00:23.39 |
*** join/#brlcad cad03
(n=48c0dea0@bz.bzflag.bz) |
00:37.19 |
IriX64 |
sdl and python installed, adrt enabled,
started a build lets see. |
00:41.49 |
``Erik |
if you want to be fast, you can cd
src/adrt |
00:41.51 |
``Erik |
and make from there |
00:42.01 |
``Erik |
you can do 'make depends' if you dont' have
libbu built yet |
00:43.50 |
brlcad |
he's not on cvs head so likely run into other
stuff |
00:44.04 |
brlcad |
plus I think I removed the libbu dep |
00:51.28 |
``Erik |
oh |
00:51.31 |
``Erik |
eh? |
00:51.58 |
``Erik |
every main() makes half an attempt to use
bu_getopt() |
00:53.17 |
brlcad |
not any more |
00:53.24 |
brlcad |
i undid that |
00:54.55 |
``Erik |
heh |
00:55.17 |
``Erik |
and here I figured it was going to become more
integrated with BRL-CAD |
01:01.20 |
brlcad |
eventually.. |
01:01.37 |
brlcad |
as soon as any of the libbu compatibility
routines become a problem, it'll be a dep |
01:01.47 |
brlcad |
but to just replace for the sake of replacing
wasn't the intent |
01:02.22 |
brlcad |
as it is, adrt should/might still compile
entirely stand-alone |
01:08.24 |
``Erik |
hm, is it getopt or getopt_long in
there? |
01:08.45 |
brlcad |
just getopt iirc |
01:09.35 |
``Erik |
ah, all the bu_getopt stuff has been removed
and the getopt_long is gaurded |
01:09.36 |
``Erik |
cool |
02:19.24 |
IriX64 |
oooo, somethings missing in
display.c |
02:19.33 |
IriX64 |
all sorts of undefines. |
02:19.40 |
IriX64 |
undefined. |
02:19.48 |
IriX64 |
all _SDL stuff. |
02:20.00 |
brlcad |
er, you have sdl installed? |
02:20.04 |
IriX64 |
yes |
02:20.24 |
brlcad |
presumably one of the first errors is
something about not finding an SDL.h header I bet |
02:20.37 |
brlcad |
might need to manually tweak
CPPFLAGS |
02:21.50 |
IriX64 |
http://www.pastebin.ca/371076 |
02:21.59 |
``Erik |
and perhaps post config.log and the affected
Makefile somewhere we can get to eventually |
02:21.59 |
IriX64 |
its at link time for observer. |
02:22.14 |
``Erik |
hm |
02:22.29 |
``Erik |
where is libSDL.dll ? |
02:22.34 |
``Erik |
(or .so or .lib or whatevr) |
02:22.48 |
IriX64 |
in local lib i think let me check. |
02:23.35 |
IriX64 |
yeah all in there. |
02:23.47 |
IriX64 |
configure found it ok. |
02:24.13 |
``Erik |
hm, in /usr/local/lib ? |
02:24.15 |
``Erik |
what's the filename? |
02:24.50 |
IriX64 |
libsdl.a .la .ddl.a 3 of them. |
02:25.06 |
IriX64 |
err dll.a |
02:26.13 |
IriX64 |
still want config.log and the makefile
``Erik? |
02:26.27 |
``Erik |
all .a or .la, huh, that means static only,
and you were trying to link it dynamically, I think... I'm not sure
on cygwin link semantics |
02:26.29 |
``Erik |
nah |
02:26.44 |
IriX64 |
ill remove shared. |
02:29.01 |
IriX64 |
http://www.pastebin.ca/371084 |
02:29.15 |
IriX64 |
anything you want added/deleted from the
config line? |
02:31.13 |
``Erik |
enable-adrt automatically enable sdl and
python... the flags are more there for defining an unusual
location |
02:31.34 |
``Erik |
disable-symbols? disable-shared? O.o |
02:31.55 |
IriX64 |
--disable-configure :P |
02:32.02 |
``Erik |
disabling debugging and enabling optimizations
are probably useless at this stage |
02:32.05 |
brlcad |
disable shared works, but not so sure about
the symbols one doing anything |
02:32.32 |
IriX64 |
seem to remember seeing it
somewhere. |
02:32.39 |
``Erik |
why disable shared? |
02:32.45 |
IriX64 |
libraries |
02:32.51 |
brlcad |
yeah, you probably should leave off
--enable-optimizations and --disable-debug and --di |
02:32.54 |
``Erik |
just cuz you saw it "somewhere" doesnt' mean
it works with BRL-CAD, homey |
02:32.54 |
brlcad |
sable-runtime-debugging |
02:33.09 |
``Erik |
./configure --help |
02:33.17 |
IriX64 |
:D |
02:33.18 |
brlcad |
disable shared works, and he actually needs it
for cygwin.. some ld problem |
02:33.29 |
``Erik |
and the more flags you give it, the less
'normal' it is, so the less able we are to help... fyi...
