I tested simple  Povray and Yafray scene.

Antialias:
Povray settings : AA:0.01 , Yafray :AA:0.05
Result : Yafray is better than Povray:(

Light diffusion:
Povray settings : Brilliance : 4
Yafray setting : Default
Result : Yafray is better than Povray:(

But I like povray. I hope that Povray will develope.
is this possible?

Happy new year all and Pov team.

Hasan.

Post a reply to this message

Attachments:
Download 'povyafray.jpg' (91 KB)

Preview of image 'povyafray.jpg'

Hasan3 wrote:
> I tested simple  Povray and Yafray scene.
> 
> Antialias:
> Povray settings : AA:0.01 , Yafray :AA:0.05
> Result : Yafray is better than Povray:(
> 
> Light diffusion:
> Povray settings : Brilliance : 4
> Yafray setting : Default
> Result : Yafray is better than Povray:(
> 
> But I like povray. I hope that Povray will develope.
> is this possible?

1) I don't see any indication in those images explaining why you think 
one is 'better'.

2) The major differences between the images most likely result from 
different scenes and render settings and not from differences in the 
renderer abilities.

3) any such comparison that does not include render times does not make 
much sense.

4) You are comparing apples and peaches - Yafray can only render 
triangles while in POV-Ray you can create the cylinders as true 
cylinders (which apparently you have not).

To sum it up: It would be great to see a qualified comparison between 
Yafray and POV-Ray.  Gilles Tran some time ago did this for C4D and 
POV-Ray (using radiosity and antialiasing test scenes and comparing 
program features).  Your two images do not give much useful information 
on this aspect though.

Christoph

-- 
POV-Ray tutorials, include files, Landscape of the week:
http://www.imagico.de/ (Last updated 31 Oct. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/

Post a reply to this message

Christoph Hormann <chr### [at] gmxde> wrote:
> Hasan3 wrote:
> > I tested simple  Povray and Yafray scene.
> >
> > Antialias:
> > Povray settings : AA:0.01 , Yafray :AA:0.05
> > Result : Yafray is better than Povray:(
> >
> > Light diffusion:
> > Povray settings : Brilliance : 4
> > Yafray setting : Default
> > Result : Yafray is better than Povray:(
> >
> > But I like povray. I hope that Povray will develope.
> > is this possible?
>
> 1) I don't see any indication in those images explaining why you think
> one is 'better'.
>
> 2) The major differences between the images most likely result from
> different scenes and render settings and not from differences in the
> renderer abilities.
>
> 3) any such comparison that does not include render times does not make
> much sense.
>
> 4) You are comparing apples and peaches - Yafray can only render
> triangles while in POV-Ray you can create the cylinders as true
> cylinders (which apparently you have not).
>
> To sum it up: It would be great to see a qualified comparison between
> Yafray and POV-Ray.  Gilles Tran some time ago did this for C4D and
> POV-Ray (using radiosity and antialiasing test scenes and comparing
> program features).  Your two images do not give much useful information
> on this aspect though.
>
> Christoph
>
> --
> POV-Ray tutorials, include files, Landscape of the week:
> http://www.imagico.de/ (Last updated 31 Oct. 2005)
> MegaPOV with mechanics simulation: http://megapov.inetart.net/


Cristoph, is the following pictures same? (Antialise settings are 0.01)

I used cube with triangles... Sources of both :

//POVRAY SOURCE
global_settings {
  max_trace_level 5
  assumed_gamma 1
  ambient_light rgb< 0,0,0>


}

#include "colors.inc"
#include "textures.inc"
#include "shapes.inc"
  //*camera //--------------------------------------------------------------
    camera {
            perspective
            location  <-4.06028292537847,10.5992586823105,-26.8592291175913>
            look_at   <-1.74267196209061,10.5992586823105,-23.4665653286463>
    }




//*LIGHT //--------------------------------------------------------------
//------------------------------------------------------------------------
  light_source {<0.238015198202164,19.6930029933815,-20.2299585531137>
        color rgb <1,1,1>*2
 media_interaction on
    fade_distance 0
    fade_power 0
  }
 #macro mat_white1 (scale1, map_type1)
texture {
    pigment {  color rgbf <1,1,1,0>}

 finish{
            ambient 0.5
            diffuse 0.7
            brilliance 4
            reflection {
            }
    }

