view effect needs optimization

Bug #1557601 reported by oriolpont
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
JessyInk
New
Undecided
Unassigned

Bug Description

I am having a look again at the script after having used it in practice several times. The view effect is very interesting because it can make prezi-like presentations. However, I have noticed that it does not tolerate well heavy use especially if the slide contains complex elements.

Do you have any clues on how to optimize it?

Revision history for this message
oriolpont (oriolpont) wrote :

A workaround that alleviates a bit the heavy load: 50ms-long timesteps are visually enough.

The script enforces time durations and so it should be skipping frames if necessary. But, for some reason, if the target rate is less demanding, it seems that the overall execution is better.

"Pop" effect would benefit from this too, but "fade" effect apparently tolerates a heavier framerate. Would it be perhaps convenient to use different framerates for these?

Revision history for this message
Hannes Hochreiner (hannes-hochreiner) wrote :

Can you provide an example file that would illustrate this behavior, please. I would also be interested in your configuration (os and browser). I just want to make sure I understand it correctly and that the effect is not limited to certain set-ups.

Revision history for this message
oriolpont (oriolpont) wrote :

Hannes, have you successfully used it long/complex/heavy presentations?

Here is one (testing) example of mine:

https://mega.nz/#!0k9BEaSQ!PmKSp_PdLDIywGulTmzcEvynErESPTs-OJAZkKP6k1I

I tested it in both chromium and firefox, in Linux. For the script, I have been using andstein's inkscape-independent mod https://github.com/oriolpont/jessyink-hackish with the script file named jessynew. I however want to discuss the issue here, because we changed nothing of the effects or code around it so it should in principle affect also your version (untested though).

I have looked also at the 2.0 branch, but I see that the effects code is not much changed so, barring the lifted overhead, should not see much optimization in this case.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.