Just as we currently have a simulator to check whether a plan will exceed motor limits, it would be useful to have simulator to estimate the required storage for a scan, working from the number of exposures and the shapes of the images produced by the detector. Producing a precise estimate could become challenging, factoring in compression ratios and so on, but even a worse-case upper bound would be useful.