Really cool #article on writing #avx #accelerated #code https://www.codeproject.com/Articles/874.. Screw writing #libraries manually to support this though, it would take an insane amount of effort. I'm very happy to let somebody else write #optimisation functions for this stuff.
#intel have just #accelerated their #raytracing #performance on their #gpu 's with a *single line of code* that #allocates #memory #locally , rather than on the #system https://www.phoronix.com/news/Intel-Vulk.. The change increases performance by *100%!*