On 08/08/2018 22:02, Mike Horvath wrote:
> On 8/8/2018 1:05 PM, Stephen wrote:
>> On 08/08/2018 15:08, Mike Horvath wrote:
>>> Okay, I uploaded my scene to here:
>>>
>>>
http://lib.povray.org/searchcollection/index2.php?objectName=Ringworld&version=2.0&contributorTag=SharkD

>>>
>>>
>>> I also attached my latest render to this post. The shadow squares' 
>>> widths now match what was apparently stated in the books. The 
>>> shadowed regions on the ring look much better now.
>>>
>>> However, I had to double the height of the shadow squares in order to 
>>> block out the sun. See the parameter `RWorld_Shadow_Square_Width`. 
>>> I'm not sure why this was necessary, as the original height was equal 
>>> to the sun's diameter, and should have fully blocked its light.
>>>
>>
>> Do the two area_light vectors <Axis_1> and <Axis_2> extend beyond the 
>> star's volume?
>>
> 
> 1. the axis lengths are equal to the sun's diameter
> 2. the area light is circular and oriented
> 3. the sun's diameter should be equal to the shadow squares' height 
> according to Niven's website

Just a check list.
I just noticed that the data I was looking at was from the game. Oops.


> 4. an array of point lights in the shape of a sphere does not suffer 
> from the issues I'm experiencing
> 

How do you do that without them clustering at the poles?


> 
>>> Oh well.
>>>
>>>
>>
>> I like the image especially the shadows cast by the shadow squares in 
>> the solar wind. :)
>>
> 
> Thanks!
> 
> 
>> Unfortunately I don't have CIE.inc or lightsys.inc installed. :(
>>
>>
> 
> I use them in nearly everything. Not sure why. Probably just habit.
> 
> 
Probably a good habit to have.



-- 

Regards
     Stephen

Post a reply to this message

On 08/08/2018 22:12, Alain wrote:
> Le 18-08-07 à 11:20, Stephen a écrit :
> 
>> Thanks Alain, I seldom use area lights and this is the first time I 
>> have used adaptive.
>> When you said samples is that what the offline help names Size_1 & 
>> Size_2, in the line;
>> AXIS_1_VECTOR, AXIS_2_VECTOR, Size_1, Size_2?
>>
> 
> Yes, Size_1 and Size_2.
> 
> AXIS_1_VECTOR and AXIS_2_VECTOR set the extent of the area_light.
> Size_1 and Size_2 set the number of components, or samples, in the 
> area_light.
> 
> When using orient, the pairs need to be of the same size. The 
> AXIS_n_VECTORs need to have the same length and be perpendicular.
> And, you need Size_1 = Size_2.
> 
> When using adaptive, always start with adaptive 0 and only increase if 
> you really need to.
> 
> 
> 

Thanks for the advice Alain


-- 

Regards
     Stephen

Post a reply to this message

On 8/8/2018 10:08 AM, Mike Horvath wrote:
> However, I had to double the height of the shadow squares in order to 
> block out the sun. See the parameter `RWorld_Shadow_Square_Width`. I'm 
> not sure why this was necessary, as the original height was equal to the 
> sun's diameter, and should have fully blocked its light.

I am dumb. Instead of specifying the area light's axis vectors in the 
form of:

     x * foo, y * foo


I was instead simply using:

     foo, foo

which is the same as:

     <foo,foo,foo>, <foo,foo,foo>

which is too long.

I fixed that and things are fine now. Doing another really big render at 
the moment. It might take over a day to render.


Mike

Post a reply to this message

On 8/8/2018 5:35 PM, Stephen wrote:
> On 08/08/2018 22:02, Mike Horvath wrote:
>> On 8/8/2018 1:05 PM, Stephen wrote:
>>> On 08/08/2018 15:08, Mike Horvath wrote:
>>>> Okay, I uploaded my scene to here:
>>>>
>>>>
http://lib.povray.org/searchcollection/index2.php?objectName=Ringworld&version=2.0&contributorTag=SharkD

>>>>
>>>>
>>>> I also attached my latest render to this post. The shadow squares' 
>>>> widths now match what was apparently stated in the books. The 
>>>> shadowed regions on the ring look much better now.
>>>>
>>>> However, I had to double the height of the shadow squares in order 
>>>> to block out the sun. See the parameter 
>>>> `RWorld_Shadow_Square_Width`. I'm not sure why this was necessary, 
>>>> as the original height was equal to the sun's diameter, and should 
>>>> have fully blocked its light.
>>>>
>>>
>>> Do the two area_light vectors <Axis_1> and <Axis_2> extend beyond the 
>>> star's volume?
>>>
>>
>> 1. the axis lengths are equal to the sun's diameter

Oops! I screwed up here. See my other post. They're fixed now.



>> 2. the area light is circular and oriented
>> 3. the sun's diameter should be equal to the shadow squares' height 
>> according to Niven's website
> 
> Just a check list.
> I just noticed that the data I was looking at was from the game. Oops.
> 

Here's a scan from the game:

https://img.fireden.net/tg/image/1464/06/1464060167901.jpg

The shadow squares seem much smaller than they should be.



> 
>> 4. an array of point lights in the shape of a sphere does not suffer 
>> from the issues I'm experiencing
>>
> 
> How do you do that without them clustering at the poles?
> 
> 

I don't. They do cluster! I should investigate a "good enough" sphere 
packing formula.



>>
>>>> Oh well.
>>>>
>>>>
>>>
>>> I like the image especially the shadows cast by the shadow squares in 
>>> the solar wind. :)
>>>
>>
>> Thanks!
>>
>>
>>> Unfortunately I don't have CIE.inc or lightsys.inc installed. :(
>>>
>>>
>>
>> I use them in nearly everything. Not sure why. Probably just habit.
>>
>>
> Probably a good habit to have.
> 
> 
>

Post a reply to this message

Here's a partial render.

