Limitations of C4D NURBS and Booleans

[flingster]

wow very informative. some very good info here from both of ya.

like to see "urban ape" come visit us sometime on cgtalk...

thanks for posting the link....really bools are rubbish in c4d (as far as i can tell...) and modelling is hard work....but the creativity of the individual means you can model anything with a bit of forethought....what i would really like to be able to do is use modelling tools within c4d without having to rely on so many different plugins....not that the plugins are bad just that integration is better.

[AdamT]

That was a good thread. I've been using the EE Pro method as a boolean substitute for some time. In my experience it's the best way to get clean, beveled holes in Cinema.

[urban ape]

Thanks guys.:beer: I will get a really big head if people keep saying nice things about me.

The reason I have come here to post this is that I have started a new thead on dvgarage about booleans that I thought you might like to look at. I have decided to stick with it on dvgarage because that is where I started the nurbs thread but I will post something on here soon.

Just as a taster here is a boolean that I have just posted in my new thread. See how clean it is, can you guess how I did it. Oh yeah by the way it was done in the OSX version of XL6 so no fancy new technology was used.

Thanks again

Leslie.

[flingster]

urban ape: welcome to cgtalk...glad you could finally make it...tap...tap....
just kidding....unfortunately a few days ago the cgtalk servers had to regress back a couple of days from backups...and there was some very interesting and informative stuff on here about said subject. unfortunately these have been wiped out...however as a result maxon have had a number of feature requests for improvements....class act....users can influence software development....so hopefully fingers crossed....maxon take some of the discussion on board.

as i said....welcome...hope you call back and post...you will find people on here always very helpful and contibutors range from absolute novice (of which i am one) to the very skilled.

[urban ape]

Flingster

Glad to be here.

WE HAVE TO START A CAMPAIGN TO GET MAXON TO GIVE US BOOLEANS THAT WORK THE WAY WE WANT THEM TO. I have just posted on dvgarage and suggested that everyone who reads my BIG BAD BOOLEANS post contacts Maxon and asks/demands better and more efficient boolean tools. Also while they are at it they should also ask for the ability to use splines to knive objects.I hope eveyone on this thread will also do the same.

Leslie

[flingster]

yup this was what happened on here....people posted requests for better bools to maxon....unfortunately i can't exactly prove it...as well the posts on the server have been set back.

hopefully enough people took up the challenge....also it wasn't just bools improvements cgtalkers wanted but a load of other stuff...i forget...(someone?).

i like the idea about splines to knife objects...there maybe a plugin for that? dunno.

well how'd ya do it then?

[urban ape]

Flingster

Well sometime ago I notice that all the stuff I read on boolean problems in C4D only dealt with cylinders, spheres and other curved surfaces so I developed this little hack. This was done in the OSX version of XL6 so no fancy new technology was used.

I have posted this image to demonstrate this. As you can see it is created from a cube primitive that has been booleaned with a cylinder, no problems. The cube then has fillet turned on so that it emulates a cylinder. Then the unwanted polygons are deleted.

As to contacting Maxon lets set it up so everytime someone contacts Maxon and tells them we want better booleans and the ability to use splines to knife objects they post on this thread just with the simple message "BEEN THERE DONE THAT". That way we will have some idea of how many have done it.

Go here and tell them http://www.maxon.net/index_e.html

Leslie

[astrofish]

BEEN THERE DONE THAT

In case people aren't aware, the way to officially submit a request to Maxon is by clicking on 'Support', on their main page, and then on 'Suggestions'

[urban ape]

OK guys I have been to Maxon and I HAVE BEEN THERE AND DONE THAT now its your turn. Demand working booleans and the ability to use splines to knife objects.

Leslie

[AdamT]

BEEN THERE DONE THAT!

[flingster]

BEEN THERE DONE THAT last time.

ps. i think H. Ikeda also did...but i'm sure he'll let us know.

