sweep 1 rail problem

Replies to This Discussion

Permalink Reply by Minh on January 11, 2010 at 6:25am

At a first look, it seems that the section should be exploded and then be joined again, but your datas are really a mess. I will have a look later this evvening on it.

Permalink Reply by Thomas on January 11, 2010 at 8:36am

Minh c'est toi? Enfin je pense que c'est toi. Si c'est le cas merci c'est cool! Oui je sais que c'est le bordel mais disons que comme je suis pas un pro j'ai tendance à faire les choses compliquées au départ et à les simplifiées. Enfin si t'arrive à y voir clair tant mieux....

Permalink Reply by Minh on January 11, 2010 at 8:40am

C'est hallucinant que l'on se retrouve ici, (d'ailleurs je ne t'avais pas reconnu sur cette photo) c'est le projet de piscine dont Julie m'a parlé? Je regarde ça ce soir, ça n'a pas l'air compliqué mais c'est juste un peu le fouillis

Permalink Reply by Thomas on January 11, 2010 at 8:45am

c'est une vieille photo, j'ai de plus en plus de mal a me voir tel que je suis :(
Mais oui c'est bien ça! En fait c'est exactement ça, pas compliqué mais fouilli...
Toute aide sera la bienvenue l'ami!
Merci beaucoup en tout cas...

Permalink Reply by Yann on January 11, 2010 at 9:02am

Hi thomas.

it's effectively "un peu le fouillis" and really heavy! if you look at the way your curves are organised just before you merge them, you will see that they are not in the same path. simplify path before merging to have a correct list for each group of curves. (use "special->panel" and link it to your output (the four trim components).
another problem seems to be due to the different number of rails and sections on your sweep component (because they have not been joined, but also because you didn't include angular edge of your main brep as rails)
Je n'ai pas le temps de regarder plus longtemps ta définition, mais selon moi, c'est sur ces deux points qu'il y a un soucis! Bonne chance

Permalink Reply by Thomas on January 11, 2010 at 9:40am

I might as well switch back to english too for the sake of finding as many helpful people as possible.
While writing the definition, I have thought of other methods, more simple and probably cleaner to generate the path curves, using sections on specigic surfaces instead of the projection/trim method.
It's probably going to be more efficient.
I'll keep the discussion updated whenever I get further (I actually don't have that much time to spend on that). Yann, I did find the way to concentrate a complex list of result in a single path but i can't remember how... Any clue on the button, or method to use?

Permalink Reply by Minh on January 11, 2010 at 9:50am

t is 'Flatten' under 'Logic' tab and 'tree' tab, therefore you could try to recreate branches through Graft button.

Permalink Reply by Yann on January 11, 2010 at 10:12am

flatten will crush your structuring. The risk is to disorganize your list. maybe "simplify" can be more recommended to make your list consistent and to keep order of your curves in anticipation of your sweep. verify you don't have null object in your tree. "prune" (component) it if necessary

Permalink Reply by Thomas on January 11, 2010 at 10:44am

I still can't get it to work, but at least I think this def is a bit clearer...

Attachments:

• APS-to3.3dm, 217 KB
• APS-to3.ghx, 129 KB

Permalink Reply by Thomas on January 11, 2010 at 10:56am

I still can't get it to work. I have used to sort item command to use only one object of both list, it still doesn't work. I have flattened the list and it doesn't work.
Once I bake both object and use the regular command sweep in rhino, it does work perfectly.
I post again the definition.
If anyone knows what's happening...

Attachments:

• APS-to3.3dm, 218 KB
• APS-to3.ghx, 166 KB

Permalink Reply by Minh on January 11, 2010 at 11:13am

Thomas,

It is better to consider intersection with frame rather than projecting lines on surface

Here a first improvement, one thing you haven't consider is to check the direction of your polycurve

an other improvement is to use graft instead of flatten, but certainly the result could be the same

Next step is to avoid small curve that is actually disturbing the sweep

(je continue plus tard, je te tiens au courant)

Attachments:

• APSto3_mm.ghx, 155 KB

Permalink Reply by maivan on January 12, 2010 at 2:48am

Hi Minh!

Thanks a lot, I am working together with Thomas, and with your "graft" help and upgrading our grasshopper we have arrived to a solution! we will show you the final result!


Thanks again!
_ivaN

• ‹ Previous
• 1
• 2
• Next ›
• Page