My method was evolutionary, put too simply - something not completely undiscussed even at the time and it's undoubtedly happening in one way or another now. I a

My method was evolutionary, put too simply - something not completely undiscussed even at the time and it's undoubtedly happening in one way or another now. I also designed a method for parallel processing dedicated to neural nets that seemed much more efficient than just "more processors", I've no idea whether NVIDIA et al have caught up with design that or passed it by now.