chiark / gitweb /
moebius2.git
9 years agobetter with more bendingness costs master
Ian Jackson [Sat, 25 Oct 2008 00:16:31 +0000 (01:16 +0100)]
better with more bendingness costs

9 years agoand the stop epsilon, from size hovering around 7.something e-4
Ian Jackson [Fri, 24 Oct 2008 18:56:00 +0000 (19:56 +0100)]
and the stop epsilon, from size hovering around 7.something e-4

9 years agoa good 246
Ian Jackson [Fri, 24 Oct 2008 18:52:50 +0000 (19:52 +0100)]
a good 246

9 years agosome history files; working on 246
Ian Jackson [Tue, 21 Oct 2008 00:08:29 +0000 (01:08 +0100)]
some history files; working on 246

9 years agoget rim twist right
Ian Jackson [Tue, 21 Oct 2008 00:00:19 +0000 (01:00 +0100)]
get rim twist right

9 years agoscale 125 seems to work ish although could be more bowl-like
Ian Jackson [Mon, 20 Oct 2008 23:57:23 +0000 (00:57 +0100)]
scale 125 seems to work ish although could be more bowl-like

9 years agofixed XBITS==4
Ian Jackson [Mon, 20 Oct 2008 00:18:32 +0000 (01:18 +0100)]
fixed XBITS==4

9 years agochanged powers on vertex displacements -- tweaked parameters for XBITS==5
Ian Jackson [Mon, 20 Oct 2008 00:11:43 +0000 (01:11 +0100)]
changed powers on vertex displacements  -- tweaked parameters for XBITS==5

9 years agowip before increase power of bendings
Ian Jackson [Sun, 19 Oct 2008 23:38:16 +0000 (00:38 +0100)]
wip before increase power of bendings

9 years agobefore abandon this version of rim twist which is nonsense
Ian Jackson [Sun, 19 Oct 2008 22:13:27 +0000 (23:13 +0100)]
before abandon this version of rim twist which is nonsense

9 years agomerge
Ian Jackson [Sun, 19 Oct 2008 21:57:45 +0000 (22:57 +0100)]
merge

9 years agonew rim_twist_cost as yet unused
Ian Jackson [Sun, 19 Oct 2008 21:57:14 +0000 (22:57 +0100)]
new rim_twist_cost as yet unused

9 years agoMerge
Ian Jackson [Sun, 19 Oct 2008 21:56:24 +0000 (22:56 +0100)]
Merge

9 years agobugfixes to view-cyclicly
Ian Jackson [Sun, 19 Oct 2008 21:56:04 +0000 (22:56 +0100)]
bugfixes to view-cyclicly

9 years agowip energy functions
Ian Jackson [Sun, 19 Oct 2008 21:37:09 +0000 (22:37 +0100)]
wip energy functions

9 years agobuild more stuff
Ian Jackson [Sun, 19 Oct 2008 21:36:58 +0000 (22:36 +0100)]
build more stuff

9 years agoshow title
Ian Jackson [Sun, 19 Oct 2008 18:25:34 +0000 (19:25 +0100)]
show title

9 years agorunes to go with new 33/64 calculations with real vertex displacement cost
Ian Jackson [Sun, 12 Oct 2008 19:42:21 +0000 (20:42 +0100)]
runes to go with new 33/64 calculations with real vertex displacement cost

9 years agodo not remove *.cfm on make clean; do remove .alternately_*
Ian Jackson [Sun, 12 Oct 2008 19:41:57 +0000 (20:41 +0100)]
do not remove *.cfm on make clean; do remove .alternately_*

9 years agoreally edgewise vertex displacement - new calculations, experimental
Ian Jackson [Sun, 12 Oct 2008 19:39:51 +0000 (20:39 +0100)]
really edgewise vertex displacement - new calculations, experimental

9 years agomerge
Ian Jackson [Thu, 9 Oct 2008 23:32:21 +0000 (00:32 +0100)]
merge

9 years agomerge
Ian Jackson [Thu, 9 Oct 2008 23:32:08 +0000 (00:32 +0100)]
merge

9 years agomerge
Ian Jackson [Thu, 9 Oct 2008 23:31:46 +0000 (00:31 +0100)]
merge

9 years agoattempts to fix 125; really need to combine edge length variation and bending costs
Ian Jackson [Thu, 9 Oct 2008 23:30:13 +0000 (00:30 +0100)]
attempts to fix 125; really need to combine edge length variation and bending costs

9 years agoview-cyclicly script - has bug in generation of .alternately filename
Ian Jackson [Wed, 8 Oct 2008 23:31:05 +0000 (00:31 +0100)]
view-cyclicly script - has bug in generation of .alternately filename

