Displacement mapping 10 fold

General discussion about the development of the open source Blender

Moderators: jesterKing, stiv

Post Reply
Money_YaY!
Posts: 442
Joined: Wed Oct 23, 2002 2:47 pm

Displacement mapping 10 fold

Post by Money_YaY! » Mon Apr 07, 2003 3:18 pm

This is Zbrush's design of displacement mapping.
I post this since somebody told me that a well designed example
would make developers try it just a little bit more.
http://www.pixolator.com/zbc-bin/ultima ... 1&t=008428

Note Zbrush is still testing it out and has not released it yet.

Synops: A very low mesh poly with a grey scale image attached creates
a detailed image " In the render not the 3d view "
So as a need to not need a super high poly mesh.

I belive this would increase Blenders modeling details power 10 fold.

Mostly I quote this as a repost of the lact of modeling tools for blender
comparied to Wings3d. Yet we can use Wings on the side. Even they
do not have displacement mapping.

Well thats all.

Dani
Posts: 143
Joined: Fri Oct 18, 2002 8:35 pm

Post by Dani » Mon Apr 07, 2003 3:41 pm

i think JMsoler wrote a diplacement script that worked on all sorts of objects, unsing the normal vector. However, I think you need a nice poly count in order to get acceptable results... The subdivision process could be launched on rendering only and not modifying the original mesh... dunno... just theories.

Maybe I should ask JMS about his script.

Dani

Money_YaY!
Posts: 442
Joined: Wed Oct 23, 2002 2:47 pm

Post by Money_YaY! » Mon Apr 07, 2003 3:59 pm

arrg it's in French. Translate.

I have downloaded this before I recall and it did not wotk Will try it again.

OSX of cource

cessen
Posts: 109
Joined: Tue Oct 15, 2002 11:43 pm

Post by cessen » Mon Apr 07, 2003 4:30 pm

Displacement mapping should be possible with external renderers (for instance, I know that Aqsis fully supports displacement mapping). This would probably be a good thing to let Green deal with in his RenderMan support.

Support of displacement mapping directly in Blender (along with a lot of other features) should probably wait until Blender 3.0, in my opinion.

Money_YaY!
Posts: 442
Joined: Wed Oct 23, 2002 2:47 pm

Post by Money_YaY! » Mon Apr 07, 2003 4:44 pm

OOOoo kay I will shut up now and try to figure out how to tie this into blender
^v^

thorax
Posts: 320
Joined: Sun Oct 27, 2002 6:45 am
Contact:

Re: Displacement mapping 10 fold

Post by thorax » Thu Apr 10, 2003 10:25 am

Money_YaY! wrote:This is Zbrush's design of displacement mapping.
I post this since somebody told me that a well designed example
would make developers try it just a little bit more.
http://www.pixolator.com/zbc-bin/ultima ... 1&t=008428

Note Zbrush is still testing it out and has not released it yet.

Synops: A very low mesh poly with a grey scale image attached creates
a detailed image " In the render not the 3d view "
So as a need to not need a super high poly mesh.

I belive this would increase Blenders modeling details power 10 fold.

Mostly I quote this as a repost of the lact of modeling tools for blender
comparied to Wings3d. Yet we can use Wings on the side. Even they
do not have displacement mapping.

Well thats all.

See my messages in the topic on UV maping is Passe..
I have a solution that is just under our noses and really easy to implement. The idea is 3D -> 2D -> 3D mapping.. The concept
that more than just 2D can be mapped and the underlying imagery doesn't have to be pixel data but that it can be perturbed or instantiated on a surface specified by UV mapping as it exists..

The idea is the same as Zbrush.. By the way The normal vector can
be used to determine displacement mapping, the hard part is handling
polygonal mismatch between displaced surfaced (fold removal) and
clever dynamic tesselation, but ultimately why use a a Z-buffer as a source for bump maps, why not combine actuall 3D geometry with image data combined with a 2D shader to influence the dynamic tesselation of a surface? It allows you to reuse what vector based drawing already exists in the 3D space as well as use procedural texturing, and so on to influence mapping of the object.. I think Zbrush's lacking is they think
of everything so simply as 2D mapping onto 3D objects.. Why do we look to static imagery to texture 3D objects? Why not use something more dynamic and infinitely precise and recursively easy to define?

Why be like Zbrush, why not be better?

Money_YaY!
Posts: 442
Joined: Wed Oct 23, 2002 2:47 pm

Post by Money_YaY! » Thu Apr 10, 2003 4:59 pm

"Why be zbrush why not be better"

Well of course. But by the time we are better They will be better.

Heh , But we have to get their sometime. Instead of talking about it.

jms
Posts: 33
Joined: Sat Oct 19, 2002 3:04 pm

Post by jms » Sun Apr 13, 2003 8:30 pm

Money_YaY! wrote:arrg it's in French. Translate.

I have downloaded this before I recall and it did not wotk Will try it again.

OSX of cource
You have to use a vertex painted mesh with uvcoord.
Zoo-3D.Blender, Ze French-Speaking Community SKB My french book about Blender.

thorax
Posts: 320
Joined: Sun Oct 27, 2002 6:45 am
Contact:

Post by thorax » Mon Apr 14, 2003 11:43 am

Money_YaY! wrote:"Why be zbrush why not be better"

Well of course. But by the time we are better They will be better.

Heh , But we have to get their sometime. Instead of talking about it.
Well displacement mapping isn't the holly grail, note it will
intersect itself too, just like real polygons, so your surfaces have to be bump-map thin..

JA-forreal
Posts: 0
Joined: Sat Mar 22, 2003 10:45 pm

Post by JA-forreal » Mon Apr 14, 2003 8:37 pm

ultimately why use a a Z-buffer as a source for bump maps, why not combine actuall 3D geometry with image data combined with a 2D shader to influence the dynamic tesselation of a surface? It allows you to reuse what vector based drawing already exists in the 3D space as well as use procedural texturing, and so on to influence mapping of the object.. I think Zbrush's lacking is they think
of everything so simply as 2D mapping onto 3D objects.. Why do we look to static imagery to texture 3D objects? Why not use something more dynamic and infinitely precise and recursively easy to define?


While working with 3d painting software I had wondered if one could combine 2d textures with the 3d geometry surface to influnce the viewed shape of the 3d surface in Blender. I never brought it up here because I thought that the Idea was alittle too far fetched. In reading you coders tech notes and I'm not even sure I can make since out of much of what you say. But I hear you anyway and the info I think I see you pointing out looks promising.

Post Reply