:) |
02:33.35 |
IriX64 |
right want to make sure now. |
02:33.50 |
IriX64 |
ty i didnt realize that. |
02:35.06 |
IriX64 |
--disable-bug-reports ;) |
02:35.09 |
brlcad |
./configure --enable-everything --enable-adrt
--disable-shared --prefix=/usr/cygcad should be all you
need |
02:35.12 |
``Erik |
like, if you were using osX.4 with no flags,
brlcad would be able to help you n/p... if you were using fbsd6
with no flags, I'd probably be able to help n/p... that's where we
live, yo... the less like our environments you get, the less we're
in the know |
02:36.05 |
IriX64 |
good to know my nect build will just be
./configure --disable-shared. |
02:36.12 |
IriX64 |
next too. |
02:36.26 |
IriX64 |
liar ;) |
02:36.33 |
brlcad |
i regularly try a variety of flag settings,
and different platforms, but it is true that os x would be one of
the most familiar regardless |
02:36.54 |
IriX64 |
motorola or apple? |
02:37.01 |
``Erik |
<-- does a bit on osX and linux, a tiny bit
on solaris... |
02:37.03 |
``Erik |
uhhhh |
02:37.09 |
``Erik |
do you mean ibm or intel? |
02:37.14 |
IriX64 |
osx |
02:37.29 |
``Erik |
<-- uses both ibm/g[45] and intel/coreduo
macs |
02:37.59 |
``Erik |
most of my dev work is on a quad core opteron
fbsd6.2 machine |
02:38.02 |
IriX64 |
model 3800+ |
02:38.54 |
IriX64 |
we have 4 boxes in this house 2 laptops and 2
desktops, *I'm only "allowed" to use one. |
02:39.12 |
IriX64 |
chilldren you know :) |
02:44.55 |
``Erik |
heh |
02:45.19 |
IriX64 |
btw thats sposed to read "WunderBox"
:) |
02:45.24 |
``Erik |
s/his h/in h/ |
02:49.01 |
IriX64 |
you're "state of the art" then
``Erik. |
02:50.27 |
IriX64 |
oon the other hand the one i'm using has a gig
of main mem and 200gig drive :) |
02:52.07 |
IriX64 |
already used 44gigs. |
02:52.32 |
``Erik |
hrm, all three of my laptops have 2g ram a
pop, and my fbsd box downstairs has around .5tb used |
02:52.45 |
``Erik |
so, uh, put it away, dude :D there's always
someone with a bigger one |
02:52.49 |
IriX64 |
whoa:) |
02:53.26 |
IriX64 |
penvy hah. |
02:58.24 |
IriX64 |
need some of natures wine, bbiab. |
03:07.06 |
Twingy |
oh yea, my PIC chip has 768 bytes of RAM, so
nya! |
03:09.49 |
``Erik |
good old 16f88's |
03:10.13 |
Twingy |
the 18F4620's are quite nice |
03:10.20 |
Twingy |
32kB program, 2kB RAM, 1kB eeprom |
03:13.50 |
IriX64 |
did I really say cursor.c ``Erik? |
06:40.39 |
*** join/#brlcad PrezKennedy
(n=Matthew@c-69-138-68-160.hsd1.md.comcast.net) |
09:32.33 |
*** join/#brlcad docelic
(n=docelic@212.15.185.136) |
09:35.25 |
*** join/#brlcad clock_
(i=clock@84-72-63-213.dclient.hispeed.ch) |
09:38.01 |
clock_ |
win 11 |
10:27.21 |
*** join/#brlcad cad03
(n=55462b6d@bz.bzflag.bz) |
14:27.59 |
*** join/#brlcad lg_
(n=lg_@88.234.13.222) |
15:05.21 |
lg_ |
hi... |
15:16.59 |
louipc |
hello |
15:17.43 |
lg_ |
a, someone alive there ;-) |
15:18.52 |
louipc |
