Odd challenge...

  • Follow


Someone asked this the other day elsewhere, and I haven't come up with a 
good solution yet.

What would be the best way to model a spider's web?

-- 
Gareee�
(Gary Tabar Jr.)


0
Reply Gareee 5/28/2006 6:58:44 PM

Gareee� wrote:

> Someone asked this the other day elsewhere, and I haven't come up with a 
> good solution yet.
> 
> What would be the best way to model a spider's web?
> 

Flat Disc, select poly, bevel out, out, out, out ,out. Select 
everything, bevel without moving, delete. Unify polygons.

This should give you the basic form in 2 point polygons.

Send it to Layout and play with some Cloth dynamics to get a little bit 
of sag.

Maybe?

Either that or just draw one and map it onto a polygon.

jw.
0
Reply James 5/28/2006 10:40:53 PM


i think the difficult bit is getting the strands to spiral out from the 
centre (i think thats how they work) and would you want the mesh to be 
solid, or a basic, no depth model? oo, am gonna have a go at it..!

i never tried cloth dynamics, is it easy enough to get on with?, whats a 
basic setup, to do the sag thing? is it with weight maps?

Iain


"James Willmott" <eatspam@mcdonalds.com> wrote in message 
news:VFpeg.12762$S7.7029@news-server.bigpond.net.au...
> Gareee� wrote:
>
>> Someone asked this the other day elsewhere, and I haven't come up with a 
>> good solution yet.
>>
>> What would be the best way to model a spider's web?
>>
>
> Flat Disc, select poly, bevel out, out, out, out ,out. Select everything, 
> bevel without moving, delete. Unify polygons.
>
> This should give you the basic form in 2 point polygons.
>
> Send it to Layout and play with some Cloth dynamics to get a little bit of 
> sag.
>
> Maybe?
>
> Either that or just draw one and map it onto a polygon.
>
> jw. 


0
Reply K 5/29/2006 11:22:20 AM

"K.Gari" <inswls@hotmail.com> wrote in message 
news:MPAeg.7482$sX1.3487@newsfe1-gui.ntli.net...
>i think the difficult bit is getting the strands to spiral out from the 
>centre (i think thats how they work) and would you want the mesh to be 
>solid, or a basic, no depth model? oo, am gonna have a go at it..!
>
> i never tried cloth dynamics, is it easy enough to get on with?, whats a 
> basic setup, to do the sag thing? is it with weight maps?

No real "desires" at all.. It's just one of those interesting things to try.

I started with a sphere, deleted 1/2, flattened that, and then beveled that, 
deleting the original polys.

Still, that looks to "mechanical" and not very web like.


-- 
Gareee�
(Gary Tabar Jr.)


0
Reply Gareee 5/29/2006 1:10:30 PM

just had a look at a web outside. it really wasnt uniform at all. really 
random, like a smashed plane of 3 n 4 sided polys. i think jittering might 
loosen it all up a bit and be more organic. still gonna try it...gettin 
sidetracked watchin films..doh

iain


"Gareee�" <gareee-no-spamm@myprovider.com> wrote in message 
news:gpCeg.25$wd3.22@fe04.lga...
> "K.Gari" <inswls@hotmail.com> wrote in message 
> news:MPAeg.7482$sX1.3487@newsfe1-gui.ntli.net...
>>i think the difficult bit is getting the strands to spiral out from the 
>>centre (i think thats how they work) and would you want the mesh to be 
>>solid, or a basic, no depth model? oo, am gonna have a go at it..!
>>
>> i never tried cloth dynamics, is it easy enough to get on with?, whats a 
>> basic setup, to do the sag thing? is it with weight maps?
>
> No real "desires" at all.. It's just one of those interesting things to 
> try.
>
> I started with a sphere, deleted 1/2, flattened that, and then beveled 
> that, deleting the original polys.
>
> Still, that looks to "mechanical" and not very web like.
>
>
> -- 
> Gareee�
> (Gary Tabar Jr.)
>
> 


