Those bugs that show up only sometimes are often thread-related. They are difficult to reproduce because they depend on very specific conditions, i.e., a thread performing a task a few milliseconds too slow because of high system load. Fingers crossed 
I don’t know i understand this correctly 
I guess you mean pipe related. I haven’t seen any thread (omp) related bugs after some minor fixes for @ralfbrown work on this.
And pixelpipe related crashes are very rare since a few versions i would say (after lots of work on pipe cache, mask distortions, raster masks …)
1 Like
You are certainly right. I haven’t read the whole thread, my comment was just based on the statement that it was hard to reproduce.
(post deleted by author)
Sorry - wrong button.
Having had multiple crashes again - sent a crash report to github - waiting.
Kind Regards
Uwe