 |
 |
|
 |
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Thanks for all the comments on what grass looked best!
I've tweaked and corrected a lot of things and made the grass a bit more
saturated. The flowers may still look like they're floating, but there's not
much to do about that I'm afraid.
But anyway, almost everybody seemed to like better the image produced by the
new technique I'm working on rather than that produced by my old Fast Grass
Include File. I'm very pleased that the new include file will not only have
the advantage that it works in animations, but also that it looks better in
general.
The catch is that it renders about 4 times slower than the old technique,
but it's still pretty fast. The attached image was rendered at 800x600 AA0.3
in about 8 minutes on my Athlon XP 1700+. As I have never tried using mesh
grass, I don't know how slow that is for comparison. Anyone got any idea how
long an equivalent image with mesh grass would take?
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated Feb 16)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
Attachments:
Download 'fgrass2_test.jpg' (87 KB)
Preview of image 'fgrass2_test.jpg'

|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Superb! Really good! I think meshes will at least use more memory resources,
even with all currently possible tricks.. About the rendertime, it depends a
lot on AA settings.. Did you use standard AA? If so, the grass texture is
not terribly fast.
Regards,
Hugo
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"Hugo" wrote:
> Superb! Really good!
Woohoo! Thanks. :)
> I think meshes will at least use more memory resources,
> even with all currently possible tricks..
Yeah, if nothing else, my method takes up practically zero memory,
completely independent on how big the field of grass is.
> About the rendertime, it depends a lot on AA settings..
> Did you use standard AA? If so, the grass texture is
> not terribly fast.
No, that's what I was afraid. I did use standard AA. I guess I'll have to
try to use some of the mesh grass macros around and see the pros and cons
myself.
Which mesh grass macro is the best for a curvy height_field like this?
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated Feb 16)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
I wrote:
> > About the rendertime, it depends a lot on AA settings..
> > Did you use standard AA? If so, the grass texture is
> > not terribly fast.
>
> No, that's what I was afraid. I did use standard AA. I
> guess I'll have to try to use some of the mesh grass
> macros around and see the pros and cons myself.
I tried to use Gilles Tran's MakeGrass macro to create a whole field of real
mesh grass, and it rendered twice as fast as my faked texture grass. :(
However, Gilles' macro can only create grass on flat surfaces AFAIK. Since
using many copies of the same mesh isn't possible to the same extent when
covering hills with grass, maybe doing that with mesh grass would still be
less efficient, or even impossible on many machines?
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated Feb 16)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
JRG's scatter macro can do curved surfaces, I'm working on a similar idea
with grass support
(see thread Vegetation Project WIP 3/27/02 by Tim Nikias)
--
__________________
RAY
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"Rune" wrote:
> Which mesh grass macro is the best for a curvy height_field like this?
Try Gilles' macro. Of course you have to modify it, as it presently does not support
heightfields. AFAIK at least one pover (Tony[B]?) has already done that. So ask him.
:)
--
Jonathan.
Home: http://digilander.iol.it/jrgpov
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"RAY" wrote:
> JRG's scatter macro can do curved surfaces, I'm working on a similar idea
> with grass support
> (see thread Vegetation Project WIP 3/27/02 by Tim Nikias)
You are probably referring to my old clutter macro. The last scatter macro I wrote is
not very suitable for such a task.
--
Jonathan.
Home: http://digilander.iol.it/jrgpov
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
> You are probably referring to my old clutter macro. The last scatter macro
I wrote is
> not very suitable for such a task.
Oops.
--
__________________
RAY
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
This grass looks a lot different than and mesh grass that I have seen. And
so, for that reason, has a lot of value even if it is slower.
-Shay
Rune <run### [at] mobilixnet dk> wrote in message
news:3ca674d4@news.povray.org...
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"JRG" wrote:
> "Rune" wrote:
> > Which mesh grass macro is the best for a
> > curvy height_field like this?
>
> Try Gilles' macro. Of course you have to modify it,
> as it presently does not support heightfields.
