I'm doing some parameter fine tuning with the help of realtime preview. Changing any parameter causes the process to be automatically applied to the preview, which is the point of this tool. But it has occurred to me that, when evaluating the effect of a checkbox, it's quite silly to recalculate the process output everytime since I'm just alterating between this set of parameters and that other set.
That lead me to think about some sort of hashing of all the parameters of a process, and assigning the resulting realtime preview to it. A list of the N more recent hashes and their associated results would be kept in memory and, when the user happens to choose an already seen set of parameters, no calculating would be needed at all, just fetch the stored result and show it in the realtime preview window.
Does this make sense?
