Jump to content
3DCoat Forums

Brushes either based on a vector displacement map or mesh


mocaw
 Share

Recommended Posts

  • Advanced Member

In another thread, several of us discussed the need for brushes that not only displace outward but also inward. Using a displacement map with the zero point set to mid gray is the traditional way.

Taros found another solution to regular displacement "up and down" but this one works in any vector. Instead of a scalar map, it uses a vector map and he showed how it's used in Mudbox to create complex geometry on the fly!

Which made me think that not only should this feature be in 3DC, but that it truly is something that just seems MADE for voxels.

If there was a way to import a mesh as a brush and have it work the same way (maybe generate a vector displacement map on the fly) and be able to slowly push in the direction of the vectors (so you can control the depth) that would be even more awesome.

To go further, if you could capture a portion of the surface of a voxel sculpt, similar to how you can during painting, and have that made into a brush that would be even more amazing.

This would make it possible to have brushes with over hangs etc. Great for making trees, grass, ears, cliffs, caves, you name it.

Link to comment
Share on other sites

  • Member

"it truly is something that just seems MADE for voxels"

+1 agree Vector Displacement Brushes are amazing

in 3DC it could be called the " Magic " brush :p:

if they can can grow ears, teethes with polys, just imagine in 3DC ... possibilities are endless with Voxels

" To go further, if you could capture a portion of the surface of a voxel sculpt, similar to how you can during painting, and have that made into a brush that would be even more amazing. " :yahoo:

Link to comment
Share on other sites

  • Advanced Member

Sorry guys but I really fail to understand where you see the advantage of using Vector Displacement Maps over Merging objects on the Pen.

Are you familiar with that option at all?

I see the Vector-Displacement like shown in the Mudbox-Ear-Video as elegant method to deal with a notorious shortcoming of meshes

(that one can not seamlessly merge complex Geometry with already existing stuff).

But with Voxels this whole problem does not exist to begin with!

In 3DCoat on can merge whatever model you have on disk - seamlessly. Here one deals with it as real Geometry, can see its orientation

and output-scale in realtime. This is awesome.

In Mudbox one has to deal with a far more abstract concept of a non-human-readable map instead which tells you for the first time what it does

when first stamped to the Model. The user first has to learn how to convert a model to such a map and each time spend time for its creation.

How is this tinkering with strangely coloured bitmap any better than using the actual Geometry which one can merge directly to an existing Voxel-Layer

or first put on another Layer and fine-tune before melting it together with existing stuff?

In my opinion on should spend time to optimize the Merge on Pen Workflow but one should not implement a (admitedly fascinating) technology

to bypass conceptual weaknesses of another Geometry-Type (Meshes).

Link to comment
Share on other sites

  • Advanced Member

Sorry guys but I really fail to understand where you see the advantage of using Vector Displacement Maps over Merging objects on the Pen.

Are you familiar with that option at all?

Yes, and it is close, but not quite as easy to use PLUS it is missing an important aspect- it does not "grow" the object with pen pressure. I'm not talking just about scale- I'm talking about the severity of the displacement. When you go grab the ear object- it just plunks it down. Granted, you could then smooth it etc. but that's not nearly as elegant. What if you were creating terrain out of a pre-defined bolder? They'll all look the same with just the object method we presently have. It also has a pre-defined vector, that while in someways is less flexible, the vector map always is orientated the right way- no need to fiddle with the settings.

I see the Vector-Displacement like shown in the Mudbox-Ear-Video as elegant method to deal with a notorious shortcoming of meshes

(that one can not seamlessly merge complex Geometry with already existing stuff).

YES- but again, just because it is particularly due to a short fall by MB tech, does not make its concept irrelevant. AFAIK though, that could be adaptive, dynamic sub divisions in MB, which are not really all that bad. Not as flexible as voxels, but still a lot more wiggle room than the divide all method.

But with Voxels this whole problem does not exist to begin with!

In 3DCoat on can merge whatever model you have on disk - seamlessly. Here one deals with it as real Geometry, can see its orientation

and output-scale in realtime. This is awesome.

Hey, I'm not saying we should get arid of this great tool, merely either have a new one or better yet augment this one. Sometimes flexibility comes at the cost of usability for certain tasks. When you need that power it's great...but sometimes it's nice to have a "dumbed down" version for working quickly in certain aspects. However, I'd argue that for certain tasks, this method is not "dumbed down" but an more elegant solution. Key words here are for certain tasks- not all.

In Mudbox one has to deal with a far more abstract concept of a non-human-readable map instead which tells you for the first time what it does

when first stamped to the Model. The user first has to learn how to convert a model to such a map and each time spend time for its creation.

This is why I suggested it be instead a layer on a pen, or created on the fly from a mesh capture. There is no reason we couldn't have a screen cap of the object that represents it. You can't say what a mesh is without opening the file or going by the name either- unless you are that good at storing matrices in you head!

How is this tinkering with strangely coloured bitmap any better than using the actual Geometry which one can merge directly to an existing Voxel-Layer

or first put on another Layer and fine-tune before melting it together with existing stuff?

Again, there is no reason you'd have to "look" at the vector map directly. And once again, a vector map tells things where they WANT to be, and can end up, not just "be there" in space like a mesh does. In this way you can "grow" points/voxels slowly towards that goal. Just like the difference between using a mesh "wrinkle" vs. using a displacement map. It's much easier to on the fly control the severity. The same goes with a vector displacement, however it allows this to happen in three directions, not just one.

In my opinion on should spend time to optimize the Merge on Pen Workflow but one should not implement a (admitedly fascinating) technology

to bypass conceptual weaknesses of another Geometry-Type (Meshes).

Well, one should not just look a solution to a different problem and write it off entirely. Their problem is merging a mesh onto another one, however, in their attempt to solve that problem, the exposed a useful adaptation- the ability to affect the severity of the 3D displacement in real time and intuitively. We are not trying to solve THEIR problem where we have none, more gleaning a generally useful concept.

Care to think of how many tools are in 3DC, that were made first in say...PS or ZB, that we take for granted? Some of them were made for fairly different us...but that doesn't mean they don't work and serve use well in 3DC!

Link to comment
Share on other sites

  • Advanced Member

Hi mocaw,

thanks for your reply!

If I understand you correctly then you would agree that On Pen merging is more telling in terms of what the

output will be. You will certainly also like the fact that one does not have to create a Map but can

create right away.

You are missing a "Grow feature" however and one could imagine quite a few others (like Tilt, Twist, Taper...).

I again don't see a reason why this shouldn't be doable in a more flexible way (= realtime editable - without need to bake

new maps) when the Merge on Pen got some Modifiers added. That was my point basically.

Link to comment
Share on other sites

  • Advanced Member

Hi mocaw,

thanks for your reply!

If I understand you correctly then you would agree that On Pen merging is more telling in terms of what the

output will be. You will certainly also like the fact that one does not have to create a Map but can

create right away.

You are missing a "Grow feature" however and one could imagine quite a few others (like Tilt, Twist, Taper...).

I again don't see a reason why this shouldn't be doable in a more flexible way (= realtime editable - without need to bake

new maps) when the Merge on Pen got some Modifiers added. That was my point basically.

Well I totally agree- if there is a way to do it that is better, either in form, function, or both, then I'm all for it! It doesn't have to be by the use of vector maps if it's better or the same even.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...