How far do you guys think the new polygon modelling tool will be taken, because I see them currently being really good for base mesh creation but not as efficient as traditional polygon modelling.
How far do you guys think the new polygon modelling tool will be taken, because I see them currently being really good for base mesh creation but not as efficient as traditional polygon modelling.
I haven't tried it myself, but it seems like a really strange design choice to me.
People have been praising it, but I can't imagine it being as good as traditional polygon modelers either no matter how much they put into it, if only because of the lack of 3D coordinates if nothing else.
And they've spent a massive amount of time and resources on it.
Hmmm, seems like the crease option is broken in R7. I have a model with a hole and in R6 the crease keeps the edges from shrinking when subdividing. In R7 it just doesn't work.
Increase your Crease edge tolerance to 180', or crease by polygroups instead.
How far do you guys think the new polygon modelling tool will be taken, because I see them currently being really good for base mesh creation but not as efficient as traditional polygon modelling.
I think it can be taken pretty far, but it isn't designed as a complete replacement by any means. Ryan Kingslien explains it using a soft-hard / loose-precise coordiate system. Zbrush had the 'loose & soft' quadrant locked down already, and all the v4 versions have been trying to control the 'loose & hard' quadrant too. ZModeler is the next step into controlling the entire loose half. It's not precise, but its there so that creative minds can dig in and start building things out of thin air.
@cryrid: Didn't work, it's already on 180 and crease by polygroups has the same result. In R6 it works fine with the same model Thanks anyway.
EDIT: Actually it does work, but not on borders; "crease all" tags all the edges, including the borders, but when subdividing it only has effect on edges that share 2 or more polygons.
How far do you guys think the new polygon modelling tool will be taken, because I see them currently being really good for base mesh creation but not as efficient as traditional polygon modelling.
I think it can go all the way (baby!). I think that some of the easier things to model in the traditional way are harder to do in Zmodeler. But also, some of the hardest things to model traditionally can be much easier in Zmodeler.
@oniLolz: Which effect are you aiming to achieve, and do you have a file you could upload somewhere? Unfortunately I'm not in a position to be able to switch between versions to compare.
I think it can go all the way (baby!). I think that some of the easier things to model in the traditional way are harder to do in Zmodeler. But also, some of the hardest things to model traditionally can be much easier in Zmodeler.
I love it!
i dont believe in this as long as i dont see a real weapon or car modeled with it. high quality, no comprimises, just like the real thing.
it seems gimiky to me like most of the recent features, that work perfectly awesome when working in a style that suits them, but utterly useless for anything else.
i have not used it though so im just speaking by the things i have seen created with it.
Interesting that all the critics of zmodeler haven't used it, what's stopping you?
I bet I can slide an edge, or perform a fast tweak faster than you can GoZ to an external app.
what stops the supporters from doing something complex in shape and quality thats not just a boxy model? :P
probably the same, time
well i am not criticising it, but i'll remain sceptical until i have time to test it or someone shows something complex being made with it.
so far what i saw have been very simple models, all youtube videos are rather basic. one video showing a complex jetengine, claiming 6 hours in 6 minutes, just imports the complex parts and does nothing that you couldn't do the same way with any modeller.
i'm willing to believe the hype but not without a proof ^^
Why does something complex have to be made in it to prove it's usefullness? I modeled everything on my current project inside of zbrush, it was faster than jumping into an external app.
R7 is a free update that finally makes zbrush 64 bit, zmodeler is just a small part of that.
In many ways Zmodeler is very similar to the Zbrush retopo tools. Artists with a very high level of "Zbrush tolerance" (=not bothered by the cumbersome subtool system, the slightly irritating navigation scheme, and many other little things) will probably power through it just fine and manage to produce great stuff with of it - since being able to stay within one unified app is a always an obvious plus. The mere fact of having to save out an OBJ can get one out of the zone, no doubt about that.
However, and objectively speaking, Zmodeler as a whole does feel unnecessarily convoluted and in many ways very antiquated - almost like a "Poser" equivalent to polygon modeling. Being able to boxmodel stuff is cool, but having to jump through the usual Zhoops in order to combine components or even simply merge vertices is a drag.
I think that people suggesting "Zbrush-only" artists to spend a little of time doing some regular polygon modeling work are simply wishing for everybody to be as efficient as possible instead of just blindly sticking to one's app just because of ... reasons ? From personal experience I tend to prefer workflow unification to happen the other way around anyways (that is to say, being able to access sculpt tools while doing regular poly modeling, rather than the opposite), as this is the paradigm shift that sped up my workflow immensely in recent years. Suggesting others to go that route is not a sign of me "hating on" Zbrush-only workflows, but rather, an attempt at sharing experience.
Now to be fair, most regular 3d apps have abysmal lowpoly toolsets by default and require extensive customization, so I cannot blame anyone from wanting to stick to something that they are confortable with ...
