 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
20 pixels per minute with 142 lines remaining comes to almost
57 hours left...*sigh* Just because I turned on 2x antialiasing
and 6 4x4 area lights and my wax media...
--
Tim Cook
http://empyrean.scifi-fantasy.com
mirror: http://personal.lig.bellsouth.net/lig/z/9/z993126
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GFA dpu- s: a?-- C++(++++) U P? L E--- W++(+++)>$
N++ o? K- w(+) O? M-(--) V? PS+(+++) PE(--) Y(--)
PGP-(--) t* 5++>+++++ X+ R* tv+ b++(+++) DI
D++(---) G(++) e*>++ h+ !r--- !y--
------END GEEK CODE BLOCK------
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Timothy R. Cook" <tim### [at] scifi-fantasy com> wrote in message
news:3db60ae0@news.povray.org...
> 20 pixels per minute with 142 lines remaining comes to almost
> 57 hours left...*sigh* Just because I turned on 2x antialiasing
> and 6 4x4 area lights and my wax media...
Dang, yeah. 4x4 area lights aren't good enough is you have wide penumbral
shadows.
I'm really wanting to get out of the sub-GHz era and join the ranks of
others with the multi-GHz CPU's. Um, sometime next year.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
hughes, b. wrote:
> Dang, yeah. 4x4 area lights aren't good enough is you have wide penumbral
> shadows.
heh usually I just use 2x2 area lights...
> I'm really wanting to get out of the sub-GHz era and join the ranks of
> others with the multi-GHz CPU's. Um, sometime next year.
1.8 GHz here...actually what I really want is a Beowulf cluster...
need to do some figuring on how to go about constructing one, and
price it out...
--
Tim Cook
http://empyrean.scifi-fantasy.com
mirror: http://personal.lig.bellsouth.net/lig/z/9/z993126
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GFA dpu- s: a?-- C++(++++) U P? L E--- W++(+++)>$
N++ o? K- w(+) O? M-(--) V? PS+(+++) PE(--) Y(--)
PGP-(--) t* 5++>+++++ X+ R* tv+ b++(+++) DI
D++(---) G(++) e*>++ h+ !r--- !y--
------END GEEK CODE BLOCK------
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> 1.8 GHz here...actually what I really want is a Beowulf cluster...
> need to do some figuring on how to go about constructing one, and
> price it out...
If you're using a 1.8 ghz machine, and doing a render at less than 1pps,
then you're most likely leaving something unoptimized. I'm on a 700mhz
machine, and I've never had to deal with a render under 5pps, if I spent
some time optimizing it. "Real Water"
(http://www.slimeland.com/images/pictures/realwater.jpg) was about 5pps, and
that uses media photons in scattering media, sphere sweeps, and some
isosurfaces.
For instance, the area lights are really nice for shadows on regular
surfaces, but do you really need them for the wax media? Probably not.
Solution: put the wax in a light_group with point lights in the positions of
the area lights. Then turn media_interaction for the area lights off. There
you go - the media is lit by point lights, while the shadows from the candle
(if that's what you're making) are calculated with area lights.
While you're at it, you might consider making the candle have no_shadow, and
duplicating its container, giving it no_image no_reflection, so that shadows
rays testing the candle don't have to travel through the actual media; they
just test against the candle shape. The difference will likely be
unnoticeable.
See, try things like that. You can speed it up.
- Slime
[ http://www.slimeland.com/ ]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Slime wrote:
> Solution: put the wax in a light_group with point lights in the positions of
> the area lights. Then turn media_interaction for the area lights off. There
> you go - the media is lit by point lights, while the shadows from the candle
> (if that's what you're making) are calculated with area lights.
Moray doesn't have light_groups (yet, don't know if it will).
Though...I might be able to do some optimization there and have
point lights with media interaction/no shadow, and area lights
with no media interaction/shadow..
BUT I've already spent this much time on the render and I doubt
that light change would gain me that much...would it? If it
would, I might change...