9 years agostopping point for 64
Ian Jackson [Wed, 8 Oct 2008 23:24:52 +0000 (00:24 +0100)]
stopping point for 64

9 years agoadjusted costs for 64
Ian Jackson [Wed, 8 Oct 2008 22:07:27 +0000 (23:07 +0100)]
adjusted costs for 64

9 years agoprimer does not need mgraph_prepare
Ian Jackson [Wed, 8 Oct 2008 00:17:25 +0000 (01:17 +0100)]
primer does not need mgraph_prepare

9 years agoallow compiling spuriously with XBITS>=6
Ian Jackson [Wed, 8 Oct 2008 00:16:36 +0000 (01:16 +0100)]
allow compiling spuriously with XBITS>=6

9 years agomemoise edge_end2 now performance is good
Ian Jackson [Tue, 7 Oct 2008 23:58:24 +0000 (00:58 +0100)]
memoise edge_end2 now performance is good

9 years agouse -O3
Ian Jackson [Tue, 7 Oct 2008 23:47:24 +0000 (00:47 +0100)]
use -O3

9 years agoactually do precomputations!
Ian Jackson [Tue, 7 Oct 2008 23:46:56 +0000 (00:46 +0100)]
actually do precomputations!

9 years agoon SIGINT, do not exit 0
Ian Jackson [Tue, 7 Oct 2008 21:15:48 +0000 (22:15 +0100)]
on SIGINT, do not exit 0

9 years agoget rid of debugging for checking OUTER iteration; leave SIGINT handler and fix to...
Ian Jackson [Tue, 7 Oct 2008 20:55:16 +0000 (21:55 +0100)]
get rid of debugging for checking OUTER iteration; leave SIGINT handler and fix to OUTER

9 years agomachinery for checking that our OUTER parallel iteration works - before we delete it
Ian Jackson [Tue, 7 Oct 2008 20:53:25 +0000 (21:53 +0100)]
machinery for checking that our OUTER parallel iteration works - before we delete it

9 years agodo not use gsl_blas_dnrm2; instead just do addition and sqrt
Ian Jackson [Mon, 29 Sep 2008 00:33:57 +0000 (01:33 +0100)]
do not use gsl_blas_dnrm2; instead just do addition and sqrt

9 years agodo not print size every time(!)
Ian Jackson [Sun, 28 Sep 2008 23:14:20 +0000 (00:14 +0100)]
do not print size every time(!)

9 years agodo not bother with barriers and stuff if nprocessors=1
Ian Jackson [Sun, 28 Sep 2008 23:13:52 +0000 (00:13 +0100)]
do not bother with barriers and stuff if nprocessors=1

9 years agox.gdb created
Ian Jackson [Sun, 28 Sep 2008 13:12:32 +0000 (14:12 +0100)]
x.gdb created

9 years agoapparently-useless patch
Ian Jackson [Sun, 28 Sep 2008 13:12:20 +0000 (14:12 +0100)]
apparently-useless patch

9 years agofake plastic imitation weights for XBITS==5
Ian Jackson [Sun, 28 Sep 2008 12:52:00 +0000 (13:52 +0100)]
fake plastic imitation weights for XBITS==5

9 years agoset threads_started
Ian Jackson [Sun, 28 Sep 2008 12:25:57 +0000 (13:25 +0100)]
set threads_started

9 years agouse barriers rather than recreating threads
Ian Jackson [Sun, 28 Sep 2008 12:06:13 +0000 (13:06 +0100)]
use barriers rather than recreating threads

9 years agodo not have threads at all if NPROCESSORS=1
Ian Jackson [Sun, 28 Sep 2008 11:11:27 +0000 (12:11 +0100)]
do not have threads at all if NPROCESSORS=1

9 years agofix threading bugs and arrangements
Ian Jackson [Sat, 27 Sep 2008 23:43:00 +0000 (00:43 +0100)]
fix threading bugs and arrangements

9 years agoadd .nprocessors.make to bzrignore
Ian Jackson [Sat, 27 Sep 2008 23:21:31 +0000 (00:21 +0100)]
add .nprocessors.make to bzrignore

9 years agoMerge multiprocessor support; fix up Makefile runes
Ian Jackson [Sat, 27 Sep 2008 23:12:47 +0000 (00:12 +0100)]
Merge multiprocessor support; fix up Makefile runes

9 years agofix broken indent change
Ian Jackson [Sat, 27 Sep 2008 22:47:55 +0000 (23:47 +0100)]
fix broken indent change

