General GUI flakiness

[daktaklakpak]

I submitted a mantis bug for the first bit I found on this, but I don't want to spam over there for something that is in placeholder status.

Two issues I'm having with it:

First, GUI response seems to be very sluggish.  Tooltips are slow to pop up, vanish, or change when hovering.  Clicking buttons has noticeable lag before any response (pause/unpause, turning FRD on or off, those kind of things).

Second, if I maximize the game window, many of the buttons are winding up laid out on top of each other.  I'm running on a 1440P monitor.  I can submit a screenshot if needed.

[x4000]

daktaklakpak: I've noticed the sluggishness of the buttons, too.

Keith: that might mean we need a 20 cycles per second sim after all.  Or maybe it's being submitted too many turns down the pike?  I'm not sure, but it seems like way more of a delay than just one cycle.  I'm not sure if it's doing batches of cycles per network send, but that might be the cause of that issue even in non-networked play.  In that event we could keep 10 cycles per second but just make 1 cycle per batch and probably have it feeling a lot better, or very close to AI War Classic.  Still might need 20 cycles per second in the end, not sure.

Second, if I maximize the game window, many of the buttons are winding up laid out on top of each other.  I'm running on a 1440P monitor.  I can submit a screenshot if needed.

Basically if you resize your window at all during play, it doesn't recalculate all the GUI offsets.  Gets me often, too.  That's on Keith's list, not sure at what priority.

[keith.lamothe]

You can cause the GUI to catch up with the resize by pressing Ctrl+F10 (reloads all external stuff). But yes, that is on my list to fix.

On the general responsiveness, it is slow. Right now it should be about a half second delay (with the later intent being more like 250 or 300 milliseconds), but I think something else is hitting it (because it feels like way more than half a second) and I need to investigate that before messing with the timing. Certainly the difference between 10 and 20 cycles per second would not account for it.