Yep I can totally see this happen too, as I went that exact same route myself ! Basically trying all the different approaches to see which one just clicks the best, coming from a regular modeling background. I think the last time I personally attempted a Zbrush-only workflow was about a year ago, and it resulted in about 50% satisfaction (being able to work in a unified environment really is great) and 50% frustration (at the realization that when taken individually, most of the steps ended up being more complicated than if they had been done in specialized packages).
It really is a never ending quest for sure ... In a sense I am very glad to see the Zbrush guys bringing Sketchup-like tools to the masses as I feel like such a workflow should have been more widespread for years now. The modeling tools in both Max and Maya really are stagnating.
(now I could start talking about how awesome the Blender workflow integration is, but that's for another thread !)
I'm able to do it just fine in 4r6 but not in 4r7 ... Not sure if a setting changed or what. I tried loading a 4r6 brush that I know worked (curveStandard) and have had no luck.
I like new nanomeshes in r7. They are really useful and long awaited. Multi mesh insert brushes made me crazy before.
But as always they did it in their alien toyish manner like wanted to invent most weird way to control instance population . A subject 3d soft went long years already to find convenient and effective approaches for .
So every time I doing nanomesh population I think why, why they just didn't do it same simple way like Vue for example, or countless object/clone painters for Max
Replies
i think it is pretty fair to say, that zremesher still is meant for sculpting meshes not for production ready meshes which will animate.
I have just been mucking around with dynamic subd and its pretty great for blocking out meshes volumes
It feels more like the old way of building head meshes, then you can sculpt on top of it
I haven't tried it myself, but it seems like a really strange design choice to me.
People have been praising it, but I can't imagine it being as good as traditional polygon modelers either no matter how much they put into it, if only because of the lack of 3D coordinates if nothing else.
And they've spent a massive amount of time and resources on it.
Is this happening to anyone else?
I think it can be taken pretty far, but it isn't designed as a complete replacement by any means. Ryan Kingslien explains it using a soft-hard / loose-precise coordiate system. Zbrush had the 'loose & soft' quadrant locked down already, and all the v4 versions have been trying to control the 'loose & hard' quadrant too. ZModeler is the next step into controlling the entire loose half. It's not precise, but its there so that creative minds can dig in and start building things out of thin air.
EDIT: Actually it does work, but not on borders; "crease all" tags all the edges, including the borders, but when subdividing it only has effect on edges that share 2 or more polygons.
EDIT2: Example
I think it can go all the way (baby!). I think that some of the easier things to model in the traditional way are harder to do in Zmodeler. But also, some of the hardest things to model traditionally can be much easier in Zmodeler.
I love it!
i dont believe in this as long as i dont see a real weapon or car modeled with it. high quality, no comprimises, just like the real thing.
it seems gimiky to me like most of the recent features, that work perfectly awesome when working in a style that suits them, but utterly useless for anything else.
i have not used it though so im just speaking by the things i have seen created with it.
I bet I can slide an edge, or perform a fast tweak faster than you can GoZ to an external app.
what stops the supporters from doing something complex in shape and quality thats not just a boxy model? :P
probably the same, time
well i am not criticising it, but i'll remain sceptical until i have time to test it or someone shows something complex being made with it.
so far what i saw have been very simple models, all youtube videos are rather basic. one video showing a complex jetengine, claiming 6 hours in 6 minutes, just imports the complex parts and does nothing that you couldn't do the same way with any modeller.
i'm willing to believe the hype but not without a proof ^^
R7 is a free update that finally makes zbrush 64 bit, zmodeler is just a small part of that.
However, and objectively speaking, Zmodeler as a whole does feel unnecessarily convoluted and in many ways very antiquated - almost like a "Poser" equivalent to polygon modeling. Being able to boxmodel stuff is cool, but having to jump through the usual Zhoops in order to combine components or even simply merge vertices is a drag.
I think that people suggesting "Zbrush-only" artists to spend a little of time doing some regular polygon modeling work are simply wishing for everybody to be as efficient as possible instead of just blindly sticking to one's app just because of ... reasons ?
Now to be fair, most regular 3d apps have abysmal lowpoly toolsets by default and require extensive customization, so I cannot blame anyone from wanting to stick to something that they are confortable with ...
It really is a never ending quest for sure ... In a sense I am very glad to see the Zbrush guys bringing Sketchup-like tools to the masses as I feel like such a workflow should have been more widespread for years now. The modeling tools in both Max and Maya really are stagnating.
(now I could start talking about how awesome the Blender workflow integration is, but that's for another thread !)
[ame]
I'm able to do it just fine in 4r6 but not in 4r7
But as always they did it in their alien toyish manner like wanted to invent most weird way to control instance population . A subject 3d soft went long years already to find convenient and effective approaches for .
So every time I doing nanomesh population I think why, why they just didn't do it same simple way like Vue for example, or countless object/clone painters for Max
Glad theres a work around, can anyone point me to a tutorial explaining this workflow? I dont really know the interface well enough.
for edges:
1)hover over an edge
2)hold down space to bring up your zmodeler actions
3)select slide
4)select edgeloop complete
now every time you click and drag an edge you'll be edgeloop slide mode. I don't think there's a way to slide polyloops though.