9 years agoparallel thing compiles now
Ian Jackson [Sat, 27 Sep 2008 15:26:20 +0000 (16:26 +0100)]
parallel thing compiles now

9 years agoadd nprocessors to bzrignore
Ian Jackson [Sat, 27 Sep 2008 15:09:22 +0000 (16:09 +0100)]
add nprocessors to bzrignore

9 years agoget flushoutput order right
Ian Jackson [Sat, 27 Sep 2008 15:09:02 +0000 (16:09 +0100)]
get flushoutput order right

9 years agouse nprocessors to set -DNPROCESSORS
Ian Jackson [Sat, 27 Sep 2008 15:08:52 +0000 (16:08 +0100)]
use nprocessors to set -DNPROCESSORS

9 years agoadd nprocessors utility
Ian Jackson [Sat, 27 Sep 2008 15:02:47 +0000 (16:02 +0100)]
add nprocessors utility

9 years agobest so far perhaps
Ian Jackson [Fri, 22 Feb 2008 22:33:15 +0000 (22:33 +0000)]
best so far perhaps

9 years agodone circumcircle but it doesn't work so well
Ian Jackson [Thu, 21 Feb 2008 19:41:27 +0000 (19:41 +0000)]
done circumcircle but it doesn't work so well

9 years agowip new energy functions
Ian Jackson [Thu, 21 Feb 2008 19:12:17 +0000 (19:12 +0000)]
wip new energy functions

9 years agoanother rune
Ian Jackson [Mon, 18 Feb 2008 01:19:27 +0000 (01:19 +0000)]
another rune

9 years agoturn it the right way up
Ian Jackson [Mon, 18 Feb 2008 01:11:39 +0000 (01:11 +0000)]
turn it the right way up

9 years agofix up rim join mistakes
Ian Jackson [Mon, 18 Feb 2008 01:03:51 +0000 (01:03 +0000)]
fix up rim join mistakes

9 years agofewer output facets - ignore ABs - this commit is an EXPERIMENT which may be reverted
Ian Jackson [Mon, 18 Feb 2008 00:35:15 +0000 (00:35 +0000)]
fewer output facets - ignore ABs - this commit is an EXPERIMENT which may be reverted

9 years agoreport on degenerate facets
Ian Jackson [Mon, 18 Feb 2008 00:34:38 +0000 (00:34 +0000)]
report on degenerate facets

9 years agofix diagram
Ian Jackson [Mon, 18 Feb 2008 00:07:14 +0000 (00:07 +0000)]
fix diagram

9 years agonow right except for at join
Ian Jackson [Sun, 17 Feb 2008 19:43:37 +0000 (19:43 +0000)]
now right except for at join

9 years agocompiles and most output facets are right; needs debugging
Ian Jackson [Sun, 17 Feb 2008 19:28:08 +0000 (19:28 +0000)]
compiles and most output facets are right; needs debugging

9 years agominor fixes to output
Ian Jackson [Sun, 17 Feb 2008 17:18:48 +0000 (17:18 +0000)]
minor fixes to output

9 years agooutput now compiles
Ian Jackson [Sun, 17 Feb 2008 17:14:07 +0000 (17:14 +0000)]
output now compiles

9 years agowip make output (STL generator) compile
Ian Jackson [Sun, 17 Feb 2008 13:18:15 +0000 (13:18 +0000)]
wip make output (STL generator) compile

9 years agodownloaded STL file format spec from WP
Ian Jackson [Sat, 16 Feb 2008 11:05:28 +0000 (11:05 +0000)]
downloaded STL file format spec from WP

9 years agomakefile representing current best effort
Ian Jackson [Sat, 19 Jan 2008 12:59:13 +0000 (12:59 +0000)]
makefile representing current best effort

9 years agocan interpolate wildly
Ian Jackson [Sat, 19 Jan 2008 11:49:19 +0000 (11:49 +0000)]
can interpolate wildly

9 years agointerpolation more flexible
Ian Jackson [Sat, 19 Jan 2008 11:10:15 +0000 (11:10 +0000)]
interpolation more flexible

9 years agosmaller minimisations
Ian Jackson [Sat, 19 Jan 2008 01:34:58 +0000 (01:34 +0000)]
smaller minimisations

9 years agobest interpolation so far
Ian Jackson [Sat, 19 Jan 2008 01:15:32 +0000 (01:15 +0000)]
best interpolation so far

9 years agolinear interpolation works; view presets
Ian Jackson [Sat, 19 Jan 2008 00:47:38 +0000 (00:47 +0000)]
linear interpolation works; view presets

