I really like it too, it must be a lot of work.

> I'm not really happy with it, from a compositional point of view,
> too many distractions, like the bookcase and the arches,
> but removing those makes the setting dull, so I'm at a loss as
> to what to do to improve it.

Maybe some focal blur, it would show better what to look at and add some 
more realistic looking to the image.
I would also improve a bit the book at first plane, scaling down a bit 
the crakle normal (+ turbulence 0.1) so that it's not so obvious that it 
is the all purpose crackle pattern and add thickness to the cover's 
corners and give them a leather texture.

> best put it away for a while and start something else :-)
Yes, I understand that, it seems that working too much on the same image 
drains out all artistic energy.

JC

Post a reply to this message

I just remember I found this page yesterday :
http://www.wozzeck.net/images/DoF_fr.html
It's a very interesting focal blur settings macro by Francois Dispot.
It's in french but the zip is down the page. It's four years old, but I 
found it yesterday, probably everybody knows it, but it was a discovery 
to me. :-))

JC

JC (Exether) wrote:
> I really like it too, it must be a lot of work.
> 
>> I'm not really happy with it, from a compositional point of view,
>> too many distractions, like the bookcase and the arches,
>> but removing those makes the setting dull, so I'm at a loss as
>> to what to do to improve it.
> 
> 
> Maybe some focal blur, it would show better what to look at and add some 
> more realistic looking to the image.
> I would also improve a bit the book at first plane, scaling down a bit 
> the crakle normal (+ turbulence 0.1) so that it's not so obvious that it 
> is the all purpose crackle pattern and add thickness to the cover's 
> corners and give them a leather texture.
> 
>> best put it away for a while and start something else :-)
> 
> Yes, I understand that, it seems that working too much on the same image 
> drains out all artistic energy.
> 
> JC
>

Post a reply to this message

JC (Exether) wrote:

> I really like it too, it must be a lot of work.

Thanks, yes quite a lot of work, evenings and weekends.

> 
> [...]
> 
> Maybe some focal blur, it would show better what to look at and add some
> more realistic looking to the image.

It's a thought, but the painting on the wall is at the same distance as the 
bookcase, and I want the painting to stay sharp. Maybe a slight ground
fog might work.

> I would also improve a bit the book at first plane, scaling down a bit
> the crakle normal (+ turbulence 0.1) so that it's not so obvious that it
> is the all purpose crackle pattern and add thickness to the cover's
> corners and give them a leather texture.

Agreed. I hadn't really realised how obvious the crackle is, and it doesn't
really look like leather either, just a first approximation.
The corners are already thicker than the book, but only by a tiny amount.
I can increase that.

> [...]
> 
> JC

-- 
Bill Hails

Post a reply to this message

Brilliant work!

-- 
Rick

Kitty5 NewMedia http://Kitty5.com
POV-Ray News & Resources http://Povray.co.uk
TEL : +44 (01270) 501101 - ICQ : 15776037

PGP Public Key
http://pgpkeys.mit.edu:11371/pks/lookup?op=get&search=0x231E1CEA

Post a reply to this message

Hi Bill!

Looks good, but the lighting needs improvement: your sun is an
extended light source, so
 - no shadows of the bars carrying the planets should be visible
   on the desk
 - the disk below the sun (on the desk) should be free of shadows
 - the shadow of the book in the foreground should be very soft
 - more than a half sphere of the planets should be lit (esp. Venus)
 - where is the shadow of the desk on the ground/walls?

Probably the best solution for the lighting is radiosity, but I
haven't worked with that yet.

I also would like so see some indication of how the planets and
the sun are floating above their supports: magnetism? transparent
material? riding on air streaming out of nozzles below the planets? electrostatic
repulsion? magic?

On the desk, between the signs for Pisces and Aries, a rectangular
area looks less "dusty" than the rest of the desk. Is that the
intended meaning?

As a last point I suggest to drastically reduce the camera's angle.
At the moment this angle introduces distortions to the picture
(visible esp. at the painting) which are correct for a viewing
distance of only some centimeters, but look wrong for normal
viewing distances.

