from nodes.py
noise = comfy.sample.prepare_noise(latent_image, seed, batch_inds)

...