Early days in a possible new project.

Rough-and-ready radiosity, need to tweak the knot dimensions / depth, maybe
have a double-helix stair instead of just the one. Although my algorithm
should have taken care of it, I have a suspicion that the steps are
bunching up a bit on the curves.  It's probably just on the insides of the
curves though, so making the steps narrower might take care of that.

Obviously, there has to be bannisters and people walking up/down (maybe on
both sides of the stairs? Hmmm...) Not sure whether to go for metal frame
railings or heavy stone columns... I could try to put a rug down the centre
of the stairs, too!

Any thoughts, anyone?

Post a reply to this message

Attachments:
Download 'trefoil1.jpg' (183 KB)

Preview of image 'trefoil1.jpg'

Bill Pragnell wrote:

> Early days in a possible new project.
> 
> Rough-and-ready radiosity, need to tweak the knot dimensions / depth,
> maybe have a double-helix stair instead of just the one. Although my
> algorithm should have taken care of it, I have a suspicion that the steps
> are
> bunching up a bit on the curves.  It's probably just on the insides of the
> curves though, so making the steps narrower might take care of that.
> 
> Obviously, there has to be bannisters and people walking up/down (maybe on
> both sides of the stairs? Hmmm...) Not sure whether to go for metal frame
> railings or heavy stone columns... I could try to put a rug down the
> centre of the stairs, too!
> 
> Any thoughts, anyone?

The beginning of an Esher-like scene?
Beyond that, I do like the concept a lot.
If you go for heavy stone bannisters then the center column/knot should be
stone too, at least that's I think.
-- 
Ger

Post a reply to this message

No thoughts.  Just stairing and stairing.... :D


Actually think I do see what you mean... The stairs look a bit steeper in
the "straight-aways" than they do in the bends.  I'm assuming that's
because they're interpolated along a spline with shorter distances between
control point_1, point_2 etc relative to the different val_1, val_2 etc?
I'm new to using splines and will most likely be using them a lot more, so
I'm curious about your approach.  Is your macro one that is supposed to
create a spline that doesn't have different distances or is your macro
meant to evenly interpolate an arbitrary spline?  Or am I off the mark
altogether and there is no spline? :D

PS. pun intended.

Charles

Post a reply to this message

Hmm, my earlier reply to this thread seems didn't seem to post properly.
Anyway....


No thoughts.  Just stairing and stairing.... :D


Actually think I do see what you mean... The stairs look a bit steeper in
the "straight-aways" than they do in the bends.  I'm assuming that's
because they're interpolated along a spline with shorter distances between
control point_1, point_2 etc relative to the different val_1, val_2 etc?
I'm new to using splines and will most likely be using them a lot more, so
I'm curious about your approach.  Is your macro one that is supposed to
create a spline that doesn't have different distances or is your macro
meant to evenly interpolate an arbitrary spline?  Or am I off the mark
altogether and there is no spline? :D


Charles

PS sorry about the bad pun

Post a reply to this message

Bill Pragnell wrote:

> Early days in a possible new project.
> 
> Rough-and-ready radiosity, need to tweak the knot dimensions / depth, maybe
> have a double-helix stair instead of just the one. Although my algorithm
> should have taken care of it, I have a suspicion that the steps are
> bunching up a bit on the curves.  It's probably just on the insides of the
> curves though, so making the steps narrower might take care of that.
> 
> Obviously, there has to be bannisters and people walking up/down (maybe on
> both sides of the stairs? Hmmm...) Not sure whether to go for metal frame
> railings or heavy stone columns... I could try to put a rug down the centre
> of the stairs, too!
> 
> Any thoughts, anyone?

This really calls for those roly-poly things that Escher dreamed up.

Regards,
John

Post a reply to this message

"Bill Pragnell" <bil### [at] hotmailcom> schreef in bericht
news:web.43a03d5963c7def4731f01d10@news.povray.org...
> Early days in a possible new project.
>
> Rough-and-ready radiosity, need to tweak the knot dimensions / depth,
maybe
> have a double-helix stair instead of just the one. Although my algorithm
> should have taken care of it, I have a suspicion that the steps are
> bunching up a bit on the curves.  It's probably just on the insides of the
> curves though, so making the steps narrower might take care of that.
>
> Obviously, there has to be bannisters and people walking up/down (maybe on
> both sides of the stairs? Hmmm...) Not sure whether to go for metal frame
> railings or heavy stone columns... I could try to put a rug down the
centre
> of the stairs, too!
>
> Any thoughts, anyone?
>
Great concept!

People: both sides, obviously (obviously?).

bannister: depends on the material of the trefoil itself, I would say, but I
would suggest something rather thin, not too massive.

rug: that would be very nice!

Thomas

Post a reply to this message

"Charles C" <nomail@nomail> wrote:
> I'm curious about your approach.  Is your macro one that is supposed to
> create a spline that doesn't have different distances or is your macro
> meant to evenly interpolate an arbitrary spline?  Or am I off the mark
> altogether and there is no spline? :D
Well, it's probably technically a spline, but I haven't adhered to any
particular set of formulae beyond the parametric knot function (the centre
column is a sphere_sweep). The knot function, although evenly-spaced in b
(0 > b > 2*pi), bunches component objects up on the curves in x,y,z just as
any trig function would (think point-plot of sin(x)). My macro tries to
evenly space them by inching along the path in very small increments of b
until the cumulative distance reaches a preset, whereupon I place a
component object and start again. It's probably a numerical line integral,
thinking about it... :)

I think the slight bunching effect still apparent is caused by the stairs
pointing inwards on the inside of the curves; the ends on the knot path are
evenly spaced, but the opposite ends are pushed together by the curve. This
effect can be minimised by shortening the width of the stair object, but I
may have to put up with it!

> PS. pun intended.
Consider it groaned! :)

Bill

Post a reply to this message

Ger <No.### [at] ThankYou> wrote:
> If you go for heavy stone bannisters then the center column/knot should be
> stone too, at least that's I think.
I agree! It'll look like a castle stairwell after a topological bomb attack.
:)

Bill

Post a reply to this message

John VanSickle <evi### [at] hotmailcom> wrote:
> This really calls for those roly-poly things that Escher dreamed up.
You refer, I take it, to (deep breath) Pedalternorotandomovens Centroculatus
Articulosus? Crawling or rolling?

Oooh, I'm getting visions of possibilities here... perhaps I could try to
insert some landings on the staircases, on the outside of the curves, with
catwalks extending radially... somewhere for the curl-ups to roll off to!
I'd have to hide the distance in fog or something. :o

....thinks...

I'll get back to you.

Bill

Post a reply to this message

Bill Pragnell wrote:
> John VanSickle <evi### [at] hotmailcom> wrote:
> 
>>This really calls for those roly-poly things that Escher dreamed up.
> 
> You refer, I take it, to (deep breath) Pedalternorotandomovens Centroculatus
> Articulosus? Crawling or rolling?

Well, rolling up stairs is a difficult trick, even for an Escher beastie.

> Oooh, I'm getting visions of possibilities here... perhaps I could try to
> insert some landings on the staircases, on the outside of the curves, with
> catwalks extending radially... somewhere for the curl-ups to roll off to!
> I'd have to hide the distance in fog or something. :o

If you can do motion blur, have the creatures falling off of the thing.

Regards,
John

Post a reply to this message