9 years agoakima no good either but check it in anyway
Ian Jackson [Fri, 18 Jan 2008 23:23:44 +0000 (23:23 +0000)]
akima no good either but check it in anyway

9 years agointerpolates but not very well
Ian Jackson [Fri, 18 Jan 2008 23:22:44 +0000 (23:22 +0000)]
interpolates but not very well

9 years agobetter machinery; before make view interpolate-compatible
Ian Jackson [Fri, 18 Jan 2008 22:12:53 +0000 (22:12 +0000)]
better machinery; before make view interpolate-compatible

9 years agointerpolate and various sizes build
Ian Jackson [Fri, 18 Jan 2008 20:33:56 +0000 (20:33 +0000)]
interpolate and various sizes build

9 years agoviewdims variable
Ian Jackson [Wed, 16 Jan 2008 21:28:20 +0000 (21:28 +0000)]
viewdims variable

9 years agoview various dimensions
Ian Jackson [Wed, 16 Jan 2008 21:27:37 +0000 (21:27 +0000)]
view various dimensions

9 years agoget rid of makedim multidim experiment
Ian Jackson [Wed, 16 Jan 2008 18:34:08 +0000 (18:34 +0000)]
get rid of makedim multidim experiment

9 years agomakedim multidim experiment
Ian Jackson [Wed, 16 Jan 2008 18:33:23 +0000 (18:33 +0000)]
makedim multidim experiment

9 years agorevert normalisation stuff
Ian Jackson [Wed, 16 Jan 2008 18:11:37 +0000 (18:11 +0000)]
revert normalisation stuff

9 years agobefore abandon normalisation effort
Ian Jackson [Wed, 16 Jan 2008 18:07:02 +0000 (18:07 +0000)]
before abandon normalisation effort

9 years agotidy up and normalise line bending cost
Ian Jackson [Wed, 9 Jan 2008 18:34:59 +0000 (18:34 +0000)]
tidy up and normalise line bending cost

9 years agodo not actually compute vertex areas
Ian Jackson [Sun, 6 Jan 2008 21:50:07 +0000 (21:50 +0000)]
do not actually compute vertex areas

9 years agoquite nice
Ian Jackson [Sun, 6 Jan 2008 20:59:54 +0000 (20:59 +0000)]
quite nice

9 years agoadd noncircular, now it collapses
Ian Jackson [Sun, 6 Jan 2008 20:38:53 +0000 (20:38 +0000)]
add noncircular, now it collapses

9 years agoprinting improvements
Ian Jackson [Sun, 6 Jan 2008 20:36:52 +0000 (20:36 +0000)]
printing improvements

9 years agominor improvements
Ian Jackson [Sun, 6 Jan 2008 20:32:20 +0000 (20:32 +0000)]
minor improvements

9 years agorun at sane speed, turn all on, need to rethink cost function a bit
Ian Jackson [Sun, 6 Jan 2008 18:18:23 +0000 (18:18 +0000)]
run at sane speed, turn all on, need to rethink cost function a bit

9 years agoexploit symmetry
Ian Jackson [Sun, 6 Jan 2008 18:11:21 +0000 (18:11 +0000)]
exploit symmetry

9 years agotuning
Ian Jackson [Fri, 4 Jan 2008 19:21:31 +0000 (19:21 +0000)]
tuning

9 years agoget sign right, but it is still wrong
Ian Jackson [Fri, 4 Jan 2008 19:21:11 +0000 (19:21 +0000)]
get sign right, but it is still wrong

9 years agoallow pausing updates
Ian Jackson [Fri, 4 Jan 2008 19:21:00 +0000 (19:21 +0000)]
allow pausing updates

9 years agonew energy functions
Ian Jackson [Fri, 4 Jan 2008 02:15:46 +0000 (02:15 +0000)]
new energy functions

9 years agoimprovements?
Ian Jackson [Fri, 4 Jan 2008 01:32:24 +0000 (01:32 +0000)]
improvements?

9 years agonew energy calculation but anarres crashes
Ian Jackson [Thu, 3 Jan 2008 20:13:06 +0000 (20:13 +0000)]
new energy calculation but anarres crashes

9 years agowip new graph layout stuff before abandoning
Ian Jackson [Thu, 3 Jan 2008 19:11:04 +0000 (19:11 +0000)]
wip new graph layout stuff before abandoning

9 years agoexperiment with other scalings for shortest path
Ian Jackson [Wed, 2 Jan 2008 21:25:56 +0000 (21:25 +0000)]
experiment with other scalings for shortest path