3DC V3.2 - The Future

[philnolan3d]

Well, Multi-resolution is currently in the works, so I assume that would solve this problem.

[AbnRanger]

Well, Multi-resolution is currently in the works, so I assume that would solve this problem.

But it's planned for Surfacing mode in Voxels. What I'm talking about is removing (or mitigating as much as possible) the extra hours of work AFTER you complete a sculpt in Voxels. If the Sculpt (Geometry) Room had the same level of tools and similar speed as the Surfacing/Voxel tools do, then people who import models wouldn't have to do twice the work a ZB or MB user does, by having to Retopologize and do UV's.

Like I said...accurate snapping (of the original imported model) in the Retopo room would be a more simple answer to that problem, as Retoplogizing would be a mere issue of tweaking instead of having to start from scratch. Right now, Voxel sculpting is TWICE the work geometry sculpting is in MB or ZB. It's only beneficial if you decide to start out a model in Voxels (with primitives). How efficient is it to work on brush speed in the Voxel Room, if it still requires you to have to spend hours to RE-Model something you imported?

[ifxs]

One possible answer is to have accurate snapping and more controls of snapping in the Retopo Room, so that the original model can be used to conform to the new voxel sculpt. Currently, trying to do so, even snapping to a voxel sculpt with minimal changes, causes a rat's nest. If Mudbox can offer a full set of geometry sculpting tools and be as smooth as butter, why not 3DC?

seems like it would be a given that the tools from the voxel room will make it into a poly editing/creation room, as technically the surface tools in the voxel room are work on a mesh as it is.

regarding SNAP:

I agree with you AbnRanger, although to be honest, I didnt purchase 3D Coat so I could work with poly sculpting(that title is held by ZB at the moment IMO), I use 3D Coat for work(or workflows) that is best done with Voxel sculpting as it's focus. When I snap to my voxel mesh in the retopo room, I OFTEN get errors when using the snap function on my retopo mesh, this is frustrating as it can far-too-often call for a complete manual retopo, UGH! Vertices often jump to unfavorable places on the voxel sculpt, sometimes making the retpo mesh unusable due to so many badly snapped vertices :-( For me, this happens much of the time with a SOLID voxel sculpt, and nearly ALL the time for a hollow/skin voxel sculpt.

I use lightwave, silo & blender for my poly modeling now. Though IMO 3D Coat is close behind these, I personally prefer the snap/retopo functions of silo, and especially the SHRINK-WRAP modifier/function in Blender. If you haven't had the joy of modeling/retopo in blender using the shrinkwrap modifer over a high res mesh, you're truly missing out(IMO). Much respect to Silo, it's truly a poly modeling specialist, but Blender's use of the shrinkwrap modifier allows for variable application of the snapping function using weights, and it stays dynamic(like a 3DS Max modifier) allowing you to not "freeze" the snap at any point until you choose to apply the modifier(usually just for exporting to another package). Also, you can have "layers" of shrinkwrap on top of each other in the modifier stack, each layer applying itself to the underlying sculpt using weights to control their application amount, allowing you to have a specific shrink weight set for the head of a figure, another for the torso, another for the hands, etc... IMO this truly offers superior control, and could likely be added to 3D coat as "Snap Layers" if Andrew wanted to reference/use the shrink-wrap code in blender(as its freely available to peruse). Weight controlled "snap layers" in the retopo room would really help, allowing for greater control and less errors in snapping IMO.

SInce the ShrinkWrap code is freely available, depending on the specific (GPL or not)license of that code, it might be freely available to use/add in 3D Coat(already written in C(i think)), perhaps Andrew could reference the code for the shrinkwrap mofidier and allow us in 3D Coat to have a "dynamic" application of snapping that allows for "weighting" to tighten or loosen the snapping if it's too difficult or time consuming to update the snap code to prevent the bevy of errors that I(and others) experience when snapping to a voxel sculpt.

http://wiki.blender.org/index.php/User:Jaguarandi/SummerOfCode2008/ShrinkwrapModifier

A video demo for those unfailiar with the 3DS & Blender style retopo modifier method

************************************************************************************************************

That being said, I'm trying to offer constructive comments, as I have nothing but respect for 3D Coat, as its taken over the majority of my 3D sculpting work in very little time due to it's keen design/implementation/innovation! Thanks Andrew.