0
Reply K 5/29/2006 1:17:53 PM

It depends on the spider :-)  Charles

K.Gari wrote:
> just had a look at a web outside. it really wasnt uniform at all. really 
> random, like a smashed plane of 3 n 4 sided polys. i think jittering might 
> loosen it all up a bit and be more organic. still gonna try it...gettin 
> sidetracked watchin films..doh
> 
> iain

0
Reply Chilla 5/29/2006 5:24:11 PM

Chilla wrote:

> It depends on the spider :-)  Charles

Yeah, cuz a Black Widow's web, looks like crap LOL
Just a bunch of strands going all over the place in seemingly no pattern 
at all.
And of course, those common types that make their webs that look more 
like carpet and a tube or funnel.
I used to love to drop other bugs in their webs when I was a kid and 
watch the action LOL
Of course, I had pet Ant Lions and other pet bugs too :)
I am curious to see what you all come up with though, as this is just one
of those fun little projects to mess around with for no reason, except 
to play :)

...................md :)

>
> K.Gari wrote:
>
>> just had a look at a web outside. it really wasnt uniform at all. 
>> really random, like a smashed plane of 3 n 4 sided polys. i think 
>> jittering might loosen it all up a bit and be more organic. still 
>> gonna try it...gettin sidetracked watchin films..doh
>>
>> iain
>
>


-- 
-- 
Check out my Tutorials:

MD arts
Mark Dunakin
md@md-arts.com
http://www.md-arts.com 
0
Reply Mark 5/29/2006 6:33:12 PM

Checkout the following topic links:

http://forums.cgsociety.org/showthread.php?t=218799
http://www.digitalproducer.com/pages/creating_a_spider_web_with_light.htm

Jim


"Gareee�" <gareee-no-spamm@myprovider.com> wrote in message 
news:bnneg.351$Ba4.52@fe07.lga...
> Someone asked this the other day elsewhere, and I haven't come up with a 
> good solution yet.
>
> What would be the best way to model a spider's web?
>
> -- 
> Gareee�
> (Gary Tabar Jr.)
>
> 


0
Reply Jim 5/29/2006 8:31:45 PM

"Jim Woodruff" <me@sayno2spam.com> wrote in message 
news:GUIeg.1886$f76.569@dukeread06...
> Checkout the following topic links:
>
> http://forums.cgsociety.org/showthread.php?t=218799
> http://www.digitalproducer.com/pages/creating_a_spider_web_with_light.htm
>
> Jim

Nice to know I'm not the only one that was perplexed by this project.

Thanks for those links!


-- 
Gareee�
(Gary Tabar Jr.)


0
Reply Gareee 5/29/2006 9:08:16 PM

Gareee wrote:
> Someone asked this the other day elsewhere, and I haven't come up with a 
> good solution yet.
>
> What would be the best way to model a spider's web?
>
>   
I had some luck using edge bevel with a cone. After merging a bunch of 
points, and deleting polys with more than 2 points, I was left with a 
lot of 2point polys as expected.

Only thing: how would I add extra points to the polys (automatically) to 
allow for sagging, if the web is wet� as with dew?

-- 
��: }	theartist
0
Reply theartist 5/30/2006 10:30:49 PM

Gareee wrote:
> Someone asked this the other day elsewhere, and I haven't come up with a 
> good solution yet.
>
> What would be the best way to model a spider's web?
>
>   
I have found what works for me is creating a few splines and extruding 
them like this:

draw out a spline with a number of points on one axis
do a radial array with a lot of iterations
then select all the points in order using a spiral motion from the 
center (except the ones on the outer ring)
create a second spline from those points.
At this time, select the spiral spline, and give it a surface of 
"in-betweens".
inverse your selection, and select all the pole splines.
create a new surface for these splines named "poles"
select all your points, and all your splines.
split polys (map this to a key, and hold it down)
At this point you'll have a bunch of 2-point splines.
(you could jitter with scaling 10% on 2-axis here)
Then use the magnet on it to stretch it around on one axis.
Extrude perpendicular to the axis of the array.
set point values to the array axis (this will squash everything flat on 
"z" for instance)
merge points and unify polys (you should have many 2-pt polys)
crank LW, set some cloth effects using silk, and you're good to go.

