Work in progress in this image (as well for the project).
Since i am still learning pov-ray the results are still not great to my taste.
Scaling of objects still not accurate

Features :
-Used lightSysIVb for sun lighting (al=10, illuminating from behind)
-Added high pressure sodium lamp emissive spectrum definition
-objects separated using an halton sequence
-rail ballast usig isosurface based on crackle


Feel free to comment and thank you again to the community for the great and fast
like light support ! ;-)

RodV

Post a reply to this message

Attachments:
Download 'rail.png' (1458 KB)

Preview of image 'rail.png'

> Work in progress in this image (as well for the project).
> Since i am still learning pov-ray the results are still not great to my taste.
> Scaling of objects still not accurate
>
> Features :
> -Used lightSysIVb for sun lighting (al=10, illuminating from behind)
> -Added high pressure sodium lamp emissive spectrum definition
> -objects separated using an halton sequence
> -rail ballast usig isosurface based on crackle
>
>
> Feel free to comment and thank you again to the community for the great and fast
> like light support ! ;-)
>
> RodV
>

I remember your original effort, and the improvement is HUGE!

The high tension pilons could be made taller and spaced more whidely. 
It's reasonable to make them 4 to 6 times higher than the light posts.
Normaly, the cooling towers like those shown are much larger. Normaly, 
they are taller that the high tention lines by a factor of 1.5 to 3 or 4.
The tanks on the right seems correctly scaled relative to the lamps, 
but, I don't remember seeing any so reflective... I don't remember 
seeing any been reflective at all, but always painted with a mostly 
flat, from weathering, paint with the occasional wusty stains.
The R1-R2 (reactor) building can be about any size, so, it's relative 
scalling can't be reliably estimated.
The wind turbines are supposed to be large or huge. They look correct, 
but maybe a little to close to the tanks, or maybe it's just their 
actual size that make them appear that close...




Alain

Post a reply to this message

On 28-6-2012 1:15, rodv92 wrote:
> Feel free to comment and thank you again to the community for the great and fast
> like light support ! ;-)

Adding to what Alain already said:
- The blades of the wind turbines seem too thin and/or too long to me. 
They will not survive any significant wind nor are they efficient enough 
to catch any wind in the first place probably.
- The landscape is in need of some relief. Even a flat desert is not 
"flat". A gentle height_field or isosurface will probably do wonders.
- Textures. Time to give some variations to the textures which are now 
too uniform. Use colour or pigment maps.
- All power lines, including those for the trains should be thinner I 
think, with a thickness variation between the main power line and the 
supporting ones.
- Sunlight is coming from due left (not from behind). Try to avoid such 
geometrically perfect fits to the landscape. I would suggest to move the 
sun towards a position further away giving some interesting shadows on 
the structures (and the landscape) by back lighting.

Otherwise, this is becoming an interesting scene indeed. As Alain said, 
keep in mind the different sizes of the structures, especially in 
relation to each other. Always build from a common scale base (feet, 
meters, pov units) and take into account "real" sizes where possible.

Thomas

Post a reply to this message

Thomas de Groot <tho### [at] degrootorg> wrote:
> On 28-6-2012 1:15, rodv92 wrote:
> > Feel free to comment and thank you again to the community for the great and fast
> > like light support ! ;-)
>
> Adding to what Alain already said:
> - The blades of the wind turbines seem too thin and/or too long to me.
> They will not survive any significant wind nor are they efficient enough
> to catch any wind in the first place probably.
> - The landscape is in need of some relief. Even a flat desert is not
> "flat". A gentle height_field or isosurface will probably do wonders.
> - Textures. Time to give some variations to the textures which are now
> too uniform. Use colour or pigment maps.
> - All power lines, including those for the trains should be thinner I
> think, with a thickness variation between the main power line and the
> supporting ones.
> - Sunlight is coming from due left (not from behind). Try to avoid such
> geometrically perfect fits to the landscape. I would suggest to move the
> sun towards a position further away giving some interesting shadows on
> the structures (and the landscape) by back lighting.
>
> Otherwise, this is becoming an interesting scene indeed. As Alain said,
> keep in mind the different sizes of the structures, especially in
> relation to each other. Always build from a common scale base (feet,
> meters, pov units) and take into account "real" sizes where possible.
>
> Thomas

Thanks Thomas and Alain for your advices.
I am aware of the scaling problems and unfortunately i used some pretty akward
scaling rules, so i think i will code a script to rescale all items.

Since the scale is very high (20 = 1m) and provided that it is a scene with an
animation in mind (point of view of the driving post in the locomotive) the
translation vectors are large, and it poses problem with the size of the sky
dome of LightSys.

It seems clear now that rescaling is mandatory, and maybe it will even increase
parsing performance...

For the desert floor, i think of a f_wrinkle based isosurface and some rock
generating macro, and will use sparse vegetation everywhere.

Thanks !

Post a reply to this message

> On 28-6-2012 1:15, rodv92 wrote:
>> Feel free to comment and thank you again to the community for the
>> great and fast
>> like light support ! ;-)
>
> Adding to what Alain already said:
> - The blades of the wind turbines seem too thin and/or too long to me.
> They will not survive any significant wind nor are they efficient enough
> to catch any wind in the first place probably.
> - The landscape is in need of some relief. Even a flat desert is not
> "flat". A gentle height_field or isosurface will probably do wonders.
> - Textures. Time to give some variations to the textures which are now
> too uniform. Use colour or pigment maps.
> - All power lines, including those for the trains should be thinner I
> think, with a thickness variation between the main power line and the
> supporting ones.
> - Sunlight is coming from due left (not from behind). Try to avoid such
> geometrically perfect fits to the landscape. I would suggest to move the
> sun towards a position further away giving some interesting shadows on
> the structures (and the landscape) by back lighting.
>
> Otherwise, this is becoming an interesting scene indeed. As Alain said,
> keep in mind the different sizes of the structures, especially in
> relation to each other. Always build from a common scale base (feet,
> meters, pov units) and take into account "real" sizes where possible.
>
> Thomas

If you look carefully at real wind turbines, you'll notice that the 
blades are indeed prety thin relative to the lenght.
Here, the bigger difference from real turbines is that the thickest part 
is slightly past the mid point (about 55% radius) while normaly, it's 
much closer to the axis, around 10% radius, and often show a slightly 
pointy bit. Leading edge straight all the way, oposite edge curves 
backward to the thickest point at about 10 to 15% radius, then taper 
down in an almost straight line to the narrow tip.

If the turbines have 3 blades, it's because it proven to be more 
effecient than 4 of the same desing. 3 blades turbines give MORE power 
than 4 bladed ones and 5 bladed ones are not practical: To much drag for 
the power, the pilon need to be almost 3 times bigger than for a 3 
bladed model.
There are some 2 blades desings where the blades are no bigger. They 
give only slightly less power with 1/3 less blade area and need thiner 
and lighter pilons, and there have even been some single, thin, blade 
prototypes built... Those look prety strange with the counter weight.



Alain

Post a reply to this message