|
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I posted an image in p.b.i titled "What causes this effect."
Can anyone explain the striated (?) effect on the inserts. I noticed that the
vertical cylinders in my Orrery (WIP) Larger insert looked like several
cylinders stacked one on top of another and to the side. So I increased the AA:
threshold 0, Rays 9x9, depth 5, and they were still there. Can't be that then.
The Main image has no AA but the camera angle is 9.8 whilst the inserts have an
angle of 39.6 (Moray settings.
As I'll want a largish field of view what causes it and is there is a work
around.
Regards
Stephen
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Fri, 04 Jul 2003 18:17:34 GMT, mca### [at] aolcom (S McAvoy) wrote:
> So I increased the AA:
> threshold 0, Rays 9x9, depth 5
Try off jitter.
Try larger size of image.
And ?
ABX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Fri, 04 Jul 2003 20:21:28 +0200, ABX <abx### [at] abxartpl> wrote:
>On Fri, 04 Jul 2003 18:17:34 GMT, mca### [at] aolcom (S McAvoy) wrote:
>> So I increased the AA:
>> threshold 0, Rays 9x9, depth 5
>
>Try off jitter.
Sorry I did not say, jitter = 0.0 I never use it, or so I thought. I copied the
settings below from the Moray generated ini file and I see that jitter is on. If
that is the problem Maybe Lutz will include a switch if I ask nicely
Width = 1024
Height = 768
Start_Column = 0.232
End_Row = 0.491
End_Column = 0.293
Start_Row = 0.417
Antialias = On
Antialias_Threshold = 0.0
Jitter=On
Jitter_Amount = 0.0
Antialias_Depth = 9
Sampling_Method = 2
>Try larger size of image.
Is 1024 x 768 too small
>And ?
What can you mean, Very nicely put. I will RTFM.
I hope :-}
Regards
Stephen
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi again!
I can't believe your options:
> Antialias = On
> Antialias_Threshold = 0.0
> Antialias_Depth = 9
> Sampling_Method = 2
The meaning of this is:
Do antialiasing (Antialias = On) on *EVERY* pixel (Antialias_Threshold = 0.0),
use the recursive subdivision method (Sampling_Method = 2) and make 9 sub-
divisions (ALWAYS because Antialias_Threshold = 0.0). This doesn't give 9x9
rays per pixel, but (2^9+1) x (2^9+1) = 263169 rays per pixel! (Some of them
are reused at the next pixel.)
Here are the (partial) statistics resulting from rendering a totally empty
scene at 4x3 with different AA settings. Especially look at Samples/pixel,
Peak memory used and PPS (pixel per second).
-1- Antialias_Threshold = 0.0, Antialias_Depth = 9
Rendering using command line '+A0.0 +AM2 +R9 -J0.0 -F +W4 +H3'.
Persistence of Vision(tm) Ray Tracer Version 3.5 win32 (.icl)
Image resolution 4 by 3 (rows 1 to 3, columns 1 to 4).
Antialiasing.........On (Method 2, Threshold 0.000, Depth 9, Jitter 0.00)
Scene contains 0 frame level objects; 0 infinite.
Pixels: 12 Samples: 3149313 Smpls/Pxl: 262442.75
Rays: 3149313 Saved: 0 Max Level: 1/5
Peak memory used: 6476331 bytes
CPU time used: kernel 0.01 seconds, user 3.05 seconds, total 3.06 seconds
Render averaged 3.92 PPS over 12 pixels
-2- Antialias_Threshold = 0.0001, Antialias_Depth = 9
Rendering using command line '+A0.001 +AM2 +R9 -J0.0 -F +W4 +H3'.
Persistence of Vision(tm) Ray Tracer Version 3.5 win32 (.icl)
Image resolution 4 by 3 (rows 1 to 3, columns 1 to 4).
Antialiasing.........On (Method 2, Threshold 0.000, Depth 9, Jitter 0.00)
Scene contains 0 frame level objects; 0 infinite.
