 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>
> I followed your advice and tried the Dome generater program. After about
>15 different tests/renders/rejections I have yet to find a shape that comes
>close to what I am hoping for. Perhaps the master could supply me with a
>couple of examples of what to enter in the different fields so that I may
>learn from his wisdome ?
The enclosed image (very quick and dirty, no AA, no good texture... just to
show you) is defined by
Base = EpiCycloide
Slope = Droite-Droite
Windows = Sans
To be honnest, this image has been done with *MY* version of the domenator
:-} (didn't I send it to you ?) where you can choose more parameters and so
the sort of cone given by the previous definition has it's head, (sorry, too
much bloody, I meant top) cut off (as you know, meshes are not good in CSG).
Here I used a v in the range 0.5 to 1 and then scaled by 2 in the up
direction.
>
> On a side note:
> One undesirable characteristic of the Dome program is that it is
producing
>a lot of degenerate triangles. You might think about adding a degenerate
>triangle testing/removal feature in the program to eliminate these. I had
>one file with over 350 degenerate triangles and that is a lot to have to
edit
>out of the file by hand. Especialy in files that are over 1+ megs in size.
>I am not sure how to implement such a routine but there are many here
>that I am sure could help you with that. It is a common problem for
triangle
>generating utilities and the fix I believe is an easy one to add.
>
Very easy : cross product is the answer (will be done as soon as I can)
Philippe
Post a reply to this message
Attachments:
Download 'LampShade.jpg' (8 KB)
Preview of image 'LampShade.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ph Gibone wrote:
>
> The enclosed image (very quick and dirty, no AA, no good texture... just to
> show you) is defined by
> Base = EpiCycloide
> Slope = Droite-Droite
> Windows = Sans
>
> To be honnest, this image has been done with *MY* version of the domenator
> :-} (didn't I send it to you ?) where you can choose more parameters and so
> the sort of cone given by the previous definition has it's head, (sorry, too
> much bloody, I meant top) cut off (as you know, meshes are not good in CSG).
> Here I used a v in the range 0.5 to 1 and then scaled by 2 in the up
> direction.
Even this is not exactly what I am looking for. The pattern in the image
you attached is only using a half wave from what I can see and I want
a full sine wave pattern.
Yours looks like this:
__ __
/ \ / \
/ \ / \
/ \ / \
/ \/ \
---------------------
and I am trying to get this:
__ __
/ \ / \
/ \ / \
---/----- \------/------\----
/ \ / \
__/ \__/ \__
Picture to yourself both of the above samples with nice smooth curves that
striaght lines simply do not do justice to.
I have hacked out a method that uses an open cone object and then clip
it in half. I use two of these one the inverse of the other. Then it's a
simple (ha ha) translate and rotate routine with a while loop to completes
the shape. There are a few hidden challenges doing it this way but the
results are looking acceptable once the parameters are setup right.
>> It is a common problem for triangle generating utilities and the
>> fix I believe is an easy one to add.
> Very easy : cross product is the answer (will be done as soon as I can)
Glad to here that it is something you can fix with reletive ease.
Thanks for your help and your continued support.
--
Ken Tyler
mailto://tylereng@pacbell.net
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Even this is not exactly what I am looking for. The pattern in the image
>you attached is only using a half wave from what I can see and I want
>a full sine wave pattern.
>
>Yours looks like this:
> __ __
> / \ / \
> / \ / \
> / \ / \
>/ \/ \
>---------------------
>
>
>and I am trying to get this:
> __ __
> / \ / \
> / \ / \
> ---/----- \------/------\----
> / \ / \
>__/ \__/ \__
>
>
>Picture to yourself both of the above samples with nice smooth curves that
>striaght lines simply do not do justice to.
Change the "epicycloid" base to "napperon" and you'll get it ! If I can find
a few minutes today I'll post a small image
> I have hacked out a method that uses an open cone object and then clip
>it in half. I use two of these one the inverse of the other. Then it's a
>simple (ha ha) translate and rotate routine with a while loop to completes
>the shape. There are a few hidden challenges doing it this way but the
>results are looking acceptable once the parameters are setup right.
>
You can't miss !
> Glad to here that it is something you can fix with reletive ease.
It is done : i'm testing it before releasing
>Thanks for your help and your continued support.
Uncle Ken thankful to me ? My 3d life's fullfilled :-} (in fact not 100%
joke, I really have great respect and admiration for your knowledge and for
the way you share it, thanks Mr Ken)
Philippe
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I took the time (in fact a few minutes), here the base is "Napperon"
Philippe
Post a reply to this message
Attachments:
Download 'LampShade2.jpg' (7 KB)
Preview of image 'LampShade2.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ph Gibone wrote:
>
> I took the time (in fact a few minutes), here the base is "Napperon"
>
> Philippe
Now your cookin". That is indeed much closer to that which I have been
endeavoring to produce.
Thank you sir.
--
Ken Tyler
mailto://tylereng@pacbell.net
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ph Gibone wrote:
> Change the "epicycloid" base to "napperon" and you'll get it !
I will accept this challenge.
> > Glad to here that it is something you can fix with reletive ease.
> It is done : i'm testing it before releasing
Superb !
> >Thanks for your help and your continued support.
>
> Uncle Ken thankful to me ? My 3d life's fullfilled :-} (in fact not 100%
> joke, I really have great respect and admiration for your knowledge and for
> the way you share it, thanks Mr Ken)
>
> Philippe
Ah gee. (blush).
--
Ken Tyler
mailto://tylereng@pacbell.net
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Wed, 05 May 1999 17:32:16 -0700, Ken <tyl### [at] pacbell net> wrote:
> Even this is not exactly what I am looking for. The pattern in the image
>you attached is only using a half wave from what I can see and I want
>a full sine wave pattern.
>
>Yours looks like this:
[pictures clipped]
>Picture to yourself both of the above samples with nice smooth curves that
>striaght lines simply do not do justice to.
Ken, was there something in my include file that didn't work for you? It
should have generated something very like what you're looking for, with
the added benefit of double illumination. If you can't figure it out, just
tell me the dimensions you're looking for and I'll make the necessary
modifications for you.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ron Parker wrote:
> Ken, was there something in my include file that didn't work for you? It
> should have generated something very like what you're looking for, with
> the added benefit of double illumination. If you can't figure it out, just
> tell me the dimensions you're looking for and I'll make the necessary
> modifications for you.
Ron,
Sorry it took so long for me to get to your suggestion but I aviod
patch object unless I can control them in a modelling environment.
So any way I downloaded your source last night and after a bit of work
isolated the ridged part of the cap. After a few modifications and a
total lack of comprehension of what exacly your source is doing I don't
believe that it will work as structured and my lack of ability to control
it's overall shape is limiting. The angle or taper of the shape is not
steep enough and I am not particularly happy with the straight skirt or
it's length at the top (or bottom depending on orientation) of the shape.
It is a very nicely modelled bottle cap however and I have had enough
suggestions and a few successful attempts of my own that I just don't
need another lamp shade at the point in time. I have around 10 different
designs for one lamp and I think that is enough for one lamp.
Thank you for the offer regardless.
--
Ken Tyler
mailto://tylereng@pacbell.net
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
