From 16cf483e43f968b0872f3d827fb858d087145e06 Mon Sep 17 00:00:00 2001 From: lkcl Date: Wed, 11 Sep 2019 01:16:47 +0100 Subject: [PATCH] --- ztrans_proposal.mdwn | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/ztrans_proposal.mdwn b/ztrans_proposal.mdwn index 08cfae28f..ae4d0344c 100644 --- a/ztrans_proposal.mdwn +++ b/ztrans_proposal.mdwn @@ -531,7 +531,11 @@ high-performance or correctly-rounding): ASINH( x ) = ln( x + SQRT(x**2+1)) -# Reciprocal +# Evaluation and commentary + +This section will move later to discussion. + +## Reciprocal Used to be an alias. Some implementors may wish to implement divide as y times recip(x). @@ -539,7 +543,7 @@ Others may have shared hardware for recip and divide, others may not. To avoid penalising one implementor over another, recip stays. -# To evaluate: should LOG be replaced with LOG1P (and EXP with EXPM1)? +## To evaluate: should LOG be replaced with LOG1P (and EXP with EXPM1)? RISC principle says "exclude LOG because it's covered by LOGP1 plus an ADD". Research needed to ensure that implementors are not compromised by such @@ -551,7 +555,7 @@ a decision > ok, they stay in as real opcodes, then. -# ATAN / ATAN2 commentary +## ATAN / ATAN2 commentary Discussion starts here: -- 2.30.2