}
#end
#declare Obje2=
  mesh   {
  smooth_triangle {
<-8.52331524112877,0,-31.2845295567918>,
<0,1903.345,0>,
<26.0865135577754,0,23.7098157920835>,
<0,1903.345,0>,
<-8.52331524112877,0,23.7098157920835>,
<0,1903.345,0>
  }
  smooth_triangle {
<-8.52331524112877,0,-31.2845295567918>,
<0,1903.345,0>,
<26.0865135577754,0,23.7098157920835>,
<0,1903.345,0>,
<26.0865135577754,0,-31.2845295567918>,
<0,1903.345,0>
  }
  smooth_triangle {
<-8.52331524112877,26.070864714638,-31.2845295567918>,
<0,1903.345,0>,
<26.0865135577754,26.070864714638,23.7098157920835>,
<0,1903.345,0>,
<-8.52331524112877,26.070864714638,23.7098157920835>,
<0,1903.345,0>
  }
  smooth_triangle {
<-8.52331524112877,26.070864714638,-31.2845295567918>,
<0,1903.345,0>,
<26.0865135577754,26.070864714638,23.7098157920835>,
<0,1903.345,0>,
<26.0865135577754,26.070864714638,-31.2845295567918>,
<0,1903.345,0>
  }
  smooth_triangle {
<-8.52331524112877,26.070864714638,-31.2845295567918>,
<0,0,-902.3082>,
<26.0865135577754,0,-31.2845295567918>,
<0,0,-902.3082>,
<26.0865135577754,26.070864714638,-31.2845295567918>,
<0,0,-902.3082>
  }
  smooth_triangle {
<-8.52331524112877,26.070864714638,-31.2845295567918>,
<0,0,-902.3082>,
<26.0865135577754,0,-31.2845295567918>,
<0,0,-902.3082>,
<-8.52331524112877,0,-31.2845295567918>,
<0,0,-902.3082>
  }
  smooth_triangle {
<-8.52331524112877,26.070864714638,23.7098157920835>,
<0,0,-902.3082>,
<26.0865135577754,0,23.7098157920835>,
<0,0,-902.3082>,
<26.0865135577754,26.070864714638,23.7098157920835>,
<0,0,-902.3082>
  }
  smooth_triangle {
<-8.52331524112877,26.070864714638,23.7098157920835>,
<0,0,-902.3082>,
<26.0865135577754,0,23.7098157920835>,
<0,0,-902.3082>,
<-8.52331524112877,0,23.7098157920835>,
<0,0,-902.3082>
  }
  smooth_triangle {
<-8.52331524112877,26.070864714638,23.7098157920835>,
<-1433.75,0,0>,
<-8.52331524112877,0,-31.2845295567918>,
<-1433.75,0,0>,
<-8.52331524112877,26.070864714638,-31.2845295567918>,
<-1433.75,0,0>
  }
  smooth_triangle {
<-8.52331524112877,26.070864714638,23.7098157920835>,
<-1433.75,0,0>,
<-8.52331524112877,0,-31.2845295567918>,
<-1433.75,0,0>,
<-8.52331524112877,0,23.7098157920835>,
<-1433.75,0,0>
  }
  smooth_triangle {
<26.0865135577754,26.070864714638,-31.2845295567918>,
<1433.75,0,0>,
<26.0865135577754,0,23.7098157920835>,
<1433.75,0,0>,
<26.0865135577754,26.070864714638,23.7098157920835>,
<1433.75,0,0>
  }
  smooth_triangle {
<26.0865135577754,26.070864714638,-31.2845295567918>,
<1433.75,0,0>,
<26.0865135577754,0,23.7098157920835>,
<1433.75,0,0>,
<26.0865135577754,0,-31.2845295567918>,
<1433.75,0,0>
  }
  mat_white1(<-0.25,0.33,0.25 >,2)