[ifxs]

ZBrush and MB artists can go in with their model, sculpt away and when they are done, either export out the hi-res mesh or bake down to normal/displacement maps. With 3DC, you're only half-way done....with 2 major steps to have to do. We're talking about hours more work. Why that is being ignored is beyond me.

For some of my current 3D Coat projects, I'm using the following workflow, this might help you out if you arent already aware of it:

1: create a base mesh(nearly 50% finished already) in my chosen poly modeling app

2: merging it into the voxel room

3: sculpt in vox room on merged mesh as I see fit

4: from there I right click on the Vox tree layer and EXPORT to OBJ

4b: exporting from the vox tree instead of retopoing allows me to automatically poly reduce the actual vox room mesh(see it with the W key held down in the vox room)

5: Import the reduced mesh back into my rendering/modeling app of choice for further work, etc.

Until there is a viable snap feature in 3D Coat that doesn't destroy my quadrangulated or manually created mesh by making the verts just too much when snapping, I'm using this workflow.

P.S. If andrew implemented a "distance limitation" in world space for each vert when snapping, I'm sure he could cure many of the "spastic" jumping of specific verts when snapping. I'm sure this could be implemented quite easily.

[philnolan3d]

Like I said...accurate snapping (of the original imported model) in the Retopo room would be a more simple answer to that problem, as Retoplogizing would be a mere issue of tweaking instead of having to start from scratch. Right now, Voxel sculpting is TWICE the work geometry sculpting is in MB or ZB. It's only beneficial if you decide to start out a model in Voxels (with primitives). How efficient is it to work on brush speed in the Voxel Room, if it still requires you to have to spend hours to RE-Model something you imported?

I do see your point about the sculpting room, for the most part right now it's a useless room (not entirely of course). I don't think I'd goes as far as saying that retopo is double the amount of work. It's an extra step for sure, but the retopo room is very fast compared to sculpting in small details (in any program).

[AbnRanger]

I do see your point about the sculpting room, for the most part right now it's a useless room (not entirely of course). I don't think I'd goes as far as saying that retopo is double the amount of work. It's an extra step for sure, but the retopo room is very fast compared to sculpting in small details (in any program).

Voxel Clay sculpting is a great alternative and avoids some issues ZB and Mudbox users face with Geometry sculpting...IF you start there. However, if you have MB or ZB and you have a low-mid poly model to work with, you're best bet is to use either one of those instead of 3DC...as that workflow is far more efficient and faster than importing a model into 3DC. We've already discussed what might be done to remedy that.

Ptex is a major advancement forward, and I applaud Andrew for jumping right on it...as it poses a legitimate alternative to doing UV's. But I hope he will soon focus on making the traditional workflow (import model > sculpt/detail > bake normal & displacement maps) as efficient as it is in MB and ZB.

[digman]

Stronger or more accurate snapping tool, I agree with having. I have snapped a number of low-polygon models successfully. Snapping works well on low polygon models as long as your polygons are not packed to close together but any areas with lots of tight edge loops like around mouths, ears and eyes causes problems. Also the extremities of models have issues as well. Things like finger nails are a no go. If accurrate snapping could be accomplished that would be great as we should not have to worry about tight edge loops etc. Right now I sculpt and then resnap, sculpt and resnap again as the model takes shape. This works well but if the initial low polygon import into the retopo room or additional resnapping does not snap correctly then there is no more you can do. We need this whether we get subdivision history or not added to the surface tools.

[Nemoid_]

Well what i think is that performance in sculpting/painting and the app as a whole needs to be improved, before adding new stuff.

3DCoat has to be on par with ZB and Mudbox, to be successful.

[Javis]

I still would like to see Geometry sculpting tools receive the same equivalents as you find in Voxel Clay Sculpting. This way, those who bring in models from another program, which is most often the case, do not have to resort to modeling again (RE-topology), and doing UV's all over again. Right now, that is the weakest part of the program and hasn't been touched since it's early days. Voxel Sculpting is nice if you plan to start there from a primitive and such, but for those that want to bring in their own model, it forces more work upon you. ZBrush and MB artists can go in with their model, sculpt away and when they are done, either export out the hi-res mesh or bake down to normal/displacement maps. With 3DC, you're only half-way done....with 2 major steps to have to do. We're talking about hours more work. Why that is being ignored is beyond me.

