Kumar Appaiah

Thank you. This was, indeed, the issue. Kumar

Dear Asymptote users, In the following code, I expect that the resulting PDF has four pages with animated images: import animate; size(3cm); settings.outformat="pdf"; animation a = animation(global=true); draw((0,0)--(1,1)); a.add(); draw((1,0)--(0,1)); a.add(); draw((0,0)--(1,0.5)); a.add(); draw((1,0.5)--(0,1)); a.add(); write(a.pictures.length); // outputs 4 a.movie(BBox(1mm)); However, I end up with only a single-page PDF. What could be the mistake I am making? Thanks. Kumar

Many thanks for the fix! Glad to have found an old bug! :-) Kumar

Dear Asymptote users, I am missing something in my understanding of the following: import graph; path ctft = (-3.8,0.2){(0,0)}..{(0,0)}(0,1){(0,0)}..{(0,0)}(3.8,0.2); write(maxtimes(ctft)); // displays 2 1 draw(ctft); write(maxtimes(ctft)); // displays 2 1 xaxis("$f$",xmin=-4,xmax=4, arrow=Arrow); yaxis("$X(f)$", autorotate=false,arrow=Arrow,ymax=2,p=fontsize(8pt)); write(maxtimes(ctft)); // displays 0 0, WHY?? Why does the last call return 0 and 0? I presume that my assumption that the path properties...

I've uploaded a new version of IT++ to Debian with the above patch. Thanks.

Ok, so removing SSE2_SHUFF and replacing the entry with 0x1bU directly works. I will use this fix in Debian.

Replacing SSE_SHUFF with 27 solves the problem, though I don't know if that is right. This may also be connected: https://stackoverflow.com/questions/48726032/using-a-variable-to-index-a-simd-vector-with-mm256-extract-epi32-intrinsic

Replacing SSE_SHUFF with 27 solves the problem, though I don't know if that is right. This may also be connected: https://stackoverflow.com/questions/48726032/using-a-variable-to-index-a-simd-vector-with-mm256-extract-epi32-intrinsic