 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
A while back, (July actually), I had mentioned that I was rendering a scene on a
big multi-processor server. The scene was created by generating an image map
for the terrain, and modeling it as an isosurface treating the image as a
heightfield. I then added in a sea, a raft, and some crates floating on the
water.
The rendering is 1920x1080, very high anti-aliasing settings, and very detailed
radiosity settings. It finished 5 days ago while I was on vacation. Here is
the summary output:
----------------------------------------------------------------------------
Render Statistics
Image Resolution 1920 x 1080
----------------------------------------------------------------------------
Pixels: 2626329 Samples: 39348223 Smpls/Pxl: 14.98
Rays: 217018915 Saved: 14080837 Max Level: 10/10
----------------------------------------------------------------------------
Ray->Shape Intersection Tests Succeeded Percentage
----------------------------------------------------------------------------
CSG Intersection 15425335120 3141422415 20.37
CSG Union 15425335120 3230265453 20.94
Isosurface 58726964545 20544073426 34.98
Isosurface Container 64835402022 58872333276 90.80
Isosurface Cache 197763 142 0.07
Mesh 202928875 202928758 100.00
Sphere 189405535387 81155901845 42.85
Bounding Box 1532537787747 754640197288 49.24
----------------------------------------------------------------------------
Isosurface roots: 58659402950
Function VM calls: 19127723381339
----------------------------------------------------------------------------
Media Intervals: 39020599651 Media Samples: 375672022941 (9.63)
Shadow Ray Tests: 61251494093 Succeeded: 45924870608
Shadow Cache Hits: 21833521
Reflected Rays: 59614334 Total Internal: 254037
Refracted Rays: 28747738
Transmitted Rays: 30300931
----------------------------------------------------------------------------
Radiosity samples calculated: 704767 (0.63 %)
discarded due to low quality: 16
retained for re-use: 704751
Radiosity samples reused: 111961780
Radiosity sample rays shot: 56381360
Radiosity octree nodes: 66553
Radiosity octree samples/node: 10.59
Radiosity blocks examined: 100696927850
Radiosity blocks passed test 0: 100696927850 (100.00 %)
Radiosity blocks passed test 1: 24302952231 (24.13 %)
Radiosity blocks passed test 2: 15586132111 (15.48 %)
Radiosity blocks passed test 3: 1166923119 (1.16 %)
Radiosity blocks passed test 4: 687023123 (0.68 %)
Radiosity blocks passed test 5: 647393076 (0.64 %)
Radiosity blocks rejected: 100049534774 (99.36 %)
----------------------------------------------------------------------------
Radiosity Depth 0 calculated: 704767 (0.63 %)
Radiosity Depth 0 reused: 111961780
Radiosity Depth 0 rays shot: 56381360
----------------------------------------------------------------------------
Radiosity (final) calculated: 702430 (0.62 %)
Radiosity (final) reused: 111960854
Radiosity (final) rays shot: 56194400
----------------------------------------------------------------------------
Pass Depth 0 Total
----------------------------------------------------------------------------
1 151 151
2 519 519
3 1667 1667
Final 702430 702430
----------------------------------------------------------------------------
Total 704767 704767
Weight 0.142
----------------------------------------------------------------------------
----------------------------------------------------------------------------
Render Time:
Photon Time: No photons
Radiosity Time: 0 hours 5 minutes 18 seconds (318.983 seconds)
using 64 thread(s) with 19470.589 CPU-seconds total
Trace Time: 1725 hours 48 minutes 48 seconds (6212928.610 seconds)
using 64 thread(s) with 387130191.433 CPU-seconds total
POV-Ray finished
The terrain is interesting, but I'm not real happy with it. The water came out
fantastic, (IMHO). I'm not sure I'll do anything like this again, as I'm sure I
nearly melted the server for 2+1/2 months to finish...
-- Chris R.