One possible answer is to have accurate snapping and more controls of snapping in the Retopo Room, so that the original model can be used to conform to the new voxel sculpt. Currently, trying to do so, even snapping to a voxel sculpt with minimal changes, causes a rat's nest. If Mudbox can offer a full set of geometry sculpting tools and be as smooth as butter, why not 3DC?

I completely agree with your post here Don. I really feel the "Sculpt" tab tools need some love, and more work. I really like sculpting with the polygons as well, and there are times when it is flat out needed.

On that note, I don't see why the current surface voxel tools (since they are poly sculpting tools anyway) could not be copied over to the "Sculpt" tab, it would be really nice when the surface voxel tools are changed or new ones added, that somehow these two sets of tools are inherently linked.

[Nemoid_]

IMHO, polygon sculpting could be greatly improved, but i don't think it will reach ZB performance wise, just because ZB uses its own method to deal with polygon amount through pixol technology, which allows to bypass open GL to work with the mesh internally.

This IMO is why ZB hadles huge amounts of polygons without needing an uber fantastic machine, cuda or whatever. BTW with a great machine and videocard u reach impressive performance.

Voxel is the technology that distinguishes 3Dcoat from other apps, and so, i think what should actually be improved is process in which you import model and voxelize it. The less time it requires, the better.

[AbnRanger]

IMHO, polygon sculpting could be greatly improved, but i don't think it will reach ZB performance wise, just because ZB uses its own method to deal with polygon amount through pixol technology, which allows to bypass open GL to work with the mesh internally.

This IMO is why ZB hadles huge amounts of polygons without needing an uber fantastic machine, cuda or whatever. BTW with a great machine and videocard u reach impressive performance.

Voxel is the technology that distinguishes 3Dcoat from other apps, and so, i think what should actually be improved is process in which you import model and voxelize it. The less time it requires, the better.

Well, Mudbox 2009 overtook ZBrush in raw poly-sculpting speed, and without any "Pixol" tricks...so there's no reason to not have robust poly/geometry sculpting in the Sculpt tab...that mirrors the Surface sculpting tools in the Voxel room. With that and Ptex...I can see 3DC sales begin to explode...it's just a matter of getting the word out in the industry.

Having more ways to attack a model is what is going to put 3DC over the top eventually...not trying to beat ZBrush at its own game, necessarily.

[Nemoid_]

Well, Mudbox 2009 overtook ZBrush in raw poly-sculpting speed, and without any "Pixol" tricks...so there's no reason to not have robust poly/geometry sculpting in the Sculpt tab...that mirrors the Surface sculpting tools in the Voxel room. With that and Ptex...I can see 3DC sales begin to explode...it's just a matter of getting the word out in the industry.

Having more ways to attack a model is what is going to put 3DC over the top eventually...not trying to beat ZBrush at its own game, necessarily.

Well, Mudbox overtakes it but only in powerful machines, i think, while Zb works well in less powerful machines and pixol serves the app for othe purposes than merely overriding Open GL... however, of 3DCoat reaches Mudbox level would be actually awesome.

[AbnRanger]

Well, Mudbox overtakes it but only in powerful machines, i think, while Zb works well in less powerful machines and pixol serves the app for othe purposes than merely overriding Open GL... however, of 3DCoat reaches Mudbox level would be actually awesome.

The speed of the surface tools are already there, for the most part, and that's why I'd like to see the Sculpt Room/Tab be a mirror of those tools. That way you have serious capability no matter what workflow you choose. Right now, both MB and ZB are pretty linear. Carrying the Surface tools over to the Geometry Sculpting Room would put 3DC in a class by itself, IMHO.

[liseng]

The speed of the surface tools are already there, for the most part, and that's why I'd like to see the Sculpt Room/Tab be a mirror of those tools. That way you have serious capability no matter what workflow you choose. Right now, both MB and ZB are pretty linear. Carrying the Surface tools over to the Geometry Sculpting Room would put 3DC in a class by itself, IMHO.

This is exactly what I am looking for and what would speed up my 'workflow' and would make things so much easier...as I would need less applications to work with...

[WillBellJr]

I agree about not requiring 3DC to beat ZBrush at its own game.