I set mine to effect the y axis of a "web" built on the z. it sagged the 
silk appropriately, and should only take a little while to set up.

-- 
��: }	theartist
0
Reply theartist 5/31/2006 7:02:33 AM

Hmmm...

Well, I actually had to do spider webs for a Halloween project a couple 
years ago...  I used flat planes subdivided into tons of little triangles - 
double sided.  Then I mapped an image of a spider web to both sides and used 
cloth simulation to give it some saggyness and interaction with some wind. 
I made a few strips of triangles with more mapped web pieces that could wave 
in the breeze.

Now, if you actually wanted to physically model the web, complete with all 
those hundreds (or thousands, depending on the type of spider web), it seems 
silly to manually do this.  I would write an LScript or small application to 
generate the mesh...  Something to map out all the points where the silk 
strands would intersect and what strands would be suspended between which 
points.  I would generate the mesh out of splines and then there's lots of 
other things that can be done with it to turn it into all points or polygons 
or whatever.

If you're just doing one or two webs as a one time shot and you don't need 
them to be very complex, then I'd just build them by hand (because it would 
take some time to write the above script/application).  Theartist's extruded 
spline approach would probably be a good way to start.

"theartist" <theartist@[remove]ev1.net> wrote in message 
news:dcbfg.683$0v4.490@tornado.texas.rr.com...
> Gareee wrote:
>> Someone asked this the other day elsewhere, and I haven't come up with a 
>> good solution yet.
>>
>> What would be the best way to model a spider's web?




0
Reply Dark 5/31/2006 5:52:24 PM

> Now, if you actually wanted to physically model the web, complete with all 
> those hundreds (or thousands, depending on the type of spider web), it seems 
> silly to manually do this. 

Not very hard.  Make a SPP (single point poly) and Lathe it by turning it 7,200
degrees, with 960 segments, and offset on Z 1 meter.  Apply Taper Evenly to
scrunch in the center, then Set Value to 0 on Z.  Bingo, instant spiral of 2 pt
polys.

Next, and this is the more tedious manual stuff, select two points and hit P
to make a poly segment and continue out from the center to make 18 or so randomly
spaced spokes.  Use the Make Points tool to add points to extend 6 of those spokes
and then add a few random runners between these size main spokes for the remaining
spokes to attach to.

Oh heck... here's my quick 5 minute spider web object and 1 hr tweak job of
applying ClothFX.  Had to make a radial gradient weight map to get more
"spring" tension in toward the center (applied to the Compress Stretch param).

Scene and object is in the zip at:  http://chrusion.com/public_files



Dean A. Scott, mfa
---------------------------------------
School of Visual Art and Design
southern adventist university
---------------------------------------
http://www.southern.edu/~dascott

-- 
Posted via a free Usenet account from http://www.teranews.com

0
Reply Dean 6/2/2006 7:32:30 PM

Dean A. Scott wrote:
>
>> Now, if you actually wanted to physically model the web, complete 
>> with all those hundreds (or thousands, depending on the type of 
>> spider web), it seems silly to manually do this. 
>
> Not very hard.  Make a SPP (single point poly) and Lathe it by turning 
> it 7,200
> degrees, with 960 segments, and offset on Z 1 meter.  Apply Taper 
> Evenly to
> scrunch in the center, then Set Value to 0 on Z.  Bingo, instant 
> spiral of 2 pt
> polys.
>
> Next, and this is the more tedious manual stuff, select two points and 
> hit P
> to make a poly segment and continue out from the center to make 18 or 
> so randomly
> spaced spokes.  Use the Make Points tool to add points to extend 6 of 
> those spokes
> and then add a few random runners between these size main spokes for 
> the remaining
> spokes to attach to.
>
> Oh heck... here's my quick 5 minute spider web object and 1 hr tweak 
> job of
> applying ClothFX.  Had to make a radial gradient weight map to get more
> "spring" tension in toward the center (applied to the Compress Stretch 
> param).
>
> Scene and object is in the zip at:  http://chrusion.com/public_files
>
>
>
> Dean A. Scott, mfa
> ---------------------------------------
> School of Visual Art and Design
> southern adventist university
> ---------------------------------------
> http://www.southern.edu/~dascott
>
Nice moves.  Good solution having your web stretch asymmetrically.
 Looks like you set up a weight-map to control your cloth dynamics. Care 
to set up a different thread and talk about that?  I'm new to dynamics