Pixels: 12 Samples: 20 Smpls/Pxl: 1.67
Rays: 20 Saved: 0 Max Level: 1/5
Peak memory used: 6476331 bytes
CPU time used: kernel 0.00 seconds, user 0.36 seconds, total 0.36 seconds
Render averaged 33.29 PPS over 12 pixels
-3- Antialias_Threshold = 0.0001, Antialias_Depth = 3
Rendering using command line '+A0.0001 +AM2 +R3 -J0.0 -F +W4 +H3'.
Persistence of Vision(tm) Ray Tracer Version 3.5 win32 (.icl)
Image resolution 4 by 3 (rows 1 to 3, columns 1 to 4).
Antialiasing.........On (Method 2, Threshold 0.000, Depth 3, Jitter 0.00)
Scene contains 0 frame level objects; 0 infinite.
Pixels: 12 Samples: 20 Smpls/Pxl: 1.67
Rays: 20 Saved: 0 Max Level: 1/5
Peak memory used: 60821 bytes
CPU time used: kernel 0.00 seconds, user 0.01 seconds, total 0.01 seconds
Render averaged 1198.32 PPS over 12 pixels
So my suggestions are:
- NEVER use Antialias_Threshold = 0.0; 0.001 will do the same (at 8 bits
per color) and save VERY much time if the scene contains areas without
change in color.
- If Sampling_Method = 2, Antialias_Depth = 3 is almost always enough. I
can't think of any use of values >5.
I couldn't reproduce your problem, so what about posting a minimal scene
(1 yellow cylinder, light, camera), .ini file and toolbar command line
settings?
Sputnik
P.S. more asteroids: http://www.f-lohmueller.de/pov_anim/astr_ani.htm
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
<fr### [at] computermuseumfh-kielde> wrote:
>Hi again!
>
>I can't believe your options:
>
What can't you believe?
1. These are the options I used.
2. I could be so stupid.
3. All of the above.
Well it took over three and a half hrs to render the smaller insert. I just
picked the most extreme settings I could and left it rendering overnight.
Here is a minimal scene. I hope I don't get shot for posting it here.
camera { // Camera Camera01
location < -10.235, -53.417, 23.301>
sky < 0.00000, 0.00000, 1.00000> // Use right handed-system
up < 0.0, 0.0, 1.0> // Where Z is up
right < 1.35783, 0.0, 0.0> // Right Vector is adjusted
to compensate for spherical (Moray) vs. planar (POV-Ray) aspect ratio
angle 39.60000 // Vertical 29.700
look_at < -30.229, 1.367, -0.976>
}
light_source {
<0.0, 0.0, 0.0>
color rgb <1.000, 1.000, 1.000>*9.9990
translate <6.697869, -12.554867, 8.987353>
}
cylinder { // Cylndr068
<0,0,1>, <0,0,0>, 1
material {
texture
{
pigment
{
color rgb <1.0, 1.0, 1.0>
}
}
}
scale <0.012183, 0.012183, 4.429599>
rotate <0.0, 0.0, 0.0>
translate <-43.897332, 0.039729, -4.532148>
}
/*
Radiosity=Off
Width = 512
Height = 384
Antialias = On
Antialias_Threshold = 0.0
Jitter=On
Jitter_Amount = 0.1
Antialias_Depth = 3
Sampling_Method = 2
*/
Regards
Stephen
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Fri, 04 Jul 2003 20:21:28 +0200, ABX <abx### [at] abxartpl> wrote:
>On Fri, 04 Jul 2003 18:17:34 GMT, mca### [at] aolcom (S McAvoy) wrote:
>Try off jitter.
>Try larger size of image.
>And ?
Any hints here? I am struggling. I have found that the further you get
horizontally from the centre of the camera's view the more pronounced the
effect. It becomes noticeable as a displacement about 7 degrees although if the
less than 2 degrees.
Regards
Stephen
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Sun, 06 Jul 2003 22:02:54 GMT, mca### [at] aolcom (S McAvoy) wrote:
>Any hints here? I am struggling.
Funny how you can struggle looking for something, give up, ask for help, then
find the answer. And the answer is "orthographic". A problem using Moray.
Thanks for all the help anyway.
Regards
Stephen
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|