EIM - UberNurbs

#7

I would vote for as much effort being put into UberMesh, and making it as strong and capable as can be. I would rather that then having resources split between multiple approaches.

_mike

#8

Mmm. What I valued most from ÜberNURBS was the ability to later knife and boolean them without the problems poly-based SDS modeling produces
but then it is not as if knifing such complex NURBS objects was foolproof at all.

Argh, I don’t know
:frowning:

#9

I agree, and I don’t want to imply that the UberNURBS tools were not wanted or needed. One of the things that I have always liked about EI in general was that although they may not have had all of the newest features/trends, what they did have was implimented very well.

I also don;’ want to imply that Kishore would not put forth his full effort to one or the other, but the reality is that developing and maintaining both would take more time, and by defintiion something esle would suffer.

Just my .02. The fact that Kishore is posting indicates that he’s looking for input, which is awesome in and of itself.

_mike

#10

I believe this is a “killer feature” for EIM.
Typical SDS modelling apps only output polys, which do not play nice with professional CAD, especially for Engineering/ Industrial Design.

The ability to have an ACIS based SDS modeling system will be extremely useful for any designer who needs CAD interoperability, and an SDS toolset.

This is a truly unique feature of EIM, which would have special appeal to many users.

My 2 cents.
Dave :wink:

#11

UberNurbs and it’s conversion to ACIS along with the knives is possibly the most important feature EIM/Telsa has.

Many users like myself switched to Concepts Unlimited (‘Shark’) and later ViaCad or something similar after development of EIM stopped. Whilst EIM is/always has been far more elegant it will need something that sets it apart to pull many users back.

I would like to see at some point the tools that are not dependent on ACIS implemented in Animator and for Telsa to be a more substantial ACIS modeller in it’s own right.

BB

#12

I’d hate to see UberNurbs be dropped.
But one option would be adding it to the next version of Tesla.

JM

#13

Check out T-Splines…available for Rhino:

http://www.tsplines.com/rhino/

Shows how far in advance EIM was when it came out in some areas. The biggest problem with Ubernurbs translation to ACIS nurbs is the overwhelming amount of complexity your object now has. Makes it almost impossible to do anything with.

I think if the surface, upon conversion, was re-skinned into one continuous nurbs surface then it would be very useful…as it stands it’s a cool feature on paper but not very useable in day to day work.

#14

Any chance that there are plans for editable “history”?

Concepts/ Shark tries, and sometime is good at this. Solidworks is supposed to be great.

i.e. boolean a cylinder out of a flat plane. Move the original cylinder, and the hole will move too.

All parametric rounding operations, deformations, etc. can have their parameters edited at any time. “Features” can be removed from the modeling history (deleting a round for instance) or new features inserted.

A very powerful tool for tweaking forms, or creating variations on complex models.

Not sure if the latest ACIS helps with this, but I hope so! :wink:
Dave

#15

Wow Paul. That’s a great link for this topic.
T-Splines provides a glimpse of what UberNurbs can be.

Wasn’t Modeler originally conceived as a modeling solution
allowing seamless interoperability between Nurbs, solid modeling,
SDS, beziers, etc.? It’s a highly appealing concept that I hope EI
continues striving for. UberNurbs, done right, could sure add to the appeal.

JM

#16

I dont know whether my answer is dealing with the Uber-Nurbs topic so much, (sorry if not) but from my point of view it could be important, not only to think about modeling features inside EIM/Tesla. More and more we get ready modeled CAD-files from apps like UG, Catia, ProE, Rhino and so on. As discussed in a different thread we now still have to route between several apps and converters to make these files become usable FACT-files in EIAS. Much too much stations...thats a big waste of time and ressources in my eyes.
As the main focus of EIAS seems to be hard body animation (in the moment),
I guess many users would highly appreciate something like step-import in EIM/Tesla
and a more direct connection to EIAS.
I dont know how complex this would be and if it is possible at all, its just my point of view.

Btw…I first didn`like the idea of renaming EIM too much, but “TESLA” is a very good choice.
Big name…great inventor…and hopefully a fantastic new modeler too :slight_smile:

Cheers
Tom

#17

yes, it is clear that many users are using eim as a FACT files translator. I am assuming SAT offers compatibility to many CAD apps. At this point STEP import is not planned. I think most CAD apps should export SAT format by now. If more request for STEP comes in the future to justify more licensing costs we can reconsider it.

#18

All good ideas. Some are going to be in next Tesla. Some in the future versions. Foundational work is being done.

We need to be able to balance new features with good release schedule.