Thanks,

-- 
��: }	theartist
0
Reply theartist 6/2/2006 8:53:25 PM

Dean A. Scott wrote:
> 
>> Now, if you actually wanted to physically model the web, complete with 
>> all those hundreds (or thousands, depending on the type of spider 
>> web), it seems silly to manually do this. 
> 
> Not very hard.  Make a SPP (single point poly) and Lathe it by turning 
> it 7,200
> degrees, with 960 segments, and offset on Z 1 meter.  Apply Taper Evenly to
> scrunch in the center, then Set Value to 0 on Z.  Bingo, instant spiral 
> of 2 pt
> polys.
> 
> Next, and this is the more tedious manual stuff, select two points and 
> hit P
> to make a poly segment and continue out from the center to make 18 or so 
> randomly
> spaced spokes.  Use the Make Points tool to add points to extend 6 of 
> those spokes
> and then add a few random runners between these size main spokes for the 
> remaining
> spokes to attach to.
> 
> Oh heck... here's my quick 5 minute spider web object and 1 hr tweak job of
> applying ClothFX.  Had to make a radial gradient weight map to get more
> "spring" tension in toward the center (applied to the Compress Stretch 
> param).
> 
> Scene and object is in the zip at:  http://chrusion.com/public_files
> 
> 
> 
> Dean A. Scott, mfa
> ---------------------------------------
> School of Visual Art and Design
> southern adventist university
> ---------------------------------------
> http://www.southern.edu/~dascott
> 
Thanks, I'm saving that one for later.
0
Reply Bill 6/2/2006 9:21:47 PM


> Looks like you set up a weight-map to control your cloth dynamics. Care 
> to set up a different thread and talk about that?  I'm new to dynamics

I'm no expert by ANY means!  The Wmap was just an experiment that happened
to work.  Thinking logically, I said, I wonder if these little FIX buttons
allow you to load a weight map?  I made one, reloaded the object, and Yep,
you can, so I said, Hmmm... I bet if I make the center of the web compress
more than the outside, will it draw the web inward?  Made a simple spot of
red (100% value) in the middle of a blue solid (-100% value) and tested.

Hey!  It worked!  A little too good.  So I repainted the map to make a smoother
gradient from red to gray to blue using the Airbrush tool set to 2% intensity
(in numeric panel) and a huge size to create a nice falloff effect as I wiggled
the mouse around the center.

If you open up the object, you'll see I made a subdivided plane as a guide to
see what I was painting as you can't see the map in Weight Map view mode on
2 pt polys.

Then, just by twiddling spring, viscosity, and compress stretch values, I was
able to get the web to settle in a natural way.  I heard/read that a Polygon
Stretch value of Zero doesn't work and to use a small fractional value to prevent
the major drooping that clothfx gives you by default.  Next a little substructure
was added and a higher viscosity to tighten up the sag.  Obviously webs don't
weigh much, so the default mass value of 1.0 was cut by 100 or so.

Again, all this is via experimentation, changing each param one at a time
to see what it does, tweaking another, and so on until the visual effect is
achieved.





