Performance of undo/redo when network.size > 60

[dcwhite]

Bad performance when network gets big, as expected. Will need to work hard on optimization.

[dcwhite]

@SCIInstitute/cibc-users @SCIInstitute/scirun-users @SCIInstitute/scirun-developers Another user poll: Has anyone noticed this slowdown? In general, how should we go about provenance in large network files?

[jessdtate]

Provenance is part of it, but there the networks can be quite slow even when freshly loaded. As for provenance, if we could set the number of steps (for undo) in preference, then set the default to 10 or so, that would probably help.

[dcwhite]

Sounds good. I'll put this on the list for 2017.

[github-actions[bot]]

Stale issue message

[dcwhite]

@jessdtate Another one that Qt5 might solve.

[jessdtate]

I just tested this on a network with 64 modules and the undo/redo step took ~ 3 s. It puts all the modules in the same location, like it does on load. Another network (120 modules) took about 6s, and a big network (>200) took about 10 s. Seems faster than qt4, but still a problem

[dcwhite]

Indeed. The implementation sucks.

[github-actions[bot]]

This issue is stale because it has been open 120 days with no activity. Remove stale label or comment or this will be closed in 60 days.

[github-actions[bot]]

This issue is stale because it has been open 240 days with no activity. Remove the stale label or comment, or this will be closed in 60 days.