@pcwalton Is that true? I haven't done a lot, but it seems like most SIMD is more "stuff" from source to binary -- it's just faster.
-
-
-
@gankro SIMD code on architectures with weird SIMD (like x86) requires a lot of algorithm contortions to consume the fewest cycles.
End of conversation
New conversation -
-
-
@pcwalton recently saw a presentation (I think it was an Intel guy) were he was openly wondering if the 2-3x bump is worth the hassle.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.