> AFAIK at least one pover (Tony[B]?) has already
> done that. So ask him. :)
Have done, but he seems to may have lost it...
Anyway, I tried some grass made by Alan Holding which works for
height_fields.
When using a medium sized field of grass, his mesh grass renders about as
fast as my faked grass, when taking into account the parsing. Both images
were finished in two minutes at 320x240. However, when increasing the number
of blades, the parse time of the mesh grass increases very much, and
exceeded half an hour in my test. The larger the field, the higher memory
requirement and longer parsing time there is for mesh grass. In contrast, my
faked grass has no parse time at all, nor any significant memory
requirement, no matter the size of the field. So my faked grass is best
suited for large fields of grass.
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated Mar 19)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
Have you tried the combo mesh grass in the foreground and your textured grass in the
background? Would they match well?
--
Jonathan.
Home: http://digilander.iol.it/jrgpov
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
> Have done, but he seems to may have lost it...
I didn't say I'd lost it, I just said I didn't know where I last saved it! I
still have it. As a matter of fact, I found it this morning and started
getting it back into working shape. I'll send it to you later today.
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"Tony[B]" wrote:
> I didn't say I'd lost it, I just said I didn't
> know where I last saved it!
That's why I said "may"! ;)
> I still have it. As a matter of fact, I found it
> this morning and started getting it back into
> working shape. I'll send it to you later today.
Thanks, I'd be interested in trying it out and do some comparing. :)
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated Mar 19)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"Rune" <run### [at] mobilixnet dk> a écrit dans le message de news: >
However, Gilles' macro can only create grass on flat surfaces AFAIK. Since
> using many copies of the same mesh isn't possible to the same extent when
> covering hills with grass, maybe doing that with mesh grass would still be
> less efficient, or even impossible on many machines?
Actually it's not very difficult for a good coder to modify the code and
make it "drop" the original grass patch on a surface following a grid,
thanks to the trace function and to the reorient macro. Check the slope
angle (to avoid putting patches where they shouldn't grow), add some
randomness and that's all.
http://www.oyonale.com/ldc/images/arbreteledetail1.jpg
The code was written for this specific picture and is not generic. I can
send it to you, it's just very ugly. Mick Hazelgrove did that with his own
macro too I guess. Note that I'm talking about covering the surface by
repeating a square, flat grass patch, not a patch that would fit the
underlying surface. The trick to optimize this is to make it in 2 steps, one
that creates the patch file and one that creates the patches coordinates
file. Then I use the coordinates file to save on the parsing time or even
add other layers of objects.
It is going to be efficient anyway because meshes are always fast.
I attached a test I did some years ago using a similar routine, but with a
car instead of a grass patch.
Your macro may be easier to use though and better-looking in some cases (in
fact I just have a project where it's going to be handy).
G.
--
**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters
Post a reply to this message
Attachments:
Download 'trafic03a.jpg' (43 KB)
Preview of image 'trafic03a.jpg'

|
 |
|  |
|  |
|
 |
|
 |
|  |
|  |
|
 |
"Gilles Tran" wrote:
> Actually it's not very difficult for a good coder to
> modify the code and make it "drop" the original grass
> patch on a surface following a grid, thanks to the
> trace function and to the reorient macro.
Aha, I thought that flat patches would not fit well on curved surfaces, so
that each blade of grass would have to be placed individually, but
apparently small flat patches fit on curved surfaces just fine.
> It is going to be efficient anyway because meshes
> are always fast.
But even when optimized, the memory requirement (and parse time) can become
huge when working with very large fields of grass. That's my niche I
guess... ;)
> Your macro may be easier to use though and better-looking
> in some cases
I don't know about the better-looking part, but it may be a bit easier to
use.
> (in fact I just have a project where it's
> going to be handy).
I'm honored! I'll try to get the include file finished within this week.
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated Mar 19)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
|
 |
|  |
|  |
|
 |
|
 |
|  |