From 76103d4b057a0787d4f310303896fd832b6438ee Mon Sep 17 00:00:00 2001 From: Juan Nunez-Iglesias Date: Fri, 19 Dec 2014 11:55:00 +1100 Subject: [PATCH] Define 'prefetching' for users --- doc/source/user_guide/numpy_images.txt | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/doc/source/user_guide/numpy_images.txt b/doc/source/user_guide/numpy_images.txt index 188218c9..c4f18450 100644 --- a/doc/source/user_guide/numpy_images.txt +++ b/doc/source/user_guide/numpy_images.txt @@ -156,8 +156,9 @@ Notes on array order Although the labeling of the axes seems arbitrary, it can have a significant effect on speed of operations. This is because modern processors never retrieve just one item from memory, but rather a -whole chunk of adjacent items. Therefore, processing elements that are -adjacent in memory one after the other is faster than processing them +whole chunk of adjacent items. (This is called prefetching.) +Therefore, processing elements that are +next to each other in memory is faster than processing them in a different order, even if the number of operations is the same: >>> def in_order_multiply(arr, scalar):