  }

 object {Obje2}

//POVRAY SOURCE END------------------------------------






---------YAFRAY SOURCE----------------------


<scene>
<shader type = "generic" name = "mat_white">
          <attributes>
                  <color r="1" g= "1" b= "1" />
                  <specular r="0" g= "0" b= "0" />
                  <reflected r="0" g= "0" b= "0" />
                  <reflected2 r="0" g= "0" b= "0" />
                  <transmitted2 r="" g= "" b= "" />
                  <hard value = "25"/>
                  <IOR value = "1"/>
                  <min_refle value = "0"/>
                  <fast_fresnel value = "off"/>
          </attributes>
</shader>
<transform m00="1.000000" m01="0.000000" m02="0.000000" m03="0.000000"
           m10="0.000000" m11="1.000000" m12="0.000000" m13="0.000000"
           m20="0.000000" m21="0.000000" m22="1.000000" m23="0.000000"
           m30="0.000000" m31="0.000000" m32="0.000000" m33="1.000000">
<object name="Model2" shadow="on"  shader_name="mat_white" >
    <attributes>
    </attributes>
    <mesh autosmooth="0" has_orco="on" >
        <points>
            <p x="-0.8523" y="-3.1285" z="0" />
            <p x="-0.8523" y="2.371" z="0" />
            <p x="2.6087" y="2.371" z="0" />
            <p x="2.6087" y="-3.1285" z="0" />
            <p x="-0.8523" y="-3.1285" z="2.6071" />
            <p x="-0.8523" y="2.371" z="2.6071" />
            <p x="2.6087" y="2.371" z="2.6071" />
            <p x="2.6087" y="-3.1285" z="2.6071" />
            <p x="-0.8523" y="-3.1285" z="2.6071" />
            <p x="2.6087" y="-3.1285" z="2.6071" />
            <p x="2.6087" y="-3.1285" z="0" />
            <p x="-0.8523" y="-3.1285" z="0" />
            <p x="-0.8523" y="2.371" z="2.6071" />
            <p x="2.6087" y="2.371" z="2.6071" />
            <p x="2.6087" y="2.371" z="0" />
            <p x="-0.8523" y="2.371" z="0" />
            <p x="-0.8523" y="2.371" z="2.6071" />
            <p x="-0.8523" y="-3.1285" z="2.6071" />
            <p x="-0.8523" y="-3.1285" z="0" />
            <p x="-0.8523" y="2.371" z="0" />
            <p x="2.6087" y="-3.1285" z="2.6071" />
            <p x="2.6087" y="2.371" z="2.6071" />
            <p x="2.6087" y="2.371" z="0" />
            <p x="2.6087" y="-3.1285" z="0" />
         </points>
        <faces>
            <f a="3" b="0" c="1" shader_name="mat_white" />
            <f a="1" b="2" c="3" shader_name="mat_white" />
            <f a="7" b="4" c="5" shader_name="mat_white" />
            <f a="5" b="6" c="7" shader_name="mat_white" />
            <f a="0" b="4" c="7" shader_name="mat_white" />
            <f a="7" b="3" c="0" shader_name="mat_white" />
            <f a="1" b="5" c="6" shader_name="mat_white" />
            <f a="6" b="2" c="1" shader_name="mat_white" />
            <f a="1" b="5" c="4" shader_name="mat_white" />
            <f a="4" b="0" c="1" shader_name="mat_white" />
            <f a="3" b="7" c="6" shader_name="mat_white" />
            <f a="6" b="2" c="3" shader_name="mat_white" />
        </faces>
    </mesh>
</object>
</transform>
<light type="pointlight" glow_intensity="0" glow_offset="0" glow_type="0"
name="Light1" power="14" cast_shadows="on">
    <from x="0.0238" y="-2.023" z="1.9693" />
    <color r="1" g="1" b="1" />
</light>
<background type="" name="world_background">
 <color r="0,056563" g="0,220815" b="0,4" />
</background>
<camera name="MAINCAM" type="perspective" resx="800" resy="600"
    focal="1.09375" aspect_ratio="1"
    dof_distance="0" aperture="0" use_qmc="on"
    bokeh_type="disk1" bokeh_bias="uniform" bokeh_rotation="0" >
    <from x="-0.406" y="-2.6859" z="1.0599" />
    <to x="-0.1743" y="-2.3467" z="1.0599" />
    <up x="-0.406" y="-2.6859" z="2.0599" />
</camera>
<background type="constant" name="world_background" >
<color r="0.056563" g="0.220815" b="0.4"/>
</background>
<render camera_name="MAINCAM"
    raydepth="4" gamma="1" exposure="0"
    AA_passes="2" AA_minsamples="8"
    AA_pixelwidth="1.5" AA_threshold="0.01" bias="0.001" clamp_rgb="on"
    background_name ="world_background"
 >
    <outfile value="C:TEST.tga" />
</render>
</scene>

Post a reply to this message

Attachments:
Download 'povyafray1.jpg' (115 KB)

Preview of image 'povyafray1.jpg'

Hasan3 wrote:
> 
> Cristoph, is the following pictures same? (Antialise settings are 0.01)
> 

Well - as you can see they are not.  The difficult part of such a 
comparison is to find which settings for both renderers are most 
similar.  This requires knowledge of both of them.

Note using smooth triangles here is most inefficient - don't try any 
performance comparisons based on that.

