Subject:
|
Re: LSynth and string
|
Newsgroups:
|
lugnet.cad
|
Date:
|
Tue, 26 Apr 2005 17:40:25 GMT
|
Viewed:
|
1488 times
|
| |
|
|
> I know I've seen black and red LEGO string, but I've not seen multi-colored
> string (I'm not very worldy that way sticking to robotics and technic).
>
> Larry produced a string segment that looks a little like one of those tiny
> stands they get elephants to step on. Its sides are composed of triangles. In
> an attempt to approximate the color difference due to diffuse and specular
> (shading and shinyness) he used two different colored triangles. This is the
> two-color issue being discussed.
>
> I think the right solution is to actually provide a tri-twist part, that can be
> made any color and used as a string segment. The 3D geometry of the tri twist
> part should handle the "two-color" issue Larry mentioned, and allow string to be
> any single color.
>
> Your proposal does sound interesting. Did you imagine three colors per strand
> in the tri-strand string?
I'm not too familiar with Lego string either but my proposal would be most
useful for situations where you have any number of colours, twisting slowly
around the string length (like a candy cane for example). I certainly don't
recall seeing Lego string like that but I've certainly seen string like that
before.
I'm guessing here, but your code will generate a vector function of some
variable (lets call it s) from either a spline code or something else. You then
use this function (and an angle function) to position and rotate each segment of
string, using eg. Larry's part.
If I'm guessing correctly, multiplying the rotation matrix by a further rotation
matrix dependent on the 'direction' of the string at s, multiplied by a
rotational factor (like 2n\pi s for n twists where s goes from 0 to 1) would
achieve what I'm suggesting. Effectively all you do is rotate the torsional
component of the string at every point. I did a little demonstration of the
method and it seems to be working (although not quite how I would expect, I
think there is a bug in my code).
Can I obtain Larry's string component anywhere? If not, would you be able to
email me the .dat?
Tim
PS. I have been looking at how to make pneumatic hose (and other stiff 'parts')
look a little more accurate. It would require some internal physical data in
addition to the strings real length (as a meta-command perhaps) but hopefully
the final work would be implementable as a simple function of the connetion
points and angles (works in Matlab so far but not in the nicest way).
|
|
Message has 1 Reply:
Message is in Reply To:
 | | Re: LSynth and string
|
| (...) I know I've seen black and red LEGO string, but I've not seen multi-colored string (I'm not very worldy that way sticking to robotics and technic). Larry produced a string segment that looks a little like one of those tiny stands they get (...) (20 years ago, 26-Apr-05, to lugnet.cad)
|
20 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
