 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Just an FYI to the folks who are interested: Christoph's radiosity fixes
have been integrated and see to work very well. A hearty 'thanks' to him
for this! I intend to do a release in the next day or two that incorporates
the new code.
On my system (8 cores), cornell.pov at 1280x1024 AA 0.3 renders in 20
seconds with a speedup ratio of about 7x. I haven't looked into what might
be chewing up the rest of the CPU time, other than to note that the cores
aren't quite 100% pegged in the first five or so seconds of the render. (I
am guessing it may be related to the early sampling and octree code but
either way I'm happy enough with it just as it is:)
I haven't tried any animations yet or anything.
-- Chris
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Cason <del### [at] deletethistoo povrayorg> wrote:
> Just an FYI to the folks who are interested: Christoph's radiosity fixes
> have been integrated and see to work very well. A hearty 'thanks' to him
> for this! I intend to do a release in the next day or two that incorporates
> the new code.
>
> On my system (8 cores), cornell.pov at 1280x1024 AA 0.3 renders in 20
> seconds with a speedup ratio of about 7x.
Whoa! clipka is really Santa under disguise! :D
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"nemesis" <nam### [at] gmailcom> wrote in message
news:web.4965f14cbea0b42de44542980@news.povray.org...
> Whoa! clipka is really Santa under disguise! :D
Hey .... and thanks Chris as well for getting to this quickly!
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Cason <del### [at] deletethistoopovrayorg> wrote:
> Just an FYI to the folks who are interested: Christoph's radiosity fixes
> have been integrated and see to work very well. A hearty 'thanks' to him
> for this! I intend to do a release in the next day or two that incorporates
> the new code.
You guys are magicians! I stand in awe of what both of you have accomplished.
Ken W.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Cason <del### [at] deletethistoopovrayorg> wrote:
> Just an FYI to the folks who are interested: Christoph's radiosity fixes
> have been integrated and see to work very well. A hearty 'thanks' to him
> for this! I intend to do a release in the next day or two that incorporates
> the new code.
You're welcome! It's all just pure selfishness anyway :)
> On my system (8 cores), cornell.pov at 1280x1024 AA 0.3 renders in 20
> seconds with a speedup ratio of about 7x. I haven't looked into what might
> be chewing up the rest of the CPU time, other than to note that the cores
> aren't quite 100% pegged in the first five or so seconds of the render. (I
> am guessing it may be related to the early sampling and octree code but
Some things that may come into play here:
- The pre-render is done in multiple passes again, and after each pass the
system waits for all threads to finish (this will get some importance when
heading for that 100% reproducability thing) before starting the next pass;
this is *one* reason why the CPU utilization doesn't reach 100% during the
first seconds. This effect is strongest of course in images that render very
quickly, but will probably not be much of an issue for renders that take long
anyway. (A potential idea for optimization here would be to measure the (CPU)
time spent on a block during each pass, and do the most "expensive" ones first
during the next pass.)
- The first seconds of the pre-render sees a very high ratio of new samples
being taken per ray shot, with each new sample requiring the radiosity cache to
be locked for a brief moment, making it more likely for a thread having to wait
for another one to release its lock. In addition, and probably more
importantly, much of the cache's octree structure used to organize the samples
still has to be "grown" in this phase, again requiring a lock each time
additional nodes are added; these locks need to be maintained for a much longer
duration than for the mere adding of samples. These are additional reasons for
the incomplete utilization.
- The splotchy artifacts seen in 3.6 radiosity were due to a flawed design that
"cut short" some sample rays; with the flaws fixed, these rays are traced
further than before, increasing rendering time accordingly. (This shouldn't
play a role on cornell.pov though.) In addition, the fixes required passing
additional data through the render engine, thereby increasing general rendering
overhead a bit.
- With increasing radiosity recursion depth, POV-Ray reduces some quality
parameters. I modified the corresponding rules a bit in favor of quality, which
of course comes at a cost of time.
> either way I'm happy enough with it just as it is:)
Glad to hear that. I hope the beta testers will be as well. Me, however, I'm not
yet ;) I'm still trying to get some speedup for high-quality shots, using an
adaptive pretrace approach. Still encountering some obstacles though.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"nemesis" <nam### [at] gmailcom> wrote:
> Whoa! clipka is really Santa under disguise! :D
You better stop this, or I might actually start believing it =B)
(So if you see me dressed all in red-and-white, you'll know what happened... ;))
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Cason wrote:
> Just an FYI to the folks who are interested: Christoph's radiosity fixes
> have been integrated and see to work very well. A hearty 'thanks' to him
> for this! I intend to do a release in the next day or two that incorporates
> the new code.
This is really good news. POV's radiosity is already the fastest I have
encountered, and I get really good results as well. I have tested
Kerkythea, YafRay and LightWave. POV-Ray's radiosity beats them all, but
by what margin I cannot say.
Thank you to Chris Cason, clipka and the rest of the POV team!
Sam
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Oh you Great Heroes of Our (Render) Time!!
Chris, Clipka, without you the world would be a different place :-)
Thanks!!
