summaryrefslogtreecommitdiff
path: root/doc/context/sources/general/manuals/luametatex/luametatex.tex
diff options
context:
space:
mode:
authorHans Hagen <pragma@wxs.nl>2020-12-03 20:01:53 +0100
committerContext Git Mirror Bot <phg@phi-gamma.net>2020-12-03 20:01:53 +0100
commit61848324b49c6fe34f9c5dd62a34d2fbfbbb16b8 (patch)
tree2bd5b88839e392589d77e586e62be2d647929e2c /doc/context/sources/general/manuals/luametatex/luametatex.tex
parent9273441a6cb7b02b3336e0a862ee311dbd3653e9 (diff)
downloadcontext-61848324b49c6fe34f9c5dd62a34d2fbfbbb16b8.tar.gz
2020-12-03 18:59:00
Diffstat (limited to 'doc/context/sources/general/manuals/luametatex/luametatex.tex')
-rw-r--r--doc/context/sources/general/manuals/luametatex/luametatex.tex21
1 files changed, 11 insertions, 10 deletions
diff --git a/doc/context/sources/general/manuals/luametatex/luametatex.tex b/doc/context/sources/general/manuals/luametatex/luametatex.tex
index e0e7f1575..a7ece66d8 100644
--- a/doc/context/sources/general/manuals/luametatex/luametatex.tex
+++ b/doc/context/sources/general/manuals/luametatex/luametatex.tex
@@ -45,14 +45,15 @@
% rpi 64 : 26.0 (overclocked)
% amd 10 fitlet : 28.2
-% But speed can differ a bit per compilation (upto .5 seconds maybe due
-% to how compiled code is organized which might influence caching. Who
-% knows ... (or cares). For instance at 20200407 I measured 10.9 seconds
-% after some new low level metapost magic was added but who knows if that
-% was the reason, because mp processing is already fast. A week later,
-% at 20200415, a by then 254 page file took 10.5 seconds, that is, we
-% were at exactly 24 pages per second but after switching to gcc9 it
-% dropped again.
+% But speed can differ a bit per compilation (upto .5 seconds maybe due to how
+% compiled code is organized which might influence caching. Who knows ... (or
+% cares). For instance at 20200407 I measured 10.9 seconds after some new low level
+% metapost magic was added but who knows if that was the reason, because mp
+% processing is already fast. A week later, at 20200415, a by then 254 page file
+% took 10.5 seconds, that is, we were at exactly 24 pages per second but after
+% switching to gcc9 it dropped again. In december 2020, with IPO enabled I crossed
+% the 26 pps barrier and went below 10 seconds but that was also after some further
+% cleanup in \LMTX.
% msvc 1899k (2% slower than mingw)
% msvc /GL 2297k (similar to mingw)
@@ -61,8 +62,8 @@
% msvc /Ob1 1763k (slower than /Ob2)
% clang 2460k (15% slower than mingw)
-% Thanks to sebastian.miele@gmail.com for close reading the manual and
-% sending fixes.
+% Thanks to sebastian.miele@gmail.com for close reading the manual and sending
+% fixes.
\enabletrackers[system.usage=summary]