I would like to make the edge of the sun a bit sharper, and also make 
the sun smaller while keeping the haze generally the same. Here is the 
sphere I use for this:


/////////////////////////////////////////////////////////////////
#declare RWorld_Corona_Radius			= 95000*4;
#declare RWorld_Corona_Object = sphere
{
     0, RWorld_Corona_Radius
     hollow
     material
     {
         texture
         {
             pigment {rgbt 1}
         }
         interior
         {
             media
             {
                 scattering {1, RWorld_Light_Point_Color/2000}
                 density
                 {
                     function {1/(x*x + y*y + z*z)/RWorld_Corona_Radius}
                     density_map
                     {
                         [0.0 rgb 0]
                         [1.0 rgb 1]
                     }
                     scale RWorld_Corona_Radius
                 }
             }
         }
     }
}
/////////////////////////////////////////////////////////


Any idea on how I can accomplish this?

Post a reply to this message

Attachments:
Download 'sdsadasdsa.png' (273 KB)

Preview of image 'sdsadasdsa.png'

On 9-8-2018 8:07, Mike Horvath wrote:
> Here's a partial render.
> 
> I would like to make the edge of the sun a bit sharper, and also make 
> the sun smaller while keeping the haze generally the same. Here is the 
> sphere I use for this:
> 
> 
> /////////////////////////////////////////////////////////////////
> #declare RWorld_Corona_Radius            = 95000*4;
> #declare RWorld_Corona_Object = sphere
> {
>      0, RWorld_Corona_Radius
>      hollow
>      material
>      {
>          texture
>          {
>              pigment {rgbt 1}
>          }
>          interior
>          {
>              media
>              {
>                  scattering {1, RWorld_Light_Point_Color/2000}
>                  density
>                  {
>                      function {1/(x*x + y*y +
z*z)/RWorld_Corona_Radius}
>                      density_map
>                      {
>                          [0.0 rgb 0]
>                          [1.0 rgb 1]
>                      }
>                      scale RWorld_Corona_Radius
>                  }
>              }
>          }
>      }
> }
> /////////////////////////////////////////////////////////
> 
> 
> Any idea on how I can accomplish this?
> 
> 

I would suggest to play with the settings of the density_map. Not 
tested, but something like:

density_map {
   [0.00 rgb 0.0]
   [0.90 rgb 0.9]
   [0.90 rgb 1.0]
}


-- 
Thomas

Post a reply to this message

On 8/9/2018 2:40 AM, Thomas de Groot wrote:
> I would suggest to play with the settings of the density_map. Not 
> tested, but something like:
> 
> density_map {
>    [0.00 rgb 0.0]
>    [0.90 rgb 0.9]
>    [0.90 rgb 1.0]
> }
> 
> 

Yeah, I thought of that too, but haven't been successful.

I also tried changing the formula so that x y and z are cubed instead of 
squared, but it turned the whole image black!

Meh.


Mike

Post a reply to this message

On 08/08/2018 12:08, Thomas de Groot wrote:

>>
> 
> I often need to take stock again and again of the gigantic distances and 
> sizes we are dealing with in a planetary system. It is something quite 
> difficult to grasp, and rendering things like the Ringworld makes us 
> forget them, reducing them to simple rings to be taken into our hands ;-)
> 

Translating from the mind's eye to reality is always a problem with 
planetary scales.

I wonder if it is possible to use a log rather than a linear scale along 
the axis.

-- 

Regards
     Stephen

Post a reply to this message

On 9-8-2018 10:10, Mike Horvath wrote:
> On 8/9/2018 2:40 AM, Thomas de Groot wrote:
>> I would suggest to play with the settings of the density_map. Not 
>> tested, but something like:
>>
>> density_map {
>>    [0.00 rgb 0.0]
>>    [0.90 rgb 0.9]
>>    [0.90 rgb 1.0]
>> }
>>
>>
> 
> Yeah, I thought of that too, but haven't been successful.
> 
> I also tried changing the formula so that x y and z are cubed instead of 
> squared, but it turned the whole image black!
> 
> Meh.
> 
> 
> Mike

I tested this and it looks acceptable to me:

#local MyScale = 10;

sphere {
   0, 1
   hollow
   texture {pigment {rgbt 1}}
   interior {
     media {
       emission 3/MyScale
       density{
         spherical
         density_map {
           [0.10 rgb <0.0, 0, 0.0>]
           [0.95 rgb <1.0, 1.0, 1.0>*0.2]
           [0.95 rgb <1.0, 1.0, 1.0>]
         }
       }
     }
   }
   scale MyScale
}

Instead of 'spherical', you may need something else more appropriate. I 
have problems with the function you use for the density map.

-- 
Thomas

Post a reply to this message

On 9-8-2018 13:11, Stephen wrote:
> On 08/08/2018 12:08, Thomas de Groot wrote:
> 
>>>
>>
>> I often need to take stock again and again of the gigantic distances 
>> and sizes we are dealing with in a planetary system. It is something 
>> quite difficult to grasp, and rendering things like the Ringworld 
>> makes us forget them, reducing them to simple rings to be taken into 
>> our hands ;-)
>>
> 
> Translating from the mind's eye to reality is always a problem with 
> planetary scales.

Indeed. It is not just a matter of scaling up or scaling down the model. 
More is involved.

> 
> I wonder if it is possible to use a log rather than a linear scale along 
> the axis.
> 

I'll leave that to my betters, but it might be an interesting and 
worthwhile experiment.

-- 
Thomas

Post a reply to this message