Some surfaces behaviour makes me confused

[[-Skint-]]

Jacek, its a good start but your model is so much more simple than mine

, have you looked at my model ? I can save it in a different format if you cannot use wf3 files ?


I know that a BB has to have the same amount of entities. i.e a square blending into the circle, each has to have same amount of parts etc. If you have a look at my model, could you explain how I remove the strange undercut as mentioned in ( 2 ).


This thread should be so much longer than the Piotrowski`s Render thread. A good render starts with a good model right

MCAD needs a " Radge " member. I have been looking through his SW model challenges this morning, and I wish we had a Pro-e WF equivalent memebr on this site haha.


Thanks for having a go straight away.

 

[Jacoolo]

no I haven`t seen Your model while I am WF 2,0 user

however the undercut might - and for sure does - come from BB tangency condition

if there would be a curve - spline the best(no curvature steps along) - with continues curvature, then You can achieve smoth transition, even without Curvature conditions in BB.

I did not make it while I wanted to make it fast.

If You set the spline right - jeff posts about Bezier curves and its equations - You can have smotth curvature transition between all shapes along trajectory.

according to miss of "ragde" kind of forum member - I miss such guy too...

but, I think if You dig some of jeff`s models, You gonna learn the same

 

[james.lynch]

I think there is some REALLY benificial hints a tips in this thread- some of the most inciteful I've seen on this forum wrtsurfacing and curve networks- PLEASE do not allow it to become cluttered with all sorts fo things and make it impossible to find anything useful..


just my $0.02c..


James

 

[[-Skint-]]

Damnit... cant you update to WF3.0

I will keep having a go at it. The difficulty is the messy junction between the base and the BB, there is just too much in between. I may have to start again and approach from a different way.


Sometimes Pro-e tells me what I can design...cos I cant tell it. Very frustrating.

 

[[-Skint-]]

less formal is better. Just make a challenge and we post it on several boards. Watch the challenge in different software packages. I got dicksham on another (he is on the Catia board here but there is no traffic) to learn something about Catia.

Bart, I agree. The way it is handled over at the PDF forum is perfect.

 

[Jacoolo]

skint, put the pic from your BB set - I want take a look on the curve network and transition conditions

According to challange suggestion - I gonna prepare such post today ivening at PDF

 

[[-Skint-]]

I think there is some REALLY benificial hints a tips in this thread- some of the most inciteful I've seen on this forum wrtsurfacing and curve networks- PLEASE do not allow it to become cluttered with all sorts fo things and make it impossible to find anything useful..


just my $0.02c..


James

James, quite right ! I have moved it to a fresh thread and edited my original post. Now go and view it and sort my problem out

 

[james.lynch]

]

James, quite right ! I have moved it to a fresh thread and edited my original post. Now go and view it and sort my problem out

- I'll see if I can take a look a little later on - busy with deadlines as always!

 

[Jacoolo]

I am very curoius if any of Logitech mice was created from start to the end in Pro\e without sculpting and scaning it before?

 

[[-Skint-]]

I am very curoius if any of Logitech mice was created from start to the end in Pro\e without sculpting and scaning it before?

Hey Jacek, I think James ( see above - www.designpartners.com ) is a master when it comes to modelling consumer electronics, and has done many logitec items I believe.

 

[Jacoolo]

without a doubt.

I just want to have clear statement - "if" any of those products are made in the way we are doing? So by curves, VSS, BB. Or maybe- as james mentioned once - they scuplts in first and then scan to Pro\e to clear the data with data doctor or different exchange tool(reverse engineering)

 

[james.lynch]

Yes I have done quite a few Logitech units - generally all are hand sculpted for ergonomic reasons and digitally scanned producing an STL. Curve networks and Surfaces are then wrapped around this scan generally to within 0.1 - 0.2mm of the scan (accountig for minor imperfections within the hand model)


the curvenetwork and surfaces are 100% independant of the scan data ensuring no references to imported data allowing modifications to becompleted for what ever reason.


James

 

[Jacoolo]

that was what I`ve wanted

now james, taking advantage I have You here - tell me, are those models done in ISDX or within ordinary Pro\e surfacing capabilities?

What do You use often - spline ors conic arcs?

 

[james.lynch]

more often than not ISDX is not used - I pretty much use ISDX only for curve building and even then it is generally very well constrained to a sketched or intersect curve. As for splines or conics - both have their place and both are very useful! I've found Jeff's contribution to curves very informative when it comes to splines.


James

 

[Jacoolo]

thx james for the great input, I wish have you more on this post

 

[james.lynch]

I read it all but havent got much time to post

 

[jeff4136]

For my talk at the 2008 international conference I am going to explain what it
takes to be a software tool at GM and hopefully some of the PTC programmers take
head so ...

Before you get too fired up about influencing PTC programmers* you might take
a quick read thru ...


[url]http://www.rzg.mpg.de/~sa2728/CADDS5-Doku/NURBS-v12-englisch .pdf[/url]


I just stumbled over it and it's an informative read for anyone interested in the type of 'low level' NURBS entity control that Rhino, for instance, exposes to the user (and more), and touches on some of the concepts (and misconceptions) being discussed here and other recent surfacing related threads.


*It's stockholders and upper management you need to influence anyway, eh? ;^)

 

[design-engine]

Jeff, thats some dense information.

Some of that seams is geared to the programmer but no. Its geard to the user. WF 4 lets you see isoparms in the preview window BTW

Edited by: design-engine

 

[jeff4136]

> dense information ... geared to the user ...


Yeah, lot of options and controls. Some of it looks a little familiar?


- - - - - - - -


I wanted to dink around with the helmet shape a bit with a VSS created
Bezier(?) surface (degree 3 b-spline surf with four CVs, U & V), explore
the range of attainable shape variations, etc. Also wanted to investigate
the behavior of a singular edge in the circumstances. Here it be if
anyone wants to play around with it along with another (4 x 13) version
that can hardly be called 'heavy'.
2008-03-14_224651_we_be_lite--wf2--.zip

 

[design-engine]

I am going to dink around with those trajpar variables more. I will import those into Alias tonight and see how the VSS looks. I am teaching a surfacing class this week and will try to incorporate those in the class. That is for sure a 3 degree surface with no isoparms.

A conic by the way is a 2 degree curve that makes for a 2 degree surface. Two cv's BTW.


Ok Howard... That model is cool. I would have never thought to get a VSS spline sketch and get the CV's to lock to those trajectories like that. have you done stuff like that on a production model? I just got schooled.




Edited by: design-engine