--
Tim Cook
http://empyrean.scifi-fantasy.com
mirror: http://personal.lig.bellsouth.net/lig/z/9/z993126
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GFA dpu- s: a?-- C++(++++) U P? L E--- W++(+++)>$
N++ o? K- w(+) O? M-(--) V? PS+(+++) PE(--) Y(--)
PGP-(--) t* 5++>+++++ X+ R* tv+ b++(+++) DI
D++(---) G(++) e*>++ h+ !r--- !y--
------END GEEK CODE BLOCK------
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Timothy R. Cook wrote:
> BUT I've already spent this much time on the render and I doubt
> that light change would gain me that much...would it? If it
> would, I might change...
screw it, I don't feel like waiting 3 days for this thing
*stops render to see if can optimize some*
--
Tim Cook
http://empyrean.scifi-fantasy.com
mirror: http://personal.lig.bellsouth.net/lig/z/9/z993126
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GFA dpu- s: a?-- C++(++++) U P? L E--- W++(+++)>$
N++ o? K- w(+) O? M-(--) V? PS+(+++) PE(--) Y(--)
PGP-(--) t* 5++>+++++ X+ R* tv+ b++(+++) DI
D++(---) G(++) e*>++ h+ !r--- !y--
------END GEEK CODE BLOCK------
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> BUT I've already spent this much time on the render and I doubt
> that light change would gain me that much...would it? If it
> would, I might change...
Depends on how much time is being spent in the wax's scattering media.
Scattering media + area lights = slooooooooowwwwww.
If you want to try it, I suggest you merely open a new instance of povray,
make a duplicate copy of the file, set your big render on low priority, and
start experimenting. Don't do it with moray though; if you really want to
optimize, you're going to get best results by hand.
- Slime
[ http://www.slimeland.com/ ]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> *stops render to see if can optimize some*
Ack! Keep a copy of the rendered file so you can resume it with +C later if
you want!
- Slime
[ http://www.slimeland.com/ ]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Slime wrote:
> Ack! Keep a copy of the rendered file so you can resume it with +C later if
> you want!
nahhhhh...it's only been rendering 2 days, hardly worth the trouble...;)
--
Tim Cook
http://empyrean.scifi-fantasy.com
mirror: http://personal.lig.bellsouth.net/lig/z/9/z993126
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GFA dpu- s: a?-- C++(++++) U P? L E--- W++(+++)>$
N++ o? K- w(+) O? M-(--) V? PS+(+++) PE(--) Y(--)
PGP-(--) t* 5++>+++++ X+ R* tv+ b++(+++) DI
D++(---) G(++) e*>++ h+ !r--- !y--
------END GEEK CODE BLOCK------
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Timothy R. Cook" wrote:
>
> 20 pixels per minute with 142 lines remaining comes to almost
> 57 hours left...*sigh* Just because I turned on 2x antialiasing
> and 6 4x4 area lights and my wax media...
Bah, you should really learn patience... ;-)
My IRTC picture has meanwhile slowed down to more than half an hour per
line, i hope it gets somewhat faster after the next 70 lines, otherwise i
will have problems finishing it in time...
Christoph
--
POV-Ray tutorials, IsoWood include,
TransSkin and more: http://www.tu-bs.de/~y0013390/
Last updated 13 Aug. 2002 _____./\/^>_*_<^\/\.______
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Wed, 23 Oct 2002 10:20:20 +0200, Christoph Hormann <chr### [at] gmxde>
wrote:
> My IRTC picture has meanwhile slowed down to more than half an hour per
> line, i hope it gets somewhat faster after the next 70 lines, otherwise i
> will have problems finishing it in time...
If you wish I can probably finish it for you for the _next_ round with my
borland compile on my portable compiling machine: Pentium 75 MHz. ;-)
ABX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
ABX wrote:
>
> If you wish I can probably finish it for you for the _next_ round with my
> borland compile on my portable compiling machine: Pentium 75 MHz. ;-)
I have my doubts. The estimated remaining render time at the current
speed is 7.1 days. That on an 1 GHz Athlon with optimized Linux gcc
compile. I think this could well exceed 2 month on your configuration...