Thomas
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
stbenge escribió:
> This is really good news. POV's radiosity is already the fastest I have
> encountered, and I get really good results as well. I have tested
> Kerkythea, YafRay and LightWave. POV-Ray's radiosity beats them all, but
> by what margin I cannot say.
I find myself thinking the same when I try every other
raytracer/renderer... (we are talking about the free ones, of course).
> Thank you to Chris Cason, clipka and the rest of the POV team!
Same here!
--
jaime
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I posted a test render in p.b-t.b. Did not experience the same speedup with
all scenes. The new beta is slower once I started adjusting global radiosity
settings.
Mike
"Chris Cason" <del### [at] deletethistoopovrayorg> wrote in
message news:4965cd6e$1@news.povray.org...
> Just an FYI to the folks who are interested: Christoph's radiosity fixes
> have been integrated and see to work very well. A hearty 'thanks' to him
> for this! I intend to do a release in the next day or two that
> incorporates
> the new code.
>
> On my system (8 cores), cornell.pov at 1280x1024 AA 0.3 renders in 20
> seconds with a speedup ratio of about 7x. I haven't looked into what might
> be chewing up the rest of the CPU time, other than to note that the cores
> aren't quite 100% pegged in the first five or so seconds of the render. (I
> am guessing it may be related to the early sampling and octree code but
> either way I'm happy enough with it just as it is:)
>
> I haven't tried any animations yet or anything.
>
> -- Chris
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Mike Hough <nos### [at] nospamcom> wrote:
> I posted a test render in p.b-t.b. Did not experience the same speedup with
> all scenes. The new beta is slower once I started adjusting global radiosity
> settings.
But the new beta (with the radiosity changes) is not published yet. How
is this possible?
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Apparently I misunderstood the situation. Will do the test again when it is
published. I thought Chris was saying that it was included in beta 30.
Interesting nonetheless.
"Warp" <war### [at] tagpovrayorg> wrote in message
news:496766da@news.povray.org...
> Mike Hough <nos### [at] nospamcom> wrote:
>> I posted a test render in p.b-t.b. Did not experience the same speedup
>> with
>> all scenes. The new beta is slower once I started adjusting global
>> radiosity
>> settings.
>
> But the new beta (with the radiosity changes) is not published yet. How
> is this possible?
>
> --
> - Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"clipka" <nomail@nomail> wrote:
> Chris Cason <del### [at] deletethistoopovrayorg> wrote:
> > Just an FYI to the folks who are interested: Christoph's radiosity fixes
> > have been integrated and see to work very well. A hearty 'thanks' to him
> > for this! I intend to do a release in the next day or two that incorporates
> > the new code.
>
> You're welcome! It's all just pure selfishness anyway :)
>
Well, be sure to thank your selfishness for me! :-P
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I've got some 32-bit EXE's of the radiosity changes up at:
http://www.povray.org/temp/pvengine-beta30-rad1.zip
No installer, sorry, there's a story behind that, will tell you later.
-- Chris
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> I've got some 32-bit EXE's of the radiosity changes up at:
Has anyone used these EXE's ?
-- Chris
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Cason <del### [at] deletethistoopovrayorg> wrote:
> > I've got some 32-bit EXE's of the radiosity changes up at:
>
> Has anyone used these EXE's ?
>
> -- Chris
After a long time many greetings to all of you!
Special Thanks to Chris and clipka for the effort providing "new" radiosity!
The shading looks much more smooth, so better compared to Beta29/30.
Also indirect lighting in dimmed regions shows better contrast. Very good!
But rendertime seems to be many times over (~10x ) with Beta30rad1 (considering
clipka's setting hint about nearest_count and low_error_factor)...
.....at least for my actual rendering:
W4500 H3000
QuadCore @2.66GHz
Vista64(!),
but also for
Core2Duo @2.4GHz
on XP32
When running rendering is done I'll post some comparing details in p.b.i...
In gratitude
TomAust
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Cason <del### [at] deletethistoopovrayorg> wrote:
> > I've got some 32-bit EXE's of the radiosity changes up at:
>
> Has anyone used these EXE's ?
>
> -- Chris
Yes, still rendering ;-)
34h and counting, I may have used exaggerated settings.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I have posted some test results to p.b-t.b. The cornell.pov scene renders
about 3 times faster with the beta than with 3.6 on my system but I can't
seem to get a good quality render using a more complex scene without
increasing the render time several fold.
Mike
"Chris Cason" <del### [at] deletethistoopovrayorg> wrote in
message news:496ffa24$1@news.povray.org...
>> I've got some 32-bit EXE's of the radiosity changes up at:
>
> Has anyone used these EXE's ?
>
> -- Chris
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I suggest a test for those with a very fast processor (pure radiosity scene):
Img: http://www.irtc.org/ftp/pub/stills/2002-08-31/golonls2.jpg
Source: http://www.irtc.org/ftp/pub/stills/2002-08-31/golonls2.zip
- radiosity abstract:
radiosity {
pretrace_start 0.08
pretrace_end 0.01
count 500
nearest_count 10 -----> change to 1?
error_bound 0.02
recursion_limit 1 ----> good
low_error_factor 0.2
gray_threshold 0.0
minimum_reuse 0.015
brightness 1
adc_bailout 0.01/2
}
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"StephenS" <nomail@nomail> wrote:
> Yes, still rendering ;-)
> 34h and counting, I may have used exaggerated settings.