this channel is pretty active, but only at
certain times and I don't think they're regular |
15:19.07 |
``Erik |
yeah, we need to eat more fiber :/ |
15:20.21 |
louipc |
hah |
16:57.43 |
brlcad |
hehe |
17:01.42 |
``Erik |
I'm gonna destroy the toilets at work,
yo |
18:07.26 |
IriX64 |
http://www.pastebin.ca/371833 |
18:15.28 |
IriX64 |
http://www.pastebin.ca/371844
<==== this is the one that gets me :) |
18:17.00 |
IriX64 |
``Erik, it produces MZ files :) |
18:17.21 |
``Erik |
uh huh? |
18:17.34 |
``Erik |
iirc, MZ is old dos executables (like .com
files) and PE is windows stuff |
18:18.06 |
``Erik |
http://en.wikipedia.org/wiki/EXE |
18:18.16 |
Maloeran |
And "PE" stands for "Portable Executable"
amusingly |
18:18.21 |
Maloeran |
Portable to windows and windows |
18:18.42 |
``Erik |
there we go, mz is the old dos executable,
then new dos is NE, then a couple os/2 ones (LE and LX), ... then
PE |
18:19.25 |
``Erik |
once upon a time, windows ran on non-x86
platforms... |
18:19.31 |
``Erik |
that's where the 'portable' comes
from |
18:19.42 |
``Erik |
a buddy of mine had NT on an alpha,
heh |
18:19.46 |
IriX64 |
``erik my mged is a PE file. |
18:20.12 |
``Erik |
as it should be... |
18:20.45 |
IriX64 |
and is :) |
18:21.28 |
IriX64 |
http://www.pastebin.ca/371853 |
18:21.53 |
IriX64 |
alphas were sometimes shipped with NT
``Erik. |
18:24.02 |
``Erik |
the interesting part was that they did dynamic
machine code translation, iirc |
18:24.23 |
``Erik |
so you could run an 'x86' program, and the
first time you did, it'd take a long time to start up, then would
run as native alpha machine code from then on |
18:24.26 |
IriX64 |
alpha/nt? |
18:24.41 |
``Erik |
yeah |
18:24.53 |
IriX64 |
didn't play with it overly long, just got the
serial ports running. |
18:25.33 |
IriX64 |
nice little box though :) |
18:36.53 |
IriX64 |
http://www.pastebin.ca/371861
<=== IriX64 laff's hysterically. |
18:37.42 |
IriX64 |
winix is born :P |
18:44.35 |
IriX64 |
http://www.pastebin.ca/371879
<=== last one, won't bug you with it again. |
18:54.38 |
IriX64 |
now who's going to translate all those auto*
tools to .bat land ;) |
19:00.32 |
brlcad |
i hear a volunteer! |
19:03.44 |
CIA-7 |
BRL-CAD: 03brlcad * 10brlcad/configure.ac:
autoconf was complaining about AM_C_PROTOTYPES being called after
AC_C_INLINE and AC_C_CONST, so move it up before them |
19:06.58 |
IriX64 |
tie him to the typer quick :) |
19:07.52 |
IriX64 |
errr you heard that silent plea for help and
you do this to me? :P |
19:10.30 |
Maloeran |
I need input on a theory. This new box doesn't
have a sound chip. If I were to forward a /dev/dsp socket to write
to the /dev/dsp of another box, I would effectively be redirecting
all audio output to the other machine's. Correct? |
19:12.46 |
IriX64 |
why not throw in a cheap sound card? |
19:13.04 |
IriX64 |
45$ canadian gets you a pci
soundblaster. |
