renderdemon

I agree also,masking is needed. The only problem I see here is that masking in pose tool is really slow,so a different system should be used. Also loading/saving mask could be cool.

ZSphere 2 seem to have a great advantage over voxel sculpting,a not destructive approach. Even if I really like voxel sculpting,posing a mesh is hard,and you can't do mistakes,you can't jump back and forward between poses.

This is a great news!

Of course at some point we have to increase resolution!

If I can give you and advice try to stay as more as possible on low resolution. Only when you are completely sure that what you are doing is going well increase res. The disadvantage with volumetric sculpting is that non linear workflow isn't so useful,so the more you stay on low res,less mistakes you do.

I also agree with this. 3dcoat makes really easy and funny starting new shapes. Doing full bodies it's so funny,the problem is the head.....it seems you never had the right amount of details. btw,nice works LJB .

It's seems extremely cool! Probably the best remeshing scheme I have seen. Btw,a thing I would like to see in 3dcoat is also a guided(bye the user)field remeshing scheme. As the programs is able to paint on the mesh,we could paint the field vectors directly(a bit like when you do in a 3d applications a painted fur map to give to the fur an initial direction). After and automatic algoritmhs could use the painted field to improve the quadrangulation(as user I would like to define the general direction and where the poles are). A solid remeshing algorithm could use both, completely automatic without the user map(as now),or with the help of this 3d direction map(probably here the difficulty is having a good way to use colors. it's not easy rapresent direction with colors) Bye

It's really a good idea. About brush feeling,I think a great thing could be having 2 modes(that gives really different result) 1)Wet mode 2)Dry mode. When you use real clay(depending of what kind of clay) you can use water to blend together the surface. With water the surface is also less hard,with a soft feeling when you manipulate it. In dry mode the surface is hard,more solid,mainpulation is different,give more precise result . Thinking better to it,it could be cool if every brush(or much better the volume layer we sculpt) has a new property going from 0 to 1 called WET that affects how the brushes work making the volume more elastic ans soft when is 1 and not elastic and hard when wet is 0(default could be 0.5) To be honest,I don't know technically how can be done,if someone has some ideas to share for helping Andrew it could be really cool(a thing like this could be a great killer application,even Zbrush doesn't have completely this kind of feeling) bye

I think Oliver Thornton has completely right. How can we brainstorming brushes improvement without thinking to how works the system? btw, a thing I would like to see is more or less the same absolute strength for the smooth brush, Depending on how low is the resolution,strength seems to be extremely high at low levels,falling down for high levels of resolution. Maybe an user option that can scale(multiply) automatically the strength when resolution go up,on low res the smooth is too strong(so it should reduced a bit),on upper res is weak(it should increased a bit). Bye

Really interesting!

Uniform isotropic distribution is the main advantage of voxel sculpting Without that it's impossible to create new geometry,I think there is no way to have topology freedom and control over the meshing(you can use different layers doing the sculpt in layers). It's the main disadvantage also. About that,people here seems not understanding that brushes will never be so good like we wish. If we think to poly sculpting(Zbrush,mudbox,blender),we can generalize that in these programs we sculpt(I mean when we do a stroke) over an arbitrary patch surface(actually more patches stiched together),like a nurbs surface but with linear interpolation. For example,if we have a plane subdivided a bit(like a patch surface) and we displaced the polys along the normals,what we have at the end? Again a patch, where some vertices are upper or lower(but the surface is continue in a LINEAR way because the connections between vertices doesn't change or break). It's like working with linear curves(polylines),even if they aren't smooth they are connected,and never stepped(stepped is a square curve,if you think to audio synth programs or 3d animation curves you can understand easily how different is a square curve compared to a linear curve). Even if c0 continuity isn't good enough to give smooth result,phong shading and subdivision algorithms make possible to have smooth and detailed surfaces,and c0 become c2 or c1,depending on where we are on the cage subdivision surface(poles). So,displacing a subdivision surface creates again a subdivision surface. In voxel it's a completely different story. Every time we do a stroke we are discretizing(like doing an audio sampling) the stroke,and the voxel resolution is the step used. So if we do again the plane previous example,we start with a plane(as voxel the plane has s thickness)and after some strokes we have a discrete(stepped) version of the actions we have done. So the only way to have a clean results is using small step(it's depend for sure of what we are doing). So,for me the only way to improve that is trying some adaptive step size for voxel,as you do stroke the program analize your action and choice the right resolution(but how this can work I really don't know) Bye

I agree with LJB,as it works now I think sketch isn't useful(IMHO a sketch tool must be able to do complex(topologically) shapes,not simmetrical stuff(Maybe using more projections) I don't think kay_Eva you can say it does 50% of work,at lest if you model something more complex than basic simmetrical shapes. I'll continue to start from a sphere....

Sketching seems interesting!