Post a reply to this message
Attachments:
Download 'scene-hd-hq-rad13-20230925.png' (3385 KB)
Preview of image 'scene-hd-hq-rad13-20230925.png'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Chris R" <car### [at] comcast net> wrote:
> A while back, (July actually), I had mentioned that I was rendering a scene on a
> big multi-processor server. The scene was created by generating an image map
> for the terrain, and modeling it as an isosurface treating the image as a
> heightfield. I then added in a sea, a raft, and some crates floating on the
> water.
>
> The rendering is 1920x1080, very high anti-aliasing settings, and very detailed
> radiosity settings. It finished 5 days ago while I was on vacation.
> ...
> The terrain is interesting, but I'm not real happy with it. The water came out
> fantastic, (IMHO). I'm not sure I'll do anything like this again, as I'm sure I
> nearly melted the server for 2+1/2 months to finish...
>
>
> -- Chris R.
Wow that's really cool! It kind of reminds me of the Myst series somehow. The
water is incredible and the foreground elements are nicely detailed! Your poor
server probably needs a vacation, but I wonder if it would look even better
scaled down from a higher resolution to clear up the noise?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Op 25-9-2023 om 16:42 schreef Chris R:
> A while back, (July actually), I had mentioned that I was rendering a scene on a
> big multi-processor server. The scene was created by generating an image map
> for the terrain, and modeling it as an isosurface treating the image as a
> heightfield. I then added in a sea, a raft, and some crates floating on the
> water.
>
> The rendering is 1920x1080, very high anti-aliasing settings, and very detailed
> radiosity settings. It finished 5 days ago while I was on vacation. Here is
> the summary output:
>
> ----------------------------------------------------------------------------
> Render Statistics
> Image Resolution 1920 x 1080
> ----------------------------------------------------------------------------
> Pixels: 2626329 Samples: 39348223 Smpls/Pxl: 14.98
> Rays: 217018915 Saved: 14080837 Max Level: 10/10
> ----------------------------------------------------------------------------
> Ray->Shape Intersection Tests Succeeded Percentage
> ----------------------------------------------------------------------------
> CSG Intersection 15425335120 3141422415 20.37
> CSG Union 15425335120 3230265453 20.94
> Isosurface 58726964545 20544073426 34.98
> Isosurface Container 64835402022 58872333276 90.80
> Isosurface Cache 197763 142 0.07
> Mesh 202928875 202928758 100.00
> Sphere 189405535387 81155901845 42.85
> Bounding Box 1532537787747 754640197288 49.24
> ----------------------------------------------------------------------------
> Isosurface roots: 58659402950
> Function VM calls: 19127723381339
> ----------------------------------------------------------------------------
> Media Intervals: 39020599651 Media Samples: 375672022941 (9.63)
> Shadow Ray Tests: 61251494093 Succeeded: 45924870608
> Shadow Cache Hits: 21833521
> Reflected Rays: 59614334 Total Internal: 254037
> Refracted Rays: 28747738
> Transmitted Rays: 30300931
> ----------------------------------------------------------------------------
> Radiosity samples calculated: 704767 (0.63 %)
> discarded due to low quality: 16
> retained for re-use: 704751
> Radiosity samples reused: 111961780
> Radiosity sample rays shot: 56381360
> Radiosity octree nodes: 66553
> Radiosity octree samples/node: 10.59
> Radiosity blocks examined: 100696927850
> Radiosity blocks passed test 0: 100696927850 (100.00 %)
> Radiosity blocks passed test 1: 24302952231 (24.13 %)
> Radiosity blocks passed test 2: 15586132111 (15.48 %)
> Radiosity blocks passed test 3: 1166923119 (1.16 %)
> Radiosity blocks passed test 4: 687023123 (0.68 %)
> Radiosity blocks passed test 5: 647393076 (0.64 %)
> Radiosity blocks rejected: 100049534774 (99.36 %)
> ----------------------------------------------------------------------------
> Radiosity Depth 0 calculated: 704767 (0.63 %)
> Radiosity Depth 0 reused: 111961780
> Radiosity Depth 0 rays shot: 56381360
> ----------------------------------------------------------------------------
> Radiosity (final) calculated: 702430 (0.62 %)
> Radiosity (final) reused: 111960854
> Radiosity (final) rays shot: 56194400
> ----------------------------------------------------------------------------
> Pass Depth 0 Total
> ----------------------------------------------------------------------------
> 1 151 151
> 2 519 519
> 3 1667 1667
> Final 702430 702430
> ----------------------------------------------------------------------------
> Total 704767 704767
> Weight 0.142
> ----------------------------------------------------------------------------
> ----------------------------------------------------------------------------
> Render Time:
> Photon Time: No photons
> Radiosity Time: 0 hours 5 minutes 18 seconds (318.983 seconds)
> using 64 thread(s) with 19470.589 CPU-seconds total
> Trace Time: 1725 hours 48 minutes 48 seconds (6212928.610 seconds)
> using 64 thread(s) with 387130191.433 CPU-seconds total
> POV-Ray finished
>
> The terrain is interesting, but I'm not real happy with it. The water came out
> fantastic, (IMHO). I'm not sure I'll do anything like this again, as I'm sure I
> nearly melted the server for 2+1/2 months to finish...