The radiosity settings where some what random in nature, something to give a
long render.
global_settings {
adc_bailout 0.003
ambient_light rgb <1.000,1.000,1.000>
assumed_gamma 2.200
irid_wavelength rgb <0.250,0.180,0.140>
max_trace_level 15
number_of_waves 10
noise_generator 3
charset ascii
radiosity {
count 950
recursion_limit 4
nearest_count 20
adc_bailout 0.010
brightness 1.000
error_bound 1.080
gray_threshold 0.000
low_error_factor 0.125
minimum_reuse 0.000
always_sample 1
pretrace_end 0.001
pretrace_start 0.080
normal 1
media 0
}
}
A partial message pane output:
Parser Statistics
Finite Objects: 8
Infinite Objects: 0
Light Sources: 0
Total: 8
Parser Time
Parse Time: 0 hours 0 minutes 0 seconds (0.000 seconds)
using 1 thread(s) with 0.000 CPU-seconds total
Bounding Time: 0 hours 0 minutes 0 seconds (0.000 seconds)
using 1 thread(s) with 0.000 CPU-seconds total
Warning: Scene contained deprecated 'assumed_gamma'. Output gamma has been
adjusted to match legacy 'assumed_gamma' due to language version being set to
3.61 or lower. However, future versions of POV-Ray will not support backward
compatibility with 'assumed_gamma'. It is strongly recommended you update your
scene file by removing 'assumed_gamma' and add suitable 'Display_Gamma' and
'File_Gamma' options to your INI file or command-line instead. See the
documentation for details.
Render Options
Quality: 9
Bounding boxes.......On Bounding threshold: 3
Antialiasing.........On (Method 1, Threshold 0.300, Depth 3, Jitter 1.00)
Rendering stopped by GUI extension
Render Statistics
Image Resolution 640 x 480
Pixels: 487991 Samples: 16650 Smpls/Pxl: 0.03
Rays: 255902538 Saved: 0 Max Level: 640/480
Ray->Shape Intersection Tests Succeeded Percentage
Box 200894103 163730488 81.50
Cone/Cylinder 89127307 12357493 13.86
Bounding Box 2808677478 1104323087 39.32
Reflected Rays: 21455955
Radiosity samples calculated: 845891 (0.52 %)
Radiosity samples reused: 161480379
Radiosity (final) calculated: 609 (0.08 %)
Radiosity (final) reused: 776949
Pass Depth 0 Depth 1 Depth 2 Depth 3 Total
1 140 3294 4375 434 8243
2 539 4253 7819 273 12884
3 2107 7599 32260 730 42696
4 5638 11817 80982 691 99128
5+ 22502 36400 591111 32318 682331
Final 609 - - - 609
Total 31535 63363 716547 34446 845891
Render Time:
Photon Time: No photons
Radiosity Time: 35 hours 45 minutes 35 seconds (128735.938 seconds)
using 12 thread(s) with 356422.745 CPU-seconds total
Trace Time: 0 hours 29 minutes 7 seconds (1747.218 seconds)
using 2 thread(s) with 3488.937 CPU-seconds total
POV-Ray finished
-
CPU time used: kernel 11.13 seconds, user 359913.86 seconds, total 359924.98
seconds.
Elapsed time 185766.48 seconds, CPU vs elapsed time ratio 1.94.
----------------------------------------------------------------------------
Bottom right lists:
99 ppm
2d 03h 36m 06s
A difference between 3.61 and 3.7b30rad1 will be posted in a binary group.
3.7b30rad1 was slightly brighter.
Complete files available.
Stephen S
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
With all that CSG this scene is really slow without radiosity but it gave me
an idea. A few years ago I created a patch that allows the definition of
arbitrary points in an area light which could be used to create things like
light domes. I completely abadoned it but I am currently rendering this
scene with it to see how it might compare in terms of rendering speed and
quality for a scene with a recursion limit of only 1. I will post the result
in p.b.i when done since it is getting off-topic for this group.
"Carlo C." <nomail@nomail> wrote in message
news:web.4971a0b6bea0b42d4c2eec1b0@news.povray.org...
>
> I suggest a test for those with a very fast processor (pure radiosity
> scene):
>
> Img: http://www.irtc.org/ftp/pub/stills/2002-08-31/golonls2.jpg
> Source: http://www.irtc.org/ftp/pub/stills/2002-08-31/golonls2.zip
>
> - radiosity abstract:
>
> radiosity {
> pretrace_start 0.08
> pretrace_end 0.01
> count 500
> nearest_count 10 -----> change to 1?
> error_bound 0.02
> recursion_limit 1 ----> good
> low_error_factor 0.2
> gray_threshold 0.0
> minimum_reuse 0.015
> brightness 1
> adc_bailout 0.01/2
> }
>
>
>
>
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I've also done some tests. Results are in p.b-t.binaries.
-Ive
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
