During acquiring multi frame Signal Imaging by STEM-EELS in DigitalMicrograph, the process time from the frame-end to the next frame is bottle-neck in the view point of fast acquisition. Though I think it is the process-time to make a new 3D box, is there any way to speed up by scripting? I'm thinking there are two ways. One is making 4D box (xyz + frames) before acquiring multi-frame SI, and the other is making new 3D box which is for next frame, during SI acquisition (parallel threading -- this is smarter??). But I don't know how to write it by DM-script. Do you have a any helpful knowledge for my purpose?
There is not really a way to improve on this by scripting, as the script interface only triggers actions that you otherwise trigger via the UI.
The speed delays, however, are not primarily caused by memory allocation of the 3D stack, but by the synchronous nature of data acquisition of multiple detectors (with varying readout speeds) as well as intermediate actions such as drift-measurements.
The best one can do at the moment by scripting is to set up and launch a regular multi-pass acquisition and copy individual data at the "end of frame" event. To my knowledge, this has been done already.