Christoph

-- 
POV-Ray tutorials, include files, Landscape of the week:
http://www.imagico.de/ (Last updated 31 Oct. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/

Post a reply to this message

> Well - as you can see they are not.  The difficult part of such a
> comparison is to find which settings for both renderers are most
> similar.  This requires knowledge of both of them.

In this case for example, it would have been good to know that POV-Ray
doesn't clip the RGB-values before averaging them for antialiasing. It has
been a discussed annoyance to some, and AFAIK there is some work done in
that area for the next release (I think Warp hinted at it, but don't go bug
him because I said so!). I think 3.5 did it differently/like Yafray and
could be used for a comparison of antialiasing.

I personally haven't tried Yafray, but just looked at its capabilities some
time ago, and didn't think it would be better than POV-Ray. Tried CSG in
Yafray? Isosurfaces? What about media, is it possible in Yafray? ...

Regards,
Tim

-- 
aka "Tim Nikias v2.0"
Homepage: <http://www.nolights.de>

Post a reply to this message

Tim Nikias wrote:
>>Well - as you can see they are not.  The difficult part of such a
>>comparison is to find which settings for both renderers are most
>>similar.  This requires knowledge of both of them.
> 
> 
> In this case for example, it would have been good to know that POV-Ray
> doesn't clip the RGB-values before averaging them for antialiasing. It has
> been a discussed annoyance to some, and AFAIK there is some work done in
> that area for the next release (I think Warp hinted at it, but don't go bug
> him because I said so!). I think 3.5 did it differently/like Yafray and
> could be used for a comparison of antialiasing.

I don't think there is any significant difference in these test images 
due to antialiasing.  The main sources for the difference in results are 
probably different textures, different light intensities or different 
gamma correction (or all of them).

Christoph

-- 
POV-Ray tutorials, include files, Landscape of the week:
http://www.imagico.de/ (Last updated 31 Oct. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/

Post a reply to this message

The very least you could do if you wanted to test AA would be to put a pure
white next to a pure black and look again. The problem in those last two is
that the scenes themselves clearly look different, so comparing any one
aspect is silly.

-s
5TF!

Post a reply to this message

"stm31415" <sam### [at] cscom> wrote:
> The very least you could do if you wanted to test AA would be to put a pure
> white next to a pure black and look again. The problem in those last two is
> that the scenes themselves clearly look different, so comparing any one
> aspect is silly.
>
> -s
> 5TF!

I agree, if you compare the walls on the right hand side of each scene where
the colours are similar the AA difference is hardly noticable, you can't
compare a white wall with a light grey one and expect to get the same
results. I have been using POV for many years (although not with meshes or
triangles very much) and never noticed any problems with poor AA quality.

Sean

Post a reply to this message

>I don't think there is any significant difference in these test images
>due to antialiasing.  The main sources for the difference in results are
>probably different textures, different light intensities or different
>gamma correction (or all of them).

>Christoph

Cristof, you may be right. I try to use same textures and gamma.


>The very least you could do if you wanted to test AA would be to put a pure
>white next to a pure black and look again. The problem in those last two is
>that the scenes themselves clearly look different, so comparing any one
>aspect is silly.

>-s
>5TF!

5TF, light source's coordinats are same. But it was white in the Pov scene.
Probably this is the reason of AA variant's .




>I agree, if you compare the walls on the right hand side of each scene where
>the colours are similar the AA difference is hardly noticable, you can't
>compare a white wall with a light grey one and expect to get the same
>results. I have been using POV for many years (although not with meshes or
>triangles very much) and never noticed any problems with poor AA quality.

>Sean

Sean, you are right. I was'nt carefull that detail. Sory.


>I personally haven't tried Yafray, but just looked at its capabilities some
>time ago, and didn't think it would be better than POV-Ray. Tried CSG in
>Yafray? Isosurfaces? What about media, is it possible in Yafray? ...
>Regards,
>Tim

Tim, Yafray hav not CGS, isosurface, prism, sweep sphere etc. Only mesh
object. And very slowly.
Of course Povray is great. But I think that Pov team must examine Yafray
Light system.

Regards

Post a reply to this message

I am content to have a faster, more flexible, better documented (IMHO)
program. If there is an AA difference (which we have yet to establish), I
don't mind. I'm not going to tell the team to do *anything* - they do more
than I would on any given day ;)

BTW, 5TF! is a statement of well wishes, like 'Peace' or 'Good-bye.' I
generally go by Sam, -s, or stm31415. But whatever- I knew who you were
talking to ;)

-s
5TF!

Post a reply to this message