| 04:07.38 | *** join/#brlcad CIA-5 (i=cia@cia.navi.cx) |
| 05:18.29 | *** join/#brlcad CIA-5 (i=cia@cia.navi.cx) |
| 07:33.36 | *** join/#brlcad CIA-5 (i=cia@cia.navi.cx) |
| 07:54.30 | *** join/#brlcad CIA-5 (i=cia@cia.navi.cx) |
| 08:01.42 | *** join/#brlcad clock_ (n=clock@zux221-122-143.adsl.green.ch) |
| 08:26.33 | *** join/#brlcad CIA-5 (i=cia@cia.navi.cx) |
| 08:42.35 | *** join/#brlcad CIA-5 (i=cia@cia.navi.cx) |
| 08:49.14 | *** join/#brlcad CIA-5 (i=cia@cia.navi.cx) |
| 09:22.44 | *** join/#brlcad CIA-5 (i=cia@cia.navi.cx) |
| 10:14.54 | *** join/#brlcad CIA-5 (i=cia@cia.navi.cx) |
| 12:10.18 | *** join/#brlcad docelic (i=docelic@ri01-212.dialin.iskon.hr) |
| 13:14.05 | *** join/#brlcad docelic (i=docelic@ri01-226.dialin.iskon.hr) |
| 13:56.53 | ``Erik | =/det |
| 15:20.50 | brlcad | ^ad does the same in fewer keystrokes |
| 15:29.40 | *** join/#brlcad docelic (i=docelic@ri02-075.dialin.iskon.hr) |
| 15:31.30 | ``Erik | ^ad is screen, /det is bx, and bx doesn't cope with sigwinch right |
| 15:37.21 | ``Erik | in scheme! w000 |
| 17:21.06 | Maloeran | Adding that to the list, hum? You have an OS and a raytracer to write in Scheme already, or so you said :) |
| 17:22.48 | Maloeran | Erik, can I assume floats are always IEEE 32 bits floats, and double IEEE 64 bits? Got to play with the binary representation to avoid internal rounding and approximations |
| 17:23.24 | Maloeran | The prep should be 2-3 times faster apparently |
| 17:46.26 | ``Erik | um, that's probably safe-ish, but we can test that using autoconf shtuffs |
| 17:57.17 | Maloeran | Yes, it just means to write a fallback path if we don't have IEEE floats |
| 17:57.32 | Maloeran | means +"I have" to write |
| 21:21.50 | *** join/#brlcad docelic (i=docelic@ri02-001.dialin.iskon.hr) |
| 22:39.07 | CIA-5 | BRL-CAD: 03mjgillich * 10brlcad/configure.ac: added check for C++ compiler and override default flags to nothing |
| 23:14.29 | CIA-5 | BRL-CAD: 03mjgillich * 10brlcad/src/external/Cubit/g-sat.cxx: BRL-CAD to ACIS (export) convertor for importing to CUBIT (1st version) -- needs review and cleanup but fully functional |
| 23:19.26 | CIA-5 | BRL-CAD: 03mjgillich * 10brlcad/src/external/Cubit/Makefile.am: initial Makefile for g-sat convertor. Has lib paths hard-coded, needs work to be configurable. |
| 23:28.52 | *** join/#brlcad docelic (i=docelic@ri01-156.dialin.iskon.hr) |
| 23:37.18 | Maloeran | Erik, feel free to try the threaded prep again. Fixed a minor potential glitch but I have doubts that was it |
| 23:37.40 | Maloeran | Oh, and the prep is now twice as fast... with a 2% graph quality loss I don't explain yet |
| 23:43.50 | ``Erik | don't, or can't? |
| 23:50.01 | Maloeran | Sorry, I mean I can't explain it yet, I just need to sleep on it |
| 23:52.11 | Maloeran | Any recommendation of LCD monitor brand for a gamer? ( Not me, christmas present ) |
| 23:58.11 | brlcad | ``Erik: we do it because having unset cflags is no more arbitrary than using -O2 -g |
| 23:59.14 | brlcad | he didn't add the cflags one -- that was already there, I added it for that very reason. he overrides cxxflags following example |