Here's an early (VERY early) look at what might end up going into the
current IRTC contest. It's a giant (The arch is 13000ft, the canyon is
well over 2 miles wide, and over a mile deep) arch bridge. This is almost
a 'storyboard' quality image at this point. The arch is a proper catenary
cirve, so this should work from an engineering standpoint, provided strong
enough materials can be developed. There will be two flat roadbeds (or
mayube railbeds?) across the canyon, one on either side of the arch.
Anyone have an ideas on how to get the height field looking better? Better
texturing will certinaly help, and I'm using smooth already, but it's
just... kinda ugly. I know a 16 bit image would help tons. Anyone willing
to take a 8bit png and 16bitize it for me?

Well, anyways, this is after, oh, maybe 5 hours of work.

Anyway, the image is actually kinda big, so I'm posting it to my website:

http://tylereaves.com/bridge.jpg

Post a reply to this message

Probably better and faster to use an isosurface instead of height field. And
with just a little bit of wind this narrow bridge will start moving forth
and back and possibly break.

Post a reply to this message

On Fri, 28 Nov 2003 01:33:09 +0100, Apache wrote:

> Probably better and faster to use an isosurface instead of height field.

Any links or doc pointers as to how one would do this? (That don't require
 higher math...)

> And
> with just a little bit of wind this narrow bridge will start moving forth
> and back and possibly break.

I disagree, for two reasons.

It could be/is built of futuristic materials (The IRTC theme is 'Future')
like carbon nanotubes, etc, that have strength that is orders of magnitude
past anything used today.

Also, it's 200ft wide, with the 3 arch tubes being 40ft in diameter. I
think it could work. The New River Gorge Bridge in West Virginia is 4
lanes (Maybe 60ft wide?) and has a single arch span of just under 2000ft.
This bridges width/length ratio may be about 3 times that, but the West
Virginia bridge was done in the 1970's with regular steel, and not
something with many times the tensile strength. I really think it could be
done.

Post a reply to this message

On Thu, 27 Nov 2003 19:56:27 -0500, Tyler Eaves wrote:

> On Fri, 28 Nov 2003 01:33:09 +0100, Apache wrote:
> 
>> Probably better and faster to use an isosurface instead of height field.
> 
> Any links or doc pointers as to how one would do this? (That don't require
>  higher math...)
> 
>> And
>> with just a little bit of wind this narrow bridge will start moving forth
>> and back and possibly break.
> 
> I disagree, for two reasons.
> 
> It could be/is built of futuristic materials (The IRTC theme is 'Future')
> like carbon nanotubes, etc, that have strength that is orders of magnitude
> past anything used today.
> 
> Also, it's 200ft wide, with the 3 arch tubes being 40ft in diameter. I
> think it could work. The New River Gorge Bridge in West Virginia is 4
> lanes (Maybe 60ft wide?) and has a single arch span of just under 2000ft.
> This bridges width/length ratio may be about 3 times that, but the West
> Virginia bridge was done in the 1970's with regular steel, and not
> something with many times the tensile strength. I really think it could be
> done.

Here's a pic:

http://www.stolaf.edu/people/dahl/Climbing/one.jpg

Post a reply to this message

apa### [at] hotmailcom news:3fc697c7$1@news.povray.org

> Probably better and faster to use an isosurface instead of height field. 

Realy? Isosurfaces are in most cases the most slow rendring objects (maybe 
after parametric) and heightfields/meshes are the fastest.

-- 
#macro g(U,V)(.4*abs(sin(9*sqrt(pow(x-U,2)+pow(y-V,2))))*pow(1-min(1,(sqrt(
pow(x-U,2)+pow(y-V,2))*.3)),2)+.9)#end#macro p(c)#if(c>1)#local l=mod(c,100
);g(2*div(l,10)-8,2*mod(l,10)-8)*p(div(c,100))#else 1#end#end light_source{
y 2}sphere{z*20 9pigment{function{p(26252423)*p(36455644)*p(66656463)}}}//M

Post a reply to this message

Here's another shot, from a different angle.

Just to give you an idea of the scale here, the flat portion of the two
decks is 90ft. The seemingly hair-thin suspension wires are actually 2ft
in diameter. The bridge would be a very good contender for the tallest
structure anywhere. The arch, at it's highest point, is 1060ft above road
level, 1145ft above the bottom of the roadway supports, and 8760ft above
the river. 

Anyways, heres the new shot. 

http://www.tylereaves.com/bridge.png 

I didn't compress it this time, so it's a hair over 250k.

Post a reply to this message

> Anyways, heres the new shot.
>
> http://www.tylereaves.com/bridge.png
>
> I didn't compress it this time, so it's a hair over 250k.

Reminds me of a project I did many moons ago, with Vivid:   Ringworld.
Now THAT's a bridge too far.

This is not my project, but very similar.
http://www.rahul.net/rootbear/graphics/ringworld/rw0.html

Post a reply to this message

On Thu, 27 Nov 2003 22:31:12 -0500, Dave VanHorn wrote:

> 
> 
>> Anyways, heres the new shot.
>>
>> http://www.tylereaves.com/bridge.png
>>
>> I didn't compress it this time, so it's a hair over 250k.
> 
> Reminds me of a project I did many moons ago, with Vivid:   Ringworld.
> Now THAT's a bridge too far.
> 
> This is not my project, but very similar.
> http://www.rahul.net/rootbear/graphics/ringworld/rw0.html

Hehe, neat, is somewhat 1995-ish ;)

Actually, I read the book a while back, so that was neat to actually 'see'
it.

Post a reply to this message

> Actually, I read the book a while back, so that was neat to actually 'see'
it.

That's why I rendered it, back in '92.
I found out some of the things they described can't be seen as in the book,
other things worked out..

It's an interesting problem of scale.

Back then, I rendered a 360 frame "fly-through" which would have been at
something like warp-9, but it took a week on my 386-33

Post a reply to this message

On 27 Nov 2003 21:36:07 -0500, "Rafal 'Raf256' Maj" <spa### [at] raf256com> wrote:
> > Probably better and faster to use an isosurface instead of height field. 
>
> Realy? Isosurfaces are in most cases the most slow rendring objects (maybe 
> after parametric) and heightfields/meshes are the fastest.

It depends. Detailed meshes and heighfileds located in complicated scenes
takes large amount of memory while a few simple isosurfaces are very small.
Meny referrences to various places in large scenes cause swaping memory cache
a lot while small isosurface written by experienced user can be efficient. So
saying anything about majority of cases is wrong without considering specific
object/scene as it was already pointed out a few times.

ABX

Post a reply to this message