#19

Hi Kishore,

I almost slept on this one… Main tool that attracted me to EIM was UberNurbs. It was at the time easier to control than most other 3D apps. My goal back then was to create an organic model to animate within EI (don’t laugh gang EI was on the road and hit the side walk for a second)

If I had to pick improvements on Uber Nurbs definite history. Multiple undo was covered, easier face selection a-la modo / silo. Deforms within Ubertools ie, select points/faces and invoke a deform window that can modify the selected area. A way to control edge meshes via triangles/smoothing at export. Faster response times, color coding of various areas to eliminate confusion. Controlled weight map shading for CAD / animation.

A feature that always seemed to frustrate my modeling process was that the model would have to become an ubermesh prior to doing a boolean. After using the boolean re-editing the cage resulted in crashes. For instance I attempted several times modeling character heads and to create nose, eye sockets and ears and mouth I would prefer to boolean those out than to navigate extrusion, expansion repeat process until I could form a nose.

So maybe adding a uber-boolean tool that can cut, or knife the ubernurbs away would help. the shortcuts that were build in helped model but I feel like there was room for improvement. If the UberNurbs was accessed I feel like it became a stand alone application within EIM. Each of the tools had to be clicked to select then return to the model cage edit return to the tools and it became tedious. A shortcut for each ubernurb tool was definitely in order. I would have saved plenty of editing time if all i have to do is click and edit.

Wow I just logged on so let me take a quick dive into modeler and I may have more feedback. Thanks for the outlook and Tesla.

#20

I found UberNurbs incredibly useful. The ability to convert to ACIS nurbs was amazingly useful. As arketype mentioned above, I consider it to be EIM’s killer feature.

Why? Several reasons:

  1. Sub-D is very good for organic, while nurbs are very good for hard-body. The ability to interconvert between the two is unique and surprisingly powerful. For example, I can model a complex semi-organic form in sub-d, and then afterwards knife, boolean, and chamfer the results. This is great for organic-mechanical models. And also great for artists and concept designers.

This is exactly the work-flow I used in creating a semi-accurate model of a violin in EIAS. Most complex parts (the violin scroll and the front/back surfaces) were modeled in sub-d, but then booleaned and chamfered extensively. It would be rather painful to cut out the holes in the top surface in a purely sub-d modeler; one would have to hand tweak the local topology extensively (ouch!) before chamfering. But as a nurbs model, it’s a piece of cake!
http://www.cs.duke.edu/~yan/page3/page4/page4.html

  1. Currently, EIM doesn’t have a huge feature advantage any where else (except a very clean user interface). By “feature advantage” I mean a feature on paper. (On the other hand, quality advantage is another story. I think EIM is great!).

EIM is “missing” the following kinds of features: topology brush, brush sculpting, texturing, paint, UV mapping, advanced poly manipulation for sub-d (see any advanced Wings3d or Silo3d video), soft-selection, combinatorial tool-set (such as Modo), dynamic construction history (see moi3d and any modern CAD program, Maya, etc.).

  1. Interestingly, Amapi3d Designer Pro has the ability to convert from Sub-D into nurbs. However, it’s not clear if e-frontiere is going to develop Amapi anymore, since they let go of their French development team. I was very sad about this, as this feature and future development is exactly why I purcahsed Amapi.

  2. In general, 3d software is becoming the center of a larger and larger universe of software, and I think the ability to interact with CAD files will become increasingly important. Just from the EIAS gallery, one can see many examples where EIAS is used for product development, commericials, and architecture. The ability to make an organic form, and then convert to ACIS nurbs will be, I think, useful and important.

My final request: Please keep UberNurbs even if it does not get actively developed in the future. Of course I would love future development, but short of that, please preserve this feature!

Sincerely,
–ChiralSym

#21

I guess the conversion feature for UberNurbs is justification enough to preserve it, at least until a substitute exists in the UberMesh world for conversion to ACIS

Thanks to everyone for feedback.

#22

Excellent news!
JM

#23

Whew!

:slight_smile:

#24

Kishore, thank you for opening this question up to the user community. My take on yuor question was obviously not in line with the majority of users, and it’s very refereshing when you have a developer actually seeking user input!

Looks like EIAS users win again. Thanks!

_mike

#25

Obviously Ubermesh should have developement priority over Ubernurbs, once Tesla is released it will be immediately compared to the current crop of SDS modelers by a lot of people.

To be honest, i’ve not used the Ubermesh/Ubernurbs tools in a long time, but it would be great to see Tesla become the all-in-one modeling solution it was always meant meant to be, good luck with it.

Reuben

#26

kishorem,

Thanks!! And best of luck in all your development efforts!

Sincerely,
–ChiralSym