Jump to content
3DCoat Forums

Idea : Associated Low Polygon Surfaces (A.L.P.S)..Andrew please read this


manticor
 Share

Recommended Posts

  • Advanced Member

Im posting here aswell as not sure where to put this idea...so sorry for doubling up.

Im really liking the direction that 3DCoat is going at the moment ,adding the colour element to voxels and surfaces is great and being able to boolean these painted models together is going to speed up the production process massively but I have been thinking that one element that is really a big time consumer still is the retopo work that comes once you have created your models ,which given that we will be able to make ever more complex and detailed models with this new technology will inevitably will take even longer.

So Im putting down an idea that Ive been mulling over in my head for the past couple of weeks as I think it would be a much needed addition to 3DCoats workflow.

The idea is basically attaching low poly surfaces to the volume layers .Sounds simple but would make a massive difference.

Say for instance you are making something quite technicaly ,a spaceship for instance .A lot of parts would be duplicated if you're using a kitbashing workflow.

So you build say some piping and then do a quick retopo and then link the low poly surface to the voxel model.An icon of some sort ( a wireframe cube,or something similar) would show up on the volume layer.Now that poly surface should be able to be altered and that would affect the voxel surface that it is bound to .( deform by cage in softimage XSI or skin wrap in 3dsmax....you get the idea ) .Simple enough idea but being able to manipulate a high res mesh with a low one enables a lot more control and possibilies .

Taking this one step further ....the model and its corresponding low poly surface can be saved as a preset in a model library .

Now it can be used to add onto another model so every time you add to it not only does it add to the voxel surface but it also adds a corresponding low poly surface in the retopo room .You can imagine how much time would be saved in the retopo room doing this.

Taking things one step further.... say you're doing a character for instance and you kit bash a character together from previously made arms,legs etc .You finish off your model and merge the whole lot to one surface.If the Autopo could " fill in the blanks" of the retopo mesh which should be quite easy for it to do as most of the guess work has been taken away as it knows how the flow lines are going then you have a complete high res model and its associated low res linked together , (or just do it by hand).

Now it would be good to create a skeletal structure from curves and bind the low poly mesh to it which would enable posing of the high res model. also expanding parts of the low poly cage could create muscle bulges and the like and as the low poly mesh stays the same topology wise it could be morphed aswell,which would be great for facial animation and any other thing like that.

And to take this idea even further and not something I'm sure is possible ...but Ill put it down here anyway.

what if the low poly surface was using ptex.It raycasts from each poly and bakes the distance to the surface to a ptex texture for each poly ( imagine if the poly surface was pushed out a short distance so there are no intersections with the high res) .Then you would have something like a negative vector displacement map, plus colour could be baked aswell to another map.Then this would make it possible to store states of the high res voxel surface completely resolution independently .Which would be great all kinds of things and you could put extra detail into areas and still be able morph and deform the voxels and never have to worry about the underlying voxel surface .

Id love to see something like this in 3DCoat...as absolutely anything would be possible in 3DCoat with this system.

And to be honest it isn't that far away as most of the tech is already there.

Link to comment
Share on other sites

  • Advanced Member

I have had similar ideas rolling in my head also. Mine have been more towards manipulating a dense sculpt via a low poly frame, especially for hard surface sculpting. The pose tool is close with the "Use free form" option, I just wish we could select a mesh from the retopo room as the free form geometry. Combine that with an option in the voxel room to "assign snap" from a layer in the retopo room to a layer in the voxel room so that the low poly will follow the sculpted surface of the voxel.

A bone-like posing system would be a welcome addition too. Curves are almost there, they just need to be able to manipulate the surrounding geometry via a falloff envelope and/or vertex weight painting. And if you created the base for a character with curves, your bones are already done.

Link to comment
Share on other sites

  • Advanced Member

Im posting here aswell as not sure where to put this idea...so sorry for doubling up.

