A conversation has arisen in the Roman PIT image simulation channel about a potential issue with how the Poisson random seeds are handled in the roman_imsim framework. It can result in coherent biases in the realized flux across the images, especially at lower flux values. We should watch the resolution of that issue, as if there is a bug there, we will get hit by it too if we don't apply a fix, since we're using the same framework.
A conversation has arisen in the Roman PIT image simulation channel about a potential issue with how the Poisson random seeds are handled in the
roman_imsim
framework. It can result in coherent biases in the realized flux across the images, especially at lower flux values. We should watch the resolution of that issue, as if there is a bug there, we will get hit by it too if we don't apply a fix, since we're using the same framework.