Robert

Yes, I think it's the same as #61, but we can keep this issue open instead. I also see it now and then, and have no idea what's happening.

That's curious... It works for me. Some things that would be useful to nail this down: - does it happen with all browsers? what browser and OS are you using?...

Might be fixed by using `encodeBorderQues` instead of `encodeAreaRoom`, and bumping `filever` for those types.

As one further datapoint, Firefox 71 on macos seems to be fine.

Another set of "ja"/"en" screenshots on macos / retina / chrome due to @montelucci  

Strangely I can't reproduce this anymore with Firefox. I can't imagine I fixed it inadvertently... In Safari I still see this, and also on pzv.jp with Firefox. Which browser are...

There seems to be more involved here than the precise puzzle height (zoom, window size, etc.). I can reproduce the problem by opening a tall puzzle and calling ``` ui.puzzle.painter.resizeCanvasByCellSize(47);...

For reference, jQuery's pointer events polyfill seems to do some manual deduplication here: https://github.com/jquery/PEP/blob/master/src/touch.js#L11 So it seems likely that the emulated mouse event can't easily be disabled.

Thanks for catching this! > Judging from the ticket and the error line (34 vs. 35), the two runs are slightly different: The test has changed in the meantime, so...

@andreasabel Do you have any info on how frequent this is?