Im really liking the direction that 3DCoat is going at the moment ,adding the colour element to voxels and surfaces is great and being able to boolean these painted models together is going to speed up the production process massively but I have been thinking that one element that is really a big time consumer still is the retopo work that comes once you have created your models ,which given that we will be able to make ever more complex and detailed models with this new technology will inevitably will take even longer.

So Im putting down an idea that Ive been mulling over in my head for the past couple of weeks as I think it would be a much needed addition to 3DCoats workflow.

The idea is basically attaching low poly surfaces to the volume layers .Sounds simple but would make a massive difference.

Say for instance you are making something quite technicaly ,a spaceship for instance .A lot of parts would be duplicated if you're using a kitbashing workflow.

So you build say some piping and then do a quick retopo and then link the low poly surface to the voxel model.An icon of some sort ( a wireframe cube,or something similar) would show up on the volume layer.Now that poly surface should be able to be altered and that would affect the voxel surface that it is bound to .( deform by cage in softimage XSI or skin wrap in 3dsmax....you get the idea ) .Simple enough idea but being able to manipulate a high res mesh with a low one enables a lot more control and possibilies .

Taking this one step further ....the model and its corresponding low poly surface can be saved as a preset in a model library .

Now it can be used to add onto another model so every time you add to it not only does it add to the voxel surface but it also adds a corresponding low poly surface in the retopo room .You can imagine how much time would be saved in the retopo room doing this.

Taking things one step further.... say you're doing a character for instance and you kit bash a character together from previously made arms,legs etc .You finish off your model and merge the whole lot to one surface.If the Autopo could " fill in the blanks" of the retopo mesh which should be quite easy for it to do as most of the guess work has been taken away as it knows how the flow lines are going then you have a complete high res model and its associated low res linked together , (or just do it by hand).

Now it would be good to create a skeletal structure from curves and bind the low poly mesh to it which would enable posing of the high res model. also expanding parts of the low poly cage could create muscle bulges and the like and as the low poly mesh stays the same topology wise it could be morphed aswell,which would be great for facial animation and any other thing like that.

And to take this idea even further and not something I'm sure is possible ...but Ill put it down here anyway.

what if the low poly surface was using ptex.It raycasts from each poly and bakes the distance to the surface to a ptex texture for each poly ( imagine if the poly surface was pushed out a short distance so there are no intersections with the high res) .Then you would have something like a negative vector displacement map, plus colour could be baked aswell to another map.Then this would make it possible to store states of the high res voxel surface completely resolution independently .Which would be great all kinds of things and you could put extra detail into areas and still be able morph and deform the voxels and never have to worry about the underlying voxel surface .

Id love to see something like this in 3DCoat...as absolutely anything would be possible in 3DCoat with this system.

And to be honest it isn't that far away as most of the tech is already there.

I agree whole heartedly...I said something along this lines a while ago...but I think you've articulated it pretty well. I think most of the functionality is already there, but it just needs to be linked together. If done properly, it would make such a big difference in workflow. In my opinion, it would make a much more organic, intuitive workflow. I think this is the one thing that will tie everything together.

I really hope it happens.

-G

Link to comment
Share on other sites

  • 4 months later...
  • Member

I totally agree, in fact I was thinking about something similar as well. Basically what I believe needs to be improved is the linear workflow, this meaning that when you make a sculpt, then retopo, then paint room, you basically can't go back to sculpting (high detail) without having to manually snap vertices in the retopo room later, and then merge again into the paint room. I'll make a workflow example: In mudbox, when I'm not happy the way an asset is looking in my 3d program of choice (maya for instance), I go back to mudbox, and can change anything, completely move the character's proportions, add or reduce detail, paint textures, and then simply re export the textures and the base mesh w/displacement map, because the base mesh adapted to my adjustments in the high poly mesh. In 3d-coat if I'm not happy with something, I have to sculpt, then in retopo room adjust my base mesh to the new sculpt, then merge again into the paint room, and then re-import my textures to keep painting... And if I make an adjustment while painting in the "sculpt room", this change won't be replicated in my voxel mesh, therefore disconecting the workflow and leaving me only with the "merged" base mesh to do what i can with it. Don't get my wrong, I believe 3d-coat is way better than mudbox right now for painting and sculpting, but as a "pipeline tool" in production, I can't help but going back to mudbox sometimes, specially when I need to make minor adjustments to an asset both in textures and sculpting at the same time, for this mudbox is just faster (in fact it is a very good pipeline tool). So, how can this be fixed in 3d-coat? By having an option to "adapt" the retopo mesh and/or the paint room mesh to the voxel room mesh. I was thinking that even an automatic vertice snapping to the voxel/surface mesh after a stroke is made, only to the vertices "touched" by the brush would be enough as a fast solution, since it wouldn't be slow cause it's only an area of the retopo mesh (that should have a limited polycount) that get's snapped gradually.