>
>
> -- Chris R.
Wow! I remember your mention of a looong render ;-) Very extreme indeed;
I don't think I shall ever do something like that. The result however,
is worthwhile. The scene is suitably alien I should think: the water
seems to be something a bit different from terrestrial seawater
(intentional?) and I like the landscape as it is. Something you might
want to explore if you want to reduce the needle sharp landscape erosion
which, I guess, originates with the image_map you generated. Drastically
smoothing that image might do the trick I believe, even if done locally
and/or to different degrees,
in order to keep some interesting features.
--
Thomas
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"19100" <nomail@nomail> wrote:
> "Chris R" <car### [at] comcastnet> wrote:
> > A while back, (July actually), I had mentioned that I was rendering a scene on a
> > big multi-processor server. The scene was created by generating an image map
> > for the terrain, and modeling it as an isosurface treating the image as a
> > heightfield. I then added in a sea, a raft, and some crates floating on the
> > water.
> >
> > The rendering is 1920x1080, very high anti-aliasing settings, and very detailed
> > radiosity settings. It finished 5 days ago while I was on vacation.
> > ...
> > The terrain is interesting, but I'm not real happy with it. The water came out
> > fantastic, (IMHO). I'm not sure I'll do anything like this again, as I'm sure I
> > nearly melted the server for 2+1/2 months to finish...
> >
> >
> > -- Chris R.
>
> Wow that's really cool! It kind of reminds me of the Myst series somehow. The
> water is incredible and the foreground elements are nicely detailed! Your poor
> server probably needs a vacation, but I wonder if it would look even better
> scaled down from a higher resolution to clear up the noise?
Myst was what inspired me to start playing with POV-Ray long ago.
I had done some lower resolution renders along the way, but seem to have lost
them. It turns out that just rendering the water, raft, and crates doesn't take
too long, but once you add in the jagged terrain reflecting in all of that bumpy
water, things get out of hand pretty quickly. The noise in the "cave" does get
considerably worse at higher resolutions.
-- Chris R.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thomas de Groot <tho### [at] degrootorg> wrote:
> Wow! I remember your mention of a looong render ;-) Very extreme indeed;
> I don't think I shall ever do something like that. The result however,
> is worthwhile. The scene is suitably alien I should think: the water
> seems to be something a bit different from terrestrial seawater
> (intentional?) and I like the landscape as it is. Something you might
> want to explore if you want to reduce the needle sharp landscape erosion
> which, I guess, originates with the image_map you generated. Drastically
> smoothing that image might do the trick I believe, even if done locally
> and/or to different degrees,
> in order to keep some interesting features.
> --
> Thomas
I can't even remember how I came up with the colors used in the sea water, but
it comes from a scattering media rather than from a pigment. I did some
research on wave patterns, but I'm not sure how well I modeled what I found, but
I liked it so I didn't play with it too much beyond that. The fun part was
getting the crates and the raft to "float" on the waves in a somewhat realistic
manner.
-- Chris R.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