Dean A. Scott, mfa
---------------------------------------
School of Visual Art and Design
southern adventist university
---------------------------------------
http://www.southern.edu/~dascott

-- 
Posted via a free Usenet account from http://www.teranews.com

0
Reply Dean 6/3/2006 12:10:53 AM

Dean A. Scott wrote:
>
>
>> Looks like you set up a weight-map to control your cloth dynamics. 
>> Care to set up a different thread and talk about that?  I'm new to 
>> dynamics
>
> I'm no expert by ANY means!  The Wmap was just an experiment that 
> happened
> to work.  Thinking logically, I said, I wonder if these little FIX 
> buttons
> allow you to load a weight map?  I made one, reloaded the object, and 
> Yep,
> you can, so I said, Hmmm... I bet if I make the center of the web 
> compress
> more than the outside, will it draw the web inward?  Made a simple 
> spot of
> red (100% value) in the middle of a blue solid (-100% value) and tested.
>
> Hey!  It worked!  A little too good.  So I repainted the map to make a 
> smoother
> gradient from red to gray to blue using the Airbrush tool set to 2% 
> intensity
> (in numeric panel) and a huge size to create a nice falloff effect as 
> I wiggled
> the mouse around the center.
>
> If you open up the object, you'll see I made a subdivided plane as a 
> guide to
> see what I was painting as you can't see the map in Weight Map view 
> mode on
> 2 pt polys.
>
> Then, just by twiddling spring, viscosity, and compress stretch 
> values, I was
> able to get the web to settle in a natural way.  I heard/read that a 
> Polygon
> Stretch value of Zero doesn't work and to use a small fractional value 
> to prevent
> the major drooping that clothfx gives you by default.  Next a little 
> substructure
> was added and a higher viscosity to tighten up the sag.  Obviously 
> webs don't
> weigh much, so the default mass value of 1.0 was cut by 100 or so.
>
> Again, all this is via experimentation, changing each param one at a time
> to see what it does, tweaking another, and so on until the visual 
> effect is
> achieved.
>
>
>
>
>
> Dean A. Scott, mfa
> ---------------------------------------
> School of Visual Art and Design
> southern adventist university
> ---------------------------------------
> http://www.southern.edu/~dascott
>
sounds like the way I work!

you can download the spider-web I did here:

http://users2.ev1.net/~theartist/online-storage/spider-web.zip

enjoy!

-- 
��: }	theartist
0
Reply theartist 6/3/2006 1:15:13 AM

Dean A. Scott wrote:

>
>> Now, if you actually wanted to physically model the web, complete 
>> with all those hundreds (or thousands, depending on the type of 
>> spider web), it seems silly to manually do this. 
>
>
> Not very hard.  Make a SPP (single point poly) and Lathe it by turning 
> it 7,200
> degrees, with 960 segments, and offset on Z 1 meter.  Apply Taper 
> Evenly to
> scrunch in the center, then Set Value to 0 on Z.  Bingo, instant 
> spiral of 2 pt
> polys.
>
> Next, and this is the more tedious manual stuff, select two points and 
> hit P
> to make a poly segment and continue out from the center to make 18 or 
> so randomly
> spaced spokes.  Use the Make Points tool to add points to extend 6 of 
> those spokes
> and then add a few random runners between these size main spokes for 
> the remaining
> spokes to attach to.
>
> Oh heck... here's my quick 5 minute spider web object and 1 hr tweak 
> job of
> applying ClothFX.  Had to make a radial gradient weight map to get more
> "spring" tension in toward the center (applied to the Compress Stretch 
> param).
>
> Scene and object is in the zip at:  http://chrusion.com/public_files

Well, I finally had the chance to check out your web, and it was done 
pretty cooly :)
And thanx too!
I wasted far too much time playing with winds though when I had other 
things to be doing.
See what you did! LOL

..............md :)

-- 
-- 
Check out my Tutorials:

MD arts
Mark Dunakin
md@md-arts.com
http://www.md-arts.com 
0
Reply Mark 6/7/2006 6:40:12 PM

17 Replies
48 Views

(page loaded in 0.134 seconds)


Reply: