This image, (not done in POV!!) is close to what I am trying to learn how 
to achieve in POV, or even in a POVMan shader.

The top grid is spherical mapped (2x4) onto a sphere as a displacement.
The grime is derived from the disp map combined with a high-pass filtered 
image of the surface's normals post displacement.
i.e. it is 2 pass. The grime map is used to blend between 2 maps, a clean 
copper and an oxidized one.

Would anyone like to take the displacement map and try to do one of these 
standard grime balls, using POV or MPOV?

Post a reply to this message

Attachments:
Download 'standard_grime_ball_1.jpg' (177 KB)

Preview of image 'standard_grime_ball_1.jpg'

"Daniel Matthews" <lig### [at] bigpondnetau> wrote in message
news:787### [at] 3-enet...
> This image, (not done in POV!!) is close to what I am trying to learn how
> to achieve in POV, or even in a POVMan shader.
>
> Would anyone like to take the displacement map and try to do one of these
> standard grime balls, using POV or MPOV?

Sure, I'll post the animation at povray.binaries.animations after posting
here.

It's not real good, partly because I couldn't get the isosurface perfect
enough, some kind of camera aligned texture offset that I wasn't able to get
rid of without messing up the physical structure of the iso.

I used the grid image you posted for the displacement but it isn't as large
as the one you apparently used.
Anyway, your "grime" sphere seems to have more diversity of texture, mine is
simply a spherical texture map.  Either that or I'm seeing reflections of
the patina in some of the pockets.

Bob H.

Post a reply to this message

Attachments:
Download 'grimey2.jpg' (12 KB)

Preview of image 'grimey2.jpg'

"Bob H." wrote:
> 
> Anyway, your "grime" sphere seems to have more diversity of texture, mine is
> simply a spherical texture map.  Either that or I'm seeing reflections of
> the patina in some of the pockets.

If you carefully look at the sphere, you will notice the image
was probably used on the six faces of the sphere :->

Post a reply to this message

news:3B8FA127.A7734110@atosorigin.com...
> "Bob H." wrote:
> >
> > Anyway, your "grime" sphere seems to have more diversity of texture,
mine is
> > simply a spherical texture map.  Either that or I'm seeing reflections
of
> > the patina in some of the pockets.
>
> If you carefully look at the sphere, you will notice the image
> was probably used on the six faces of the sphere :->

Ahhhh yes...  I see that now.  Doesn't seem to be uniformly mapped onto it
though, still pinched at the top and bottom.

Bob H.

Post a reply to this message

Yes the mapping is not the same, but anything close is fine.
I just thought that a "Standard Grime Ball" was a good way of people 
sharing and evaluating related techniques without the underlying model 
confusing comparisons. To do it correctly there should be a standard 
environment for the ball too.

Your ball is basically the same, but the layers of grime look more 
complex, do you have POV code for it? It would be great if we could set a 
fixed ball and env file/s then have people work on various grime type 
solutions. POV needs to get down and dirty! 
Then POV can have it's own library of filth and corruption.
I hope you don't mind a bit of muck raking?
:o)

Bob H. wrote:


> news:3B8FA127.A7734110@atosorigin.com...
>> "Bob H." wrote:
>> >
>> > Anyway, your "grime" sphere seems to have more diversity of texture,
> mine is
>> > simply a spherical texture map.  Either that or I'm seeing reflections
> of
>> > the patina in some of the pockets.
>>
>> If you carefully look at the sphere, you will notice the image
>> was probably used on the six faces of the sphere :->
> 
> Ahhhh yes...  I see that now.  Doesn't seem to be uniformly mapped onto it
> though, still pinched at the top and bottom.
> 
> Bob H.

Post a reply to this message

If you wish to see the hires image of the ball, I save saved it here 

http://binalong.hypermart.net/grimeball.rar

Warning, the file download is just under 6.7MB !!

This version is a 32bit 2048x2048 TIF image compressed with RAR.
There is a LOT more detail in the full render as I boosted the resolution 
of the displacement surface, it still does not show 100% of the detail on 
the map, but I thought a 40min wait on the parse stage of a scan-line 
render was getting a bit long. :o)

The image shows many small details that reveal the strengths ad weaknesses 
of the method I have described. I am sure that if it was procedural the 
results would be much better. 

