What to ask for more?
Can we get Rhino plugin demo early next year?
This would speed things up.
With the bonus feature “Rhino fillets solved” included.
(From the “how to make a best-seller plugin for Rhino” department)
At least Volkswagen does not use Catia for class A. There were tests in using Catia in design in the late 90’s, but it failed due to surface quality issues. I think BMW did that once for one model. But the effort was way too expensive.
If you are talking about the current ICEM Surf integration in Catia, this is something else, because you still need to corner fillet manually to achieve class A. This module is actually the same as the standalone platform. Dassault just likes to combine both platforms.
Basically one aspect of class A is not using multispan. Although I would not strictly follow this rule, it is still important for most cases. One reason for this: the overall curvature flow drastically gets worse if you add knots. Why? Because even if you have G3 and higher in between each Bezier segment, you will have a hard time to control it. You can apply an energy minimisation, but that doesn’t necessary solve your problem, because you cannot control and modify it. You have to live with the outcome.
It is also hard to fully match a patch like this. If you numerically test it you will see greater deviation. It’s not bad for most cases, but it is bad for class A demands. You can argue about numerical deviation a lot and if the values makes sense or not. But if I create good class B fillets, the thinking is not, hey you did that much faster so I don’t care about having this quality. They rather say, you did a bad job because I want it faster but with same quality. Please don’t see this as a critic to your plugin. Its great, but its not class A.
I think we can discuss for a long time, but there will always be, for this kind of situation, geometric / mathematical limitations. The plug-in does as much as possible, but it’s not a magic tool!
Any CAD software will never be able to do everything perfectly. For example, VSR multiblend was an attempt (better than the Rhino patch surely) but not perfect.
Catia, Ugs nx (to talk about the best) they also have limitations, otherwise they would have developed a perfect tool long ago.
The proposed XNurbs plug-in looks like it could be quite useful. However I strongly doubt it will be as fantastic as many of the claims @XNurbs has made, and which various folks have been skeptical of. My personal reaction to strong sales hype is to become somewhat negative about the product on offer. But depending on the price and implementation I might be interested in purchasing the XNurbs plug-in for Rhino. I currently use Patch to model boat hulls based on data from photogrammetry and other sources.
Just a reminder: “Class A” is only a definition of surfaces that are of good enough quality to be used as surfaces that are visible to the user. And a high continuity is basically only important to make sure reflections run smoothly over a highly glossy polysurface. So “Class A” is not a mathematical definition per say and not everybody are designing for Audi’s A8 team or for Koenigsegg. So for those who are not the nerdy-nerdy-nerdest of nerds then a lot of less than 100% nerd-proof G3 continuity-surfaces are classified as Class A.
Recap: One designers Class A requirements can differ from another, and has to be considered in this discussion.
It’d help if everyone ups an example image of what they mean by Class A with their comments.
I was thinking curvature continuous would be Class A before, but it seems to mean “Surface looks good enough for me”
Yup. That’s just a fact.
As for me I would like to emphasize two things in this discussion:
I really do apprechiate all I can learn from the Surface Masters post’s in this forum. I’m afraid I wouldn’t know all the names, but TomTom’s and Stratosfear’s informative posts included. I just learn from almost every post from you guys.
Problem is: I do surfaces, and I want them to be perfect (perfect for my own needs), but I don’t do advanced surfaces often enough to ever get as skilled as I would like, and so I can only dream of reaching the level of skills which some of you Surface Masters on this forum are capable of.
And this 1 & 2 leaves me (and I assume I’m among the majority of Rhino users in this respect) with a need for a tool - as a compensation for my lacking skills - to can fill a gap between my low level and up to at least the ankles of Surface Masters like TomTom and Stratosfear and others of that caliber.
Conclusion: XNurbs is interesting for a majority of Rhino users. I mean, potential customers.
Moreover, the Surface Masters should have credit for sharing their skills, and in so doing making us mortals aware about the skills we still lack (we really benefit from their advice, although slowly in my case), and it follows, how much we mortals, the majority of Rhino users, actually need XNurbs to fill that very gap.
Well, that’s the core of the problem as both are right. In some scenarios G0 with tangency is good enough while other scenarios requires the best possible continuity that any tool can manage to generate.
I’m not saying this or that is wrong, I just think it is important that people understanding why some users can cheer while others moan over exactly the same results while using the same terminology…
that both kills and strengthen my point at the same time…
Squish to get that double curved surface to the wall?
This part shown is per automotive definition not a part of class A. This part isn’t visible, so its considered as class B in automotive. You won’t see these patchlayouts on exterior or interior design data.
“class A” is a term originated in automotive. Sure there isn’t a fixed public definition, and people doing class A surfacing are rare minority. Again I didn’t say something negative about this plugin, I’m just saying its not class A (…in a automotive way). So even if Wikipedia does not provide an full information about whats class A and whats not, at least companies in automotive do have certain quality standards. Often too strict, but that’s how it is. I don’t say its right or wrong, just saying how it is.