You are right, thoses FGversion agnostic glitches is what we get too.
The logic is pretty we'll explained in the wiki
http://wiki.flightgear.org/Wingflexer There are some parameters to tweak to get estimations and then watch the visual results.
From FGWiki: 777 wing in turbulence show about 2-3 Hz. (I didn't research 787 videos).
So maybe with some tweeks the 787 wing flex could be more credible. I wonder if the 777 WF is working. Also wish to know how the WF is performing on YASIM.
Although the math problem is solved., it will need a control protection algo on the run-time to avoid the computacional cascade loop that collapses the machine, ej with metar changes.
Math done. It will required a FG software eng hacker or similar to tackle the issue.
If anyone could tweak and test the787 and any Yasim WF model getting same bad results, that could be a good starting point to scale up the issue with a ticket in SF.
29.92 for 787 in Merlion livery.