I don't agree much kay_Eva, I don't think linking this kind of examples prove nothing(I don't have nothing against that use but it can't be the solution for everything) Personally I don't think using another application to do volumetric sculpt isn't particularly interesting to show 3dcoat powerful. Yes,you can merge object from outside,but this doesn't mean that the sculpt tool haven't to improve,maybe sometimes you will need to manipulate with brushes merged geometry. I hope one day hard surface can be done better with better brushes and/or tools INTO 3dcoat,not because hard surfaces need to be done with volumetric sculpt(it's not the right tool)but because it could mean that the tool set is more mature. To be clear,nothing against tinker's work(I like it),but I think doesn't prove nothing,for example what will happen if someone here go on stanford's web site,download lucy(the 3d scanned data),merge it into 3d coat,volumetric sculpt will allow to do extremely clean an great models?(personally I think yes,it can be done,but will take more time because brushes need some work) Stanford lucy P.S Btw, personal thanks to Andrew,he is doing a great work and the tool is becoming everyday(literally)better,we have simply to see what are the limits of this technology,personally it's not clear to me.

This is actually the same problem I have posted on 3.00.08 beta thread. I hope Andrew will fix it,and makes select with pen not affected by undo(for the selecting area)

If I have to be honest,when I have bought 3dcoat I didn't read the EULA,I thinked was a bit more standard. As a person interested mainly in anatomy and organic sculpting , the old version of the EULA coulded be really limiting,this is also a reason why I stopped posting volumetric sculptures on the forum,I don't want to offend anyone(even if I think that my work,like the ones made by other 3d organic modellers isn't offensive). It's really important to be clear about that point,the new version of the EULA from my point of view it's much more acceptable.

As I allready said,this was a test,I have tried to understand better the limits. The volume is more or less 10 millions,and it's not enough to really sharp lines,this is the reason why I have tried to do all by hand,without alpha brushes,simply stroking lines to see what kind of stuff can be done(for example,elephants skin style is doable,pores not so much,are harder because the tiny holes are not sharp enough,skin folding need also sharper depth curves) I think that surface brushes helps a bit(but at the end you can't have more resolution),maybe with another level of subdvision(increasing again the volume resolution)the detailing can be really close to sharp line,but 40 millions for the head are a bit too much(and it will become slow). In the future(3/5 years),this can become an interesting stuff,with computer with 64 gigabyte,faster computer,faster graphics cards handling 200/300 millions can be doable. btw,thanks,it's 2/3 days full work(splitted in several evenings). bye

I can't see a reason for smoothing backside,so if you want that surface smooth affects only front facing for me it can be good.

Andrew,can you do backside masking also for the smooth brush? In general,affecting backface or not should be an user option selectable for every brush,if you can't do it for voxel mode at least for surface mode. When you do thin surfaces you'll never want affecting the backside(the only useful brush in this case is an inflate brush,that push along the geometry normals,not the view normals) now the flatten is great.

Hi,a little test I want to see what kind of detailing can be done with volumetric sculpting Doing sharp detailing is hard,I'm not satisfied,but it'not so bad(this is not finished,I have started only with tiny volumetric strokes(extrude brush),but recenty Andrew improved the voxel to surface workflow,probably using surface brushes helps a lot in detailing,if I started from scratch now probably I'll use surface for detailing. Another view Bye

Beautiful!

As the tool is now,doing clean high frequency detailing isn't possible. The way the tessellation works gives always isotropic density,this is good,but to have really sharp details probably you need a huge polycount,too huge for any hardware system. To make an example,for a head,I think that even with 8 millions you can't do really sharp details(pores and skin stuff). For medium frequency and low frequency feature, I think that volumetric sculpt is really good.

Yes,I mean that ! Btw,I like ideas for improvements,I think here all we want the same thing, making the tool always better.

Carve is useful for snapping a layers to another. I agree with other stuff,even if I think that hiding with a medium/high res works not so bad(how can you do ears and fingers without medium/high resolution?)

The more strength on surface smoothing is a great improvement,thanks! In mho,now 3 things are needed to improve the workflow as smooth and fluid as possible. --First,pose tool improvement 2 things here are bothering me. If you use the pen tool to paint the masking area("select with pen") and you do a mistake,if you do an undo,the manipulator axes are lost,you have to redo the axis line,the rotation/scaling/moving axis shouldn't be affected by undo on pen selection masking And the "select with pen" must be more precise,I don't know why,but selecting a forearm is a pain,you always have areas where the masking isn't full,so when you start the posing some part of the surface/volume(I have found that here the behaviour is the same)aren't trasformed completely but only for a small percentage,breaking the shape,and you have to redo(sometimes losing the previous selection like with the pose manipulator) So please make the undo not affecting selection and axis but only posing in this contest. ---Second,as someone has allready pointed out, surface flatten shouldn't work on backside(maybe an user option for brushes could be cool). --Third, if you find a way to switch to a low resolution to do moving and posing transformation the tool could be really easier to work with(sorry for bothering,I 'm sure you know this) A good artist can work without multiresolution but it's a limitation,it's force the user to a linear workflow,if you have fine arts experience isn't a problem but the beauty in computer art is freedom,if your target is computer artists this limitation can give really frustration(btw,I find linear workflow really interesting to improve personal skills,in this way you have to focus better on shapes) Bye