More power to cgtalkers....its good to talk.

by the sounds of it maxon develop according to the number of requests they get....personally anything they can do help modelling in anyway would be a bonus....i would also appreciate it if those who posted before suggestions about modelling toolset changes they would like to see done can do so again...because some excellent suggestions were made which probably warrant further discussion.

being not that experienced in 3d stuff....users of other products often say are this is better in whatever program and i would like to see in cinema. what i would like to know is in say xsi, lightwave or rhino what modelling tools are cool that cinema doesn't have, but could be implemented just in a maxon way...if you get what i mean. the way it works is not necessarily the thing you want...its more the feature or capability we want.
thanks all.

[creart]

BEEN THERE DONE THAT too; let's hope that it works...

[astrofish]

Flingster,
I'm not sure if this is what you were meaning, but in addition to booleans I was asking for more NURBs tools.

In particular:
1. Ability to use splines to cut arbitrary shaped holes in surfaces.
2. Ability to smoothly blend between such edges on multiple objects.

These two tools alone would vastly increase the power of the NURBs modelling.

Look at tutorial 5 on this page:
dvGarage video tutorials (http://www.dvgarage.com/garage/a2d/tut/tutpast/tutpast_1_25.php)
Note how the valve is blended in with the rest of the body. I tried making this model in C4D, and this part of it was very hard to do well. Since I attempted it I've bought Edge Extrude Pro, which
would make this much easier, but still not ideal.

[H. Ikeda]

Sure, I've done that. Trust me.;)

Of course better booleans and modeling tools are expected in a future c4d, but I think it may not be appropriate to say booleans=messes. I'm sure we can easily generate a mess with booleans in c4d, but careful control of mesh structure makes it much better in some cases.

The point is, IMO, how to control the mesh structure indirectly. In fact, primitive and nurbs objects are sorts of easy-to-use and indirect methods for controlling mesh. For example, fillets in primitives, as pointed by urban ape, have good mesh structure with few small triangles.

List of parameters to control mesh---
primitive: segments, fillet & fillet segments, cap & cap segments, regular grid option
nurbs: subdivisions, rounding steps, regular subdivision option & mesh size

[flingster]

boom thanks guys...thats what i was talking about. These were some of the things people suggested before.

_________
1. Ability to use splines to cut arbitrary shaped holes in surfaces.
2. Ability to smoothly blend between such edges on multiple objects.
_________
_________
List of parameters to control mesh---
primitive: segments, fillet & fillet segments, cap & cap segments, regular grid option
nurbs: subdivisions, rounding steps, regular subdivision option & mesh size
_________

Why can't we have these tools, is it just they were never requested, or just that maxons priorities with past releases have been in other areas? i just want anything to do with modelling to be quicker and easier...:thumbsup:

[AdamT]

I think this hasn't been addressed because the focus has been elsewhere. R6 was about a new GUI. R7 was about rendering. R8 was about animation and workflow. Seems pretty obvious that R9 must address modeling as the primary focus.

But, I think this will require a major rewrite of the program, since Maxon will have to implement real nurbs algos and/or rewrite the basic poly engine to support n-gons. I suspect this will break every plugin that has anything to do with modeling....

[astrofish]

> I suspect this will break every plugin that has anything to do with modeling....

This is a potential issue, but unless Maxon decide that they can _never_ implement n-gons because of it then it's just something that's going to have to happen.

Of course, if the functionality of the main plugins that people are using is included in the base package as part of the overall modelling improvements, then those plugins become redundant anyway.

It would be unfortunate if, after Maxon do decide to do a major modelling upgrade, we still need a whole load of extra modelling plugins for general work...

I think that there will always be scope for specialist plugins, but if there's a plugin that you are using a lot in your general work, then it really ought to be part of the base package.

I think that the poly modelling tools could be improved a lot even without adding n-gons.

Just one more thought on n-gons. I wonder how much of C4Ds rendering speed is dependent upon the fact that _all_ polys are effectively quads? (internally quads are just tris with a repeated element). Maybe it's not an issue, but maybe it is....

[AdamT]

Originally posted by astrofish
> Just one more thought on n-gons. I wonder how much of C4Ds rendering speed is dependent upon the fact that _all_ polys are effectively quads? (internally quads are just tris with a repeated element). Maybe it's not an issue, but maybe it is....

I doubt that's the fairy dust that makes the engine fast. AFAIK, even n-gon modelers tesselate to tris at render time (except for Realsoft3D).

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.