Daniel Matthews wrote:

> Yes the mapping is not the same, but anything close is fine.
> I just thought that a "Standard Grime Ball" was a good way of people
> sharing and evaluating related techniques without the underlying model
> confusing comparisons. To do it correctly there should be a standard
> environment for the ball too.
> 
> Your ball is basically the same, but the layers of grime look more
> complex, do you have POV code for it? It would be great if we could set a
> fixed ball and env file/s then have people work on various grime type
> solutions. POV needs to get down and dirty!
> Then POV can have it's own library of filth and corruption.
> I hope you don't mind a bit of muck raking?
> :o)
> 
> Bob H. wrote:
> 

>> news:3B8FA127.A7734110@atosorigin.com...
>>> "Bob H." wrote:
>>> >
>>> > Anyway, your "grime" sphere seems to have more diversity of texture,
>> mine is
>>> > simply a spherical texture map.  Either that or I'm seeing reflections
>> of
>>> > the patina in some of the pockets.
>>>
>>> If you carefully look at the sphere, you will notice the image
>>> was probably used on the six faces of the sphere :->
>> 
>> Ahhhh yes...  I see that now.  Doesn't seem to be uniformly mapped onto
>> it though, still pinched at the top and bottom.
>> 
>> Bob H.

Post a reply to this message

"Daniel Matthews" <lig### [at] bigpondnetau> wrote in message
news:675### [at] 3-enet...
> If you wish to see the hires image of the ball, I save saved it here
>
> http://binalong.hypermart.net/grimeball.rar
>
> Warning, the file download is just under 6.7MB !!

No matter, Cable Modem helps here. ;-)

It's as if the texture is shifted to the left, doesn't look right to me.

Unfortunately I did mine using POV-Ray v3.5, not available yet to the
public.  Don't go e-mailing TAG Team people about it please!  It will be out
someday soon, just be patient.

Anyway, there's some changes to the way it works from MegaPOV so it would be
pointless to post it for now.
I'm not an avid isosurface maker so I can't be sure if I'm not adjusting
things as well as could be.  If you look at the animation I posted to p.b.a.
you can easily see concentric rings that are not a part of the mpeg
encoding.  Can see them in the jpeg here too actually.
The texture_map (and it is just a texture map, spherical type) gets somehow
distorted onto the isosurface surface.  Using a high 'accuracy' (higher is
less in this case) it tends to vanish, and I used 'evaluate' to try and
adjust 'max_gradient' without good results avail.
Same goes for using a better accuracy, the surface becomes fragmented and
adjusting max_gradient doen't help.
But like I say, I'm not a expert at iso's so I don't know if there is a
genuine problem with POV or whether I'm not hitting upon the best
parameters.

Bob H.

Post a reply to this message

Bob H. wrote:

> It's as if the texture is shifted to the left, doesn't look right to me.
> 

I was trying for a smeared on corroder and a buffed back corrosion effect.
But yes at that high resolution there are problems with a hand cooked map.
:o( Hence the need for a procedural approach.

There are actually a group of different grime effects and they are some 
times combined. Some are;

Fallout.
Deposit
Splatter
Gas corrosion
Water mark
Drip and Flow
leach
 
and the related;

Buff and Polish
Scratch and Score
Dent and Ding
Chip
Peel

then there are controls like;

Lost Mask or Grime effect shadow
Directional effect

Can you think of any more?

I guess I can start on a library of ref. photos and some form of 
classification system, while I wait for the new POV and I improve my POV 
skills. If possible, doing this in pure POV code would be best?

Post a reply to this message

Daniel Matthews wrote:
> 
> This image, (not done in POV!!) is close to what I am trying to learn how
> to achieve in POV, or even in a POVMan shader.
> 
> The top grid is spherical mapped (2x4) onto a sphere as a displacement.
> The grime is derived from the disp map combined with a high-pass filtered
> image of the surface's normals post displacement.
> i.e. it is 2 pass. The grime map is used to blend between 2 maps, a clean
> copper and an oxidized one.
> 
> Would anyone like to take the displacement map and try to do one of these
> standard grime balls, using POV or MPOV?

Done with my personal patched version of pov.
It took me 4 days to get the picture mapping on the sphere as your initial
image. I found it rather horrible : on the top of the sphere, there is 4 half
rectangular picture squeezed. But at least, it reproduce your picture.

I may post the image_mapping function if there is interest.

The displacement was made using Warps' tesselation of a sphere with my
current patch in progress... Thanks to you, I have found that I should
optimise it more.

No image map (other than for the displacement) was used. The texture would
need some works, it's just an onion pattern.

Bad things: it's just a mesh. and the displacement suffered from the low
resolution as well as the jpeg original...

Post a reply to this message

Attachments:
Download 'orig.jpg' (99 KB)

Preview of image 'orig.jpg'

Here's mine :-)

Post a reply to this message

Attachments:
Download 'patina_on_copper_orb.jpg' (32 KB)

Preview of image 'patina_on_copper_orb.jpg'