Link to comment
Share on other sites

  • Contributor

+1 to these ideas!

Overall I love 3D-Coat more than Zbrush or Mudbox, but 3D-Coat has one big drawback, and that is the exact type of thing you all are talking about. There is a big disconnect between voxels and polygons, and sometimes this gap kind of ruins everything in the end. That's why my current workflow still goes like this:

1) Build semi-detailed voxel base mesh in 3D-Coat

2)Retopo in 3D-Coat

3)Send my base mesh to Zbrush or Mudbox for detailing

Then if I need to go back and make changes, I just do it in Mudbox or Zbrush because I have the subdivision levels stored there that I need. I think this workflow is a real shame because it is possible to make highly detailed and beautiful sculpts in 3D-Coat, but I never want to risk going there in case I am asked to make changes to my lower poly base mesh...

Link to comment
Share on other sites

  • 3 weeks later...

Basically what I believe needs to be improved is the linear workflow

I think most of the functionality is already there, but it just needs to be linked together

YES ! GREAT idea for development.

This is -for me too- the exact point where is necessary to be focused before a v4. release.

+1

Link to comment
Share on other sites

  • Advanced Member

Yes I would also like if you could extrude polygons (no need for any fancy operations, just simple extrude only) and then press a button in order to convert those polygons into voxels at a certain density and thickness.

Link to comment
Share on other sites

  • 1 month later...
  • Member

A bone-like posing system would be a welcome addition too. Curves are almost there, they just need to be able to manipulate the surrounding geometry via a falloff envelope and/or vertex weight painting. And if you created the base for a character with curves, your bones are already done.

I was happy to see the mannequin being added.

That said, I would really like to see a solid, and intuitive posing system in 3DCOAT as well.

We have these marvelous Voxels to sculpt with...the only other thing I would want is a solid foundation

in the form of a skeletal Posing System to build up with Voxels and be able to pose them intuitively.

Voxels and a Posing System would go together like peanut butter and chocolate. :)

Link to comment
Share on other sites

  • 1 year later...
  • New Member

+1 to these ideas!

Overall I love 3D-Coat more than Zbrush or Mudbox, but 3D-Coat has one big drawback, and that is the exact type of thing you all are talking about. There is a big disconnect between voxels and polygons, and sometimes this gap kind of ruins everything in the end. That's why my current workflow still goes like this:

1) Build semi-detailed voxel base mesh in 3D-Coat

2)Retopo in 3D-Coat

3)Send my base mesh to Zbrush or Mudbox for detailing

Then if I need to go back and make changes, I just do it in Mudbox or Zbrush because I have the subdivision levels stored there that I need. I think this workflow is a real shame because it is possible to make highly detailed and beautiful sculpts in 3D-Coat, but I never want to risk going there in case I am asked to make changes to my lower poly base mesh...

This is exactly the problem 3d coat has, one "easy" solution would be the posibility of subdivide going up and down in retopo room. Right now you can subdivide de geometry doing snap with the high res mesh, but you cannot go down into a lower res. I think this is a must, even only for retopo, because you are not completely sure the retopo mesh is going to produce the result you want when the render subdivide the geometry at render time.

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...