Christoph
--
POV-Ray tutorials, IsoWood include,
TransSkin and more: http://www.tu-bs.de/~y0013390/
Last updated 13 Aug. 2002 _____./\/^>_*_<^\/\.______
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Wed, 23 Oct 2002 10:57:01 +0200, Christoph Hormann <chr### [at] gmxde>
wrote:
> > If you wish I can probably finish it for you for the _next_ round with my
> > borland compile on my portable compiling machine: Pentium 75 MHz. ;-)
>
> I have my doubts. The estimated remaining render time at the current
> speed is 7.1 days. That on an 1 GHz Athlon with optimized Linux gcc
> compile. I think this could well exceed 2 month on your configuration...
What about making screenshots every day, and releasing it as 'speed?' in next
animation round ...
ABX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Timothy R. Cook <tim### [at] scifi-fantasycom> wrote:
> heh usually I just use 2x2 area lights...
That's like not having area lights at all.
I don't understand why people are so afraid of using really dense area
lights.
Granted, they slow down the rendering a bit, but not that much (as long
as you are using adaptive sampling).
The only place where denser area lights slow down the rendering are in
the penumbra (ie. the soft border of the shadows). In all other parts
(ie. fully illuminated or fully shadowed) it doesn't matter how dense the
light source is, as long as the adaptiveness is fixed (that is, a 4x4 area
light and a 100x100 area light will be equally fast).
In a regular scene there's probably about 1%, in the worst case perhaps
even 10% of penumbra in the image area. Big deal?
--
#macro M(A,N,D,L)plane{-z,-9pigment{mandel L*9translate N color_map{[0rgb x]
[1rgb 9]}scale<D,D*3D>*1e3}rotate y*A*8}#end M(-3<1.206434.28623>70,7)M(
-1<.7438.1795>1,20)M(1<.77595.13699>30,20)M(3<.75923.07145>80,99)// - Warp -
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
As a cautionary tale, try rendering the incdemos on an Amiga 1200 at
20MHz... Not fun. (CPU is a Motorola 68020 at 20MHz, FPU is a Motorola 68882
at 50MHz, 32MB fastRAM, POV-Ray 3.1... Takes about 8 mins for the main
textures incdemo - *without* antialiasing.)
The Amiga is a lovely thing, but raytracing really requires more speed than
that... And anyway, POV-Ray 3.5 isn't availible for it. :-(
Andrew.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> That's like not having area lights at all.
>
> I don't understand why people are so afraid of using really dense area
> lights.
The wise Warp speaks...
(Not that I've ever tried to use area lights to this day...)
Andrew.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Andrew Coppin wrote:
>
> As a cautionary tale, try rendering the incdemos on an Amiga 1200 at
> 20MHz... Not fun. (CPU is a Motorola 68020 at 20MHz, FPU is a Motorola
> 68882 at 50MHz, 32MB fastRAM, POV-Ray 3.1... Takes about 8 mins for
> the main textures incdemo - *without* antialiasing.)
Not that comparing a nine-year-old machine to what we have today is
a fair comparison...
Regards,
John
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Wed, 23 Oct 2002 04:57:01 -0400, Christoph Hormann wrote:
> ABX wrote:
> I have my doubts. The estimated remaining render time at the current
> speed is 7.1 days. That on an 1 GHz Athlon with optimized Linux gcc
> compile. I think this could well exceed 2 month on your
> configuration...
An optimized POV for Linux? Where did you get that?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Tailkinker wrote:
>
> An optimized POV for Linux? Where did you get that?
Made myself, but the current official POV-Ray (3.50c) is fairly well
optimized too.
Christoph
--
POV-Ray tutorials, IsoWood include,
TransSkin and more: http://www.tu-bs.de/~y0013390/
Last updated 13 Aug. 2002 _____./\/^>_*_<^\/\.______
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Tailkinker wrote:
>
> An optimized POV for Linux? Where did you get that?
If you don't want to compile it yourself you can get optimized binaries
from http://www.povworld.org/povray/binaries.html
--
objects.povworld.org - The POV-Ray Objects Collection
book.povworld.org - The POV-Ray Book Project
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
