From 24baf34d381a3a4febc464780ab1c06e738008e8 Mon Sep 17 00:00:00 2001 From: lkcl Date: Fri, 29 Oct 2021 12:19:17 +0100 Subject: [PATCH] --- 3d_gpu/architecture/dynamic_simd/shape.mdwn | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/3d_gpu/architecture/dynamic_simd/shape.mdwn b/3d_gpu/architecture/dynamic_simd/shape.mdwn index 1497a4d50..8183e844f 100644 --- a/3d_gpu/architecture/dynamic_simd/shape.mdwn +++ b/3d_gpu/architecture/dynamic_simd/shape.mdwn @@ -457,7 +457,7 @@ used on both LHS and RHS, it is effectively an identical case to when the RHS is an integer, because the fixed_width of the RHS is itself an integer. -A special case of this also occurs in instances where all elwidths +For certain operators (mul, shift) a special case of this also occurs in instances where all elwidths in all partitions on either LHS or RHS are identical: element widths {3,3,3,3} for example. Under such special circumstances, multiply would function correctly even for dual-priority, due to uniform scaling. -- 2.30.2