If you put it away for a while: not too long, please; I'd like to
see the next steps of its evolution!

   Sputnik

Post a reply to this message

H Frank, thanks for your comments.



> Hi Bill!
> 
> Looks good, but the lighting needs improvement: your sun is an
> extended light source, so
>  - no shadows of the bars carrying the planets should be visible
>    on the desk

I agree, but an area light introduces too vague a shadow on the
rings of saturn. I'd pinned my hopes on a parallel light, but that
requires a pont_at statement :-(
I was trying to avoid letting the planets look too much "in the room"
as opposed to "in space", the harsh shadows on Saturn's rings
help there, I think.

>  - the disk below the sun (on the desk) should be free of shadows
>  - the shadow of the book in the foreground should be very soft
>  - more than a half sphere of the planets should be lit (esp. Venus)

same point vs area light issue, maybe a fairly tight area light would do it.

>  - where is the shadow of the desk on the ground/walls?

I made the table and planets except Saturn, and even the support for
venus no_shadow, otherwise it looks a mess: *lots* of shadows
all over the place. I'd love to think that the next povray release might
have a modifier for no_shadow that specifies a percentage.

> 
> Probably the best solution for the lighting is radiosity, but I
> haven't worked with that yet.

I have radiosity turned on, but that's just to soften the colours.
I've made a few attempts at radiosity without conventional lighting
on this pic but the sun needs lots of output, I'm not sure radiosity
alone can do it.

> I also would like so see some indication of how the planets and
> the sun are floating above their supports: magnetism? transparent
> material? riding on air streaming out of nozzles below the planets?
> electrostatic repulsion? magic?

Magic. :-)
I kind of disagree on this pont, I think the lack of connection is a sort
of statement in itself, the supports are all dead center on their planets
after all.

> 
> On the desk, between the signs for Pisces and Aries, a rectangular
> area looks less "dusty" than the rest of the desk. Is that the
> intended meaning?

well spotted, it's meant to show a previous position of the book.
I should really add some smearing in the dust to the current
position, and should maybe up the density of the dust a touch.

> 
> As a last point I suggest to drastically reduce the camera's angle.
> At the moment this angle introduces distortions to the picture
> (visible esp. at the painting) which are correct for a viewing
> distance of only some centimeters, but look wrong for normal
> viewing distances.

earlier versions had a much tighter angle, but the addition of that
set of columns receding into the distance looked bad, so the
current angle is a compromise.
Maybe it's not an optimal one.

> 
> If you put it away for a while: not too long, please; I'd like to
> see the next steps of its evolution!

Thanks!
I was really getting a bit tired of it, but I can feel a fresh wave
of enthusiasm coming on. I think I should lie down :-)

> 
>    Sputnik

-- 
Bill Hails

Post a reply to this message

Rick [Kitty5] wrote:

> Brilliant work!
> 
Thanks!
-- 
Bill Hails

Post a reply to this message

Bill Hails wrote:

> H Frank, thanks for your comments.
> 

> 
>> Hi Bill!
>> 
>> Looks good, but the lighting needs improvement: your sun is an
>> extended light source, so
>>  - no shadows of the bars carrying the planets should be visible
>>    on the desk
> 
> I agree, but an area light introduces too vague a shadow on the
> rings of saturn. I'd pinned my hopes on a parallel light, but that
> requires a pont_at statement :-(
> I was trying to avoid letting the planets look too much "in the room"
> as opposed to "in space", the harsh shadows on Saturn's rings
> help there, I think.
> 
>>  - the disk below the sun (on the desk) should be free of shadows
>>  - the shadow of the book in the foreground should be very soft
>>  - more than a half sphere of the planets should be lit (esp. Venus)
> 
> same point vs area light issue, maybe a fairly tight area light would do
> it.
> 

I was just paging through the manual and found 6.5.8 "Light Groups"
It looks to be exactly what I need - separate lighting for the planets.

I'm constantly amazed at the maturity of POV-Ray, it's like every
problem I have has already been solved somewhere!

