 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
My particle system has had support for flowing for quite a while (since
March 2001), but it used a rather awkward approach. If the particles
collided with a surface with a velocity higher than a certain threshold
value, they would bounce, and otherwise they would flow. This could
produce acceptable results but the method was not very good, nor easy to
use because the user had to find a fitting threshold value.
Now, based on the feedback in my latest thread in povray.advanced-users,
I have changed my particle system to use a different collision method,
which unifies bouncing and flowing into a single simple model. So far,
it seems to work very well. Here's an animation I made with it. Hope you
like it. :)
(The reason that a few of the particles disappear is that I didn't set
their lifespan to be very high.)
And yes, I know, I should use more and smaller particles... ;)
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated May 20)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
Attachments:
Download 'flownew.mpg' (609 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I took this sort of approach in my particle system too, and it works. My
particles only ever bounce, never slide or roll (though they can have very
shallow bounces which makes them look like they are sliding). I don't think that
this will model friction with any great accuracy, which is a drawback.
Is there a noticable speed difference between the two methods?
MJL
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Verrrry nice. Set it up with more particles so we can see what the liquid
really looks like, 'cause i bet it'll be cool =)
(BTW, i assume this is a bunch of blob components scaled in a certain
direction?)
- Slime
[ http://www.slimeland.com/ ]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Mark James Lewin wrote:
> I took this sort of approach in my particle system too, and it works.
> My particles only ever bounce, never slide or roll (though they can
> have very shallow bounces which makes them look like they are
> sliding). I don't think that this will model friction with any great
> accuracy, which is a drawback.
I think that rather quickly the particles will begin to bounce in every
calculation step because the bounces become so small.
The problem is that while the particles bounce, a certain energy loss
should happen per bounce, but when they slide, the energy loss should
happen not per bounce but per time unit, so that the friction is the
same no matter if many or few calculation steps are used. But when there
is no distinction between bouncing and sliding, it can be difficult to
handle. Maybe some fixed threshold based on the incoming angle could be
used, so that when it is very small the particle is considered to be
sliding?
> Is there a noticable speed difference between the two methods?
I haven't compared, but I don't think the difference is very big. The
amount of calculations is much the same as before, so I don't know if
the new method is slower or faster.
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated May 20)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Slime wrote:
> Verrrry nice. Set it up with more particles so we can see what the
> liquid really looks like, 'cause i bet it'll be cool =)
I think it's much the same, but I'm rendering with more particles now.
> (BTW, i assume this is a bunch of blob components
> scaled in a certain direction?)
Yep! Scaled longer in the direction of movement. It's not very
realistic, but it resembles motion blur a little. :)
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated May 20)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Here's the same animation but with 6000 particles instead of 1200. On my
Athlon XP 1700 the last frame took 10 seconds to parse and the total
parse time for the 120 frames were 20 minutes. (Note to self: There's
somthing wrong. This means that all the frames took 10 seconds to parse
each, but the first frames should be much faster because of the fewer
active particles. Something to investigate.)
The total time for parsing plus rendering was 3 hours.
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated May 20)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
Attachments:
Download 'flownew2.mpg' (609 KB)
|
|
| |
| |
|
|
From: Christoph Hormann
Subject: Re: Flowing particles again! ( 608 kbu )
Date: 28 May 2002 07:47:00
Message: <3CF36E34.6175906B@gmx.de>
|
|
|
| |
| |
|
|
Rune wrote:
>
> [...]
>
> And yes, I know, I should use more and smaller particles... ;)
It looks good, but i think more important than the number of particles
would be cohesive and adhesive forces, of course modeling these would
result in an extreme slowdown.
Christoph
--
POV-Ray tutorials, IsoWood include,
TransSkin and more: http://www.tu-bs.de/~y0013390/
Last updated 05 May. 2002 _____./\/^>_*_<^\/\.______
Post a reply to this message
|
|
| |
| |
|
|
From: Rick [Kitty5]
Subject: Re: With 5 times the particles ( 608 kbu again )
Date: 28 May 2002 10:51:49
Message: <3cf39985$1@news.povray.org>
|
|
|
| |
| |
|
|
Rune <run### [at] mobilixnet dk> wrote:
> Here's the same animation but with 6000 particles instead of 1200. On
> my Athlon XP 1700 the last frame took 10 seconds to parse and the
> total parse time for the 120 frames were 20 minutes. (Note to self:
There's
> somthing wrong. This means that all the frames took 10 seconds to
> parse each, but the first frames should be much faster because of the
> fewer active particles. Something to investigate.)
looks mcuh better, especially the first 'splash', how about an over nighter
with 120000 :P
--
Rick
Kitty5 WebDesign - http://Kitty5.com
POV-Ray News & Resources - http://Povray.co.uk
TEL : +44 (01270) 501101 - FAX : +44 (01270) 251105 - ICQ : 15776037
PGP Public Key
http://pgpkeys.mit.edu:11371/pks/lookup?op=get&search=0x231E1CEA
Post a reply to this message
|
|
| |
| |
|
|
From: Rune
Subject: Re: With 5 times the particles ( 608 kbu again )
Date: 28 May 2002 14:03:26
Message: <3cf3c66e@news.povray.org>
|
|
|
| |
| |
|
|
Rick [Kitty5] wrote:
> looks mcuh better, especially the first 'splash'
Thanks!
> how about an over nighter with 120000 :P
Sure, I'll send you the scene files... ;)
Rune
--
3D images and anims, include files, tutorials and more:
Rune's World: http://rsj.mobilixnet.dk (updated May 20)
POV-Ray Users: http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Ring: http://webring.povray.co.uk
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Better, but it still looks a little flawed. You know what might be an
interesting thing to do? Give the blob's strengths that are less than the
threshold. That way, when blobs go off by themselves, they'll be invisible,
but the only place you'll see liquid is where there's a significant amount
of it in one place.
- Slime
[ http://www.slimeland.com/ ]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
