Christoph Hormann schrieb in Nachricht <3A9A8561.61198559@gmx.de>...
>It looks nice,

Thanks!

>but not much like radiosity, because the shadows have very
>constant brightness.
>

Hm, it takes a fair bit longer when I set it to use rad, so I think it does
it's job. ;-) Perhaps my other lights are to strong. I had to position one
light_source at the camera position, otherwise the details under the well's
roof wouldn't be visible.


>Maybe add some kind of wooden vehicle.
>

I thought the same sometimes while making the scene. Perhaps it would give
the scene more life.

Marc-Hendrik

Post a reply to this message

Greg M. Johnson schrieb in Nachricht <3A9A7A8A.8F377821@my-dejanews.com>...
>Each piece is perfect by itself, but the overall composition (color
>balance/contrast) is a bit off.
>


Perhaps you are right. I'm not that satisfied with it myself. Any idea, what
I could do to improve it? Should I add some more details like flower-pots or
even a market-booth - and bring some more colors in that way? Or should I
change the colors itself?

Any help would be greatly appreciated!

Marc-Hendrik

Post a reply to this message

David Fontaine schrieb in Nachricht <3A99BFDF.5ED2C4A1@faricy.net>...
>Nice scene. I think you have it down quite well (no pun intended). The
technical
>detail is good (is that just normal, or isosurface or something?),

Thank you. Most of it are isosurfaces, some 26000 for the fibonacci pattern
and some more for the rest. The bricks are a derivate of the one's I've
posted some weeks ago. If you are intressted in something special, I'll be
glad to share the code for it (the whole scene is a bit to messy to post).

>and the
>composition is very nice.

Good to know you like it. I'm not satisfied myself but I can't tell why.
Perhaps it's the lighting or the colors. I just don't know, but it has not
the same quality as "The dancing cur" IMO.

>Great use of the fibanocchi spirals!
>

I was glad I found the code for it, as I wanted to make some good ground
this time. The height of the stones follows a function, which is quite
flexible and I used the same function for an Iso that extends in the
background, 'cause my system could not stand more of the stones.

Marc-Hendrik

Post a reply to this message

Marc-Hendrik Bremer wrote:
> 
> Hm, it takes a fair bit longer when I set it to use rad, so I think it does
> it's job. ;-) Perhaps my other lights are to strong. I had to position one
> light_source at the camera position, otherwise the details under the well's
> roof wouldn't be visible.
> 

With radiosity, additional light sources should usually be used very
sparsely. A light source at the camera position is no good idea IMO. If
you use appropriate radiosity settings (recursion_limit >= 2) and
finishes, things should work.

Christoph

-- 
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other 
things on: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Christoph Hormann schrieb in Nachricht <3A9A938C.F97101B9@gmx.de>...
>With radiosity, additional light sources should usually be used very
>sparsely. A light source at the camera position is no good idea IMO. If
>you use appropriate radiosity settings (recursion_limit >= 2) and
>finishes, things should work.


I'm not happy with that light either, but I couldn't make it work. it's just
color 0.2, but makes a big difference.  Recursion_limit is at 3. The rest is
set to the default values (according to your radiosity-tutorial, but I
copied the settings in the radiosity block). My idea was that the dark slate
roof absorbs all the radiosity light. Is that possible?
I didn't care much about the finishes yet. ambient is set to 0 of course
(using MegaPov 0.6a). I once tried diffuse 1 which did not much difference
to the default 0.65. Should I use different settings?

Marc-Hendrik

Post a reply to this message

Marc-Hendrik Bremer wrote:
> 
> I'm not happy with that light either, but I couldn't make it work. it's just
> color 0.2, but makes a big difference.  Recursion_limit is at 3. The rest is
> set to the default values (according to your radiosity-tutorial, but I
> copied the settings in the radiosity block). My idea was that the dark slate
> roof absorbs all the radiosity light. Is that possible?

Of course it absorbs a lot of light, but there should still be a lot of
indirect light from the ground.  

> I didn't care much about the finishes yet. ambient is set to 0 of course
> (using MegaPov 0.6a). I once tried diffuse 1 which did not much difference
> to the default 0.65. Should I use different settings?
> 

There is not much contrast in your picture in general for an outside
sunlight scene.  Maybe increase the main light source.  Decreasing
error_bound in the radiosity block should make the radiosity effect more
differentiated, but it will not affect the overall brightness.

A sample picture without the additional light sources might help.  

Christoph

-- 
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other 
things on: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Christoph Hormann schrieb in Nachricht <3A9AA9E6.A6CCBDED@gmx.de>...

>
>A sample picture without the additional light sources might help.


OK, here is a tiny picture without the extra light (and aa). I hope it's
enough to see what's going on. It is much better than I remembered, but I
changed some things since I tried it the last time without that additional
light. And the shadows are still of constant brightness.

I still have to try your other suggestions. I'll also try to replace
background {rgb <0.4,0.4,0.6>} with a big sphere with ambient 0.5 or
something like that.

Marc-Hendrik

Post a reply to this message

Attachments:
Download 'verwaltung-klein.jpg' (43 KB)

Preview of image 'verwaltung-klein.jpg'

Marc-Hendrik Bremer wrote:
> 
> OK, here is a tiny picture without the extra light (and aa). I hope it's
> enough to see what's going on. It is much better than I remembered, but I
> changed some things since I tried it the last time without that additional
> light. And the shadows are still of constant brightness.

It looks as i would have expected it, not too dark IMO.  Maybe make the
main light slightly brighter.  

> 
> I still have to try your other suggestions. I'll also try to replace
> background {rgb <0.4,0.4,0.6>} with a big sphere with ambient 0.5 or
> something like that.
> 

That's probably a good idea, ambient 1 diffuse 0 is usual finish.  I never
tried radiosity with only background, even don't know if it has any
effect.  

Christoph

-- 
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other 
things on: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

I think this is amazing. Thanks for sharing. 

-- 
Cheers
Steve              email mailto:ste### [at] zeroppsuklinuxnet

%HAV-A-NICEDAY Error not enough coffee  0 pps. 

web http://www.zeropps.uklinux.net/

or  http://start.at/zero-pps

  9:47pm  up 24 days, 23:28,  2 users,  load average: 1.62, 1.20, 1.06

Post a reply to this message

Steve schrieb ...
>I think this is amazing. Thanks for sharing.
>


Thank you!
I'm glad that I'm able to share it here, where people care to comment about
it. Without all the suggestions and the help from all this kind folk here, I
don't think that I would be able to make scenes like this. Take the
fibonacci tiling - I did not even know what's the name for this until Alf
and Tor posted there 'fibonacci series'.

Marc-Hendrik

Post a reply to this message