>[...]
>>    Sputnik
> 

-- 
Bill Hails

Post a reply to this message

It has definitely improved.  I really like the prominances on the Sun and I
agree that the rings work very well.

I'm no expert, but here are some things I'd try if I were you:
- Change the number and position of the prominances.  The more I look at the
Sun in this picture, the more the prominances seem imbalanced somehow.
- Change that innermost planet.  It doesn't look like any planet in our
system.
It's too big and shiny to be Mercury and too small and grey to be Venus.
- That sun appears to be the only light in that room, so perhaps the room
should appear darker.  If the room is darker and the dark sides of the
planets show less ambient light, it might improve the look of the room.
Either that or add some other lights.
- Is the sun already an area light?  I think if you made it an area light
approximately the size of Earth and turned "orient" on there might be some
improvement.
- Increase the apparent detail in the wall texture and maybe change the
finish.  Right now it looks like some sort of plastic veneer.
- The far part of the table should appear grayer and less shiny than the
near part.  That's because the surface over there is being viewed at a more
oblique angle, which means there will be a greater density of dust
particles along the line of sight.  I wonder if a thin layer of scattering
media would do the trick.

Just so you know, I don't criticize when I don't like a piece.  (I often
don't criticize even when I do, however.  Laziness.)

Bill Hails wrote:
>I've been fiddling with this old pic for over a month now,
>and I need to start something new, but I thought somebody
>might like it.

Post a reply to this message

Thanks for your comments, and sorry I haven't been keeping a close eye
on this thread :-)
I've been working on it most of my spare time since I last posted, expect
an update soon.

Mack Tuesday wrote:
> It has definitely improved.  I really like the prominances on the Sun and
> I agree that the rings work very well.
> 
> I'm no expert, but here are some things I'd try if I were you:
> - Change the number and position of the prominances.  The more I look at
> the Sun in this picture, the more the prominances seem imbalanced somehow.

I've rotated the sun slightly (15*y). That improves the balance.

> - Change that innermost planet.  It doesn't look like any planet in our
> system.
> It's too big and shiny to be Mercury and too small and grey to be Venus.

dunno, I kind of like it (It's meant to be Venus). I'm not aiming for
scientific accuracy, just an impression.

> - That sun appears to be the only light in that room, so perhaps the room
> should appear darker.  If the room is darker and the dark sides of the
> planets show less ambient light, it might improve the look of the room.
> Either that or add some other lights.

I've gone round and round with the lighting, I hope I've fixed things.

> - Is the sun already an area light?  I think if you made it an area light
> approximately the size of Earth and turned "orient" on there might be some
> improvement.

For my final render I'm using a circular area light the same size as the 
sun, with orient, but it changes the render time from 8-9 hours to 4-5 days 
at 1920x1440 so I'm fixing other stuff before I do that again.

I've also made each planetary body its own light_group with a parallel
light at the sun's center, and fixed up the radiosity.

> - Increase the apparent detail in the wall texture and maybe change the
> finish.  Right now it looks like some sort of plastic veneer.

Terracotta, terracotta :-) Hopefully the lighting fixes will improve that.

> - The far part of the table should appear grayer and less shiny than the
> near part.  That's because the surface over there is being viewed at a
> more oblique angle, which means there will be a greater density of dust
> particles along the line of sight.  I wonder if a thin layer of scattering
> media would do the trick.

That's where I'm at right now, exchanging a semi-transparent disc for a 
cylinder with media, I'm also adding smudges from the book-hole to the 
book, and increasing the density of the dust a bit, to make the marks a bit 
more obvious.

> Just so you know, I don't criticize when I don't like a piece.  (I often
> don't criticize even when I do, however.  Laziness.)

Thanks!  I've been working on this since June, off and on (mostly on since I 
last posted), hopefully the end result will have been worth it.

> Bill Hails wrote:
>>I've been fiddling with this old pic for over a month now,
>>and I need to start something new, but I thought somebody
>>might like it.

-- 
Bill Hails

Post a reply to this message