19:30.59 |
*** join/#brlcad docelic
(n=docelic@212.91.114.66) |
19:42.50 |
IriX64 |
http://www.pastebin.ca/371946
<== Where I want to go with this thing, at the moment she just
"looks" pretty :) |
19:57.17 |
IriX64 |
http://www.pastebin.ca/371970
< now I return to ruminating just how big angels wings must be
:) |
19:57.56 |
IriX64 |
btw, it works in both environments, admittly I
rely on cygwin but still... |
19:58.22 |
``Erik |
'both' environment? O.o |
19:58.36 |
IriX64 |
:) |
20:03.35 |
IriX64 |
http://www.pastebin.ca/371979
<=== I'll shut up now, cheers. |
20:08.14 |
*** join/#brlcad cad18
(n=47c72e4f@bz.bzflag.bz) |
20:29.38 |
Maloeran |
Just for the record, redirecting /dev/dsp
worked nicely |
20:37.50 |
``Erik |
bah, fucking snow |
20:42.23 |
``Erik |
... |
20:42.44 |
IriX64 |
http://www.pastebin.ca/372055
<=== ``Erik, recognize this one? |
20:44.02 |
IriX64 |
hahahah this is an OS/2 32 bit
executable. |
20:49.03 |
IriX64 |
http://www.pastebin.ca/index.php |
21:01.42 |
IriX64 |
TurboC high on Cygwin hehe. |
21:13.35 |
IriX64 |
btw pretty.exe produced by that TC ran in the
cygwin environment stand alone. |
21:17.05 |
Maloeran |
Such a strange person |
22:00.45 |
*** join/#brlcad IriX64
(n=mario_du@bas3-sudbury98-1168050480.dsl.bell.ca) |
22:30.24 |
Maloeran |
I just noticed something. When the raytracing
demo renders 100fps at 800x600, only ~30% of the cpus time is spent
raytracing, the rest is "idle" ( according to 'top' ) |
22:31.16 |
Maloeran |
I'm guessing it's SDL's surface blitting being
a serious bottleneck |
22:32.40 |
Maloeran |
Nevermind, SDL is just a small part of it.
Blocking until the frame is complete instead of pushing and
buffering several frames at the same time is the problem |
22:40.06 |
IriX64 |
farkles, does BRL-CAD support building share
*and static libs on the same run? |
22:47.07 |
brlcad |
IriX64: of course it does, and is what it does
by default |
22:50.59 |
IriX64 |
ty lucy :) |
22:51.55 |
brlcad |
seriously, you know the issue with cygwin ..
if you're not going to investigate properly, why do you
ask? |
22:53.41 |
IriX64 |
seperate thread for everybody? ie main
launches them all and waits? |
22:54.37 |
Maloeran |
The proper is at the end of a frame, only a
handful of threads might still be working, the others are done and
there's no work left for this frame |
22:55.14 |
Maloeran |
In practice, I can't get my cpu use above 80%
with 8 cores |
22:55.23 |
brlcad |
you're just now running into this
problem? |
22:55.26 |
IriX64 |
have it wait on a universal done signale, ok
got yours now i reset the flag and wait again. |
22:55.48 |
``Erik |
he just now got enough hw to find the
problem... |
22:55.50 |
Maloeran |
Yes brlcad, it was not much of an issue with
just 2 cores |
22:56.10 |
Maloeran |
It's not a problem with the engine, just the
demos |
22:56.27 |
brlcad |
I know, i'm just a bit surprised |