Granted, I'm only an indie making games and don't have the drop-dead requirements of a professional production haus, but I'd prefer to see 3DC have its own feature set that works well for what it does.

I tend to agree 3DC should focus mostly on Voxels since 3DC was first at making them popular.

Though I also agree with what's been said about the poly sculpting needing some updates.

Frankly I feel the poly sculpting and voxel rooms should be folded into one Sculpting room where you can just choose which one you're working on (or both if possible) using the same tools.

If Andrew can eventually get 3DC to match ZB in polygon performance fine, but I'd be just as happy in the mid-range 1-8 million for my needs.

Anyone who needs what ZB can do already has it, why waste energy trying to match it - 3DC needs its own niche and needs to hella good at whatever it does.

The way I see it:

Get 3DC down to three rooms - Sculpt, Poly, Paint.

Let me sculpt - using imported base meshes OR starting with a voxel - let me use the same tools for both.

(Having differing tools for voxel and scuplt is confusing and makes learning (and remembering) the app harder.

This IMO, is 3DC's bread and budder; this has to work the best for peeps to take it seriously.

If sculpting with Voxels way better than with polygons, then scrap the poly sculpting and make absolutely solid turning everything into a voxel - imported base meshes should be >faithfully< converted to a voxel object and in the end after voxel sculpting, let me get my object back out as a decent polygon model.

Let me build / edit a polygon model OR retopo imported models using the same tools for both. (This is why I had said earlier that I'd like to be able to create polygon models in 3DC, the majority of the toolset is already there with retopo...)

Let me paint, generate maps and export - this is pretty much there.

-Will

[splodge]

IMHO, polygon sculpting could be greatly improved, but i don't think it will reach ZB performance wise, just because ZB uses its own method to deal with polygon amount through pixol technology, which allows to bypass open GL to work with the mesh internally.

This IMO is why ZB hadles huge amounts of polygons without needing an uber fantastic machine, cuda or whatever. BTW with a great machine and videocard u reach impressive performance.

Voxel is the technology that distinguishes 3Dcoat from other apps, and so, i think what should actually be improved is process in which you import model and voxelize it. The less time it requires, the better.

ZBrush achieves good performance through using subdivision surfaces. This means it can temporarily drop down to lower detail level when the user is rotating the on screen model. When 3D Coat supports subdivision surfaces then it should be able match ZBrush's performance.

[Nemoid_]

well voxel technology is the one allowing wonders like for example trimming the mesh, good booleans, and model regardless from mesh resolutions, so it actually frees the user from limitations polygon sculpting has. so this is 3d coat main selling point.

btw if polygon sculpting gets enhanced to the level of ZB it is even better and no one will complain for sure

[splodge]

well voxel technology is the one allowing wonders like for example trimming the mesh, good booleans, and model regardless from mesh resolutions, so it actually frees the user from limitations polygon sculpting has. so this is 3d coat main selling point.

btw if polygon sculpting gets enhanced to the level of ZB it is even better and no one will complain for sure

Yep, voxels are wonderful. But they can be a problem when you need detail. If you need to make a detailed human figure from voxels then you're gonna be hitting something like 32 million polygons. Trying to rotate around that many polygons can be quite painful.

So while voxels may appear to be a key selling point of 3D Coat, I'm sure there's quite a few professional artists that have tried 3D Coat and realized the limitations of voxels and so they've kept their credit card in their wallet.

[ifxs]

Yep, voxels are wonderful. But they can be a problem when you need detail. If you need to make a detailed human figure from voxels then you're gonna be hitting something like 32 million polygons. Trying to rotate around that many polygons can be quite painful.

So while voxels may appear to be a key selling point of 3D Coat, I'm sure there's quite a few professional artists that have tried 3D Coat and realized the limitations of voxels and so they've kept their credit card in their wallet.

I completely agree with your points. We need everything in the voxel room optimized, multithreaded, etc; all merging, converting, displaying/redraw. After PTEX is at least 75% integrated, I put my 100% support behind voxels refining and development. I advocate for 3DC whenever i have a fitting opportunity to do so. That being said, I use tools that allow for me to achieve the results I want with as few hurdles as possible in the workflow to do so. RIght now 3DC does just that for me, and others I know, with its awesome implementation of voxels. Honestly, Andrew really knew what he was doing when he took 3DC heavily in the voxel direction. If another program gets a leg up on 3DC in this area, I foresee most of the beneficial points I bring up to others when advocating 3DC to be moot. I would rather that not be, as I like the general ethic and and community of 3DC, and Andrew is a great maestro on this 3DC journey(IMO).

1/2 of my voxel workflow is creating original voxel sculpts starting in 3DC, the other 1/2 of my workflow in voxels usually involves merging/importing a nearly complete mesh into the vox room, sculpting details on it, then quadrantulating or exporting(for either STL printing or realtime work). During this process, to achieve the details I need, I'm EASILY pushing it past 20 million polys. At 22+million polys 3DC just grinds to a COMPLETE halt/freeze many times, especially when I push it close-to/past 30 million. I need 25+ million to achieve the higher details work I'm doing in other sculpting apps, the kind of detail I NEED FROM ANY SCULPTING APP, ESPECIALLY for HD-res(compositing) map baking.

Yes, I know this is just my perspective, I dont mean to offend with the follow thoughts, as I would too like the sculpt room to be upgraded...AFTER THE VOX ROOM IS REFINED: I could honestly care less if the sculpting room gets any immediate attention, as there are many apps that are free that can perform more sculpting functionality than 3D Coat & other 3d sculpting apps. It's not that I would like the sculpting room to be at least as advanced as the voxel room, but I jsut think there are some major performance issues with the voxel room that are more important right now. If I really wanted to work in polys, honestly I would use ZB/Mudbox/(or even)Blender, not 3D Coat. On the other hand, I purchased 3D Coat because IMO it represents a paradigm change/revolution.

I personally dont think it is a wise choice to delay an immediate performance upgrade in the vox room in exchange for months(or more) of work by Andrew just to get 3D Coat's sculpt room up to par with an app like ZB/MB/ETC. A focus on voxels, optimizing & advancing the voxel workflow ASAP would likely be the best move for the forward looking aspects of 3DC; as it's the aspect of the program that will keep many artists using 3DC, and suggesting it's use to other studios and artists. I would definitely like 3DC to have a sculpt room faster & more robust than ZB/MB, but I'd be hard pressed to believe that 3DC will infringe on much of the ZB/MB market-share if it's just going to copy them while the keep expanding their toolsets(especially with Z-Spheres2+ in ZB).

I'd have a different tone if there was a large team working on 3DC, but pragmatically speaking, 3DC has a "temporary" edge with its voxel capabilities, if Pixologic decides to really expand Z-Spheres there's likely to be little edge at all for 3DC to tout. Please, to all the 3DC developers, make the best of it while you have it (Andrew and others). Make this voxel workflow one that others will not easily be able to match/beat and you just might have 3DC make a MAJOR impact on the market before the other apps catch on to the sheer genius of 3DC and implement voxels or voxel-like technology, which z-Spheres2 is beginning to to IMO, honestly, if ZB would have released a comparable voxel capability greater tahn Z-Spheres 2(II), i dont think i would have bought 3DC, I'm glad I didm, but I'm just being honest here. I would be VEYR surprised if I was alone in saying that.

******************

On the note of developers creating innovative technologies that encroach on voxels for FREE, has anyone see the array sketching just created for blender 2.5? It's getting close to z-Spheres 2, for FREE and opensource, which is good for many studio pipelines!!

Check it out if you already havent done so and are interested!

http://vimeo.com/9084198

above is the link to the video demo of array sketching(nearly z-spheres) in blender, for free and opensource(sorry for the redundancy). You can now use z-spheres like tech, with auto quadrangulation, and then sculpt with poly limits well over 15-20 mill in blender 2.5, complete with painting and normal baking to low res meshes(all in blender)...

Please keep optimizing/building the vox room in 3DC, please. It's only a matter of time, other 3D apps see what 3DC is doing, and have MANY MORE DEVS to work on voxel workflows in their individual apps. 3DC can likely retain the crown and continue its growth with a strong voxel direction, all in my opinion of course.

There is also work that has begun on voxel sculpting dev in blender 2.5 as well

LINK for one of the DEVS beginning work on voxel modeling in blender here

[parel]

Many devs does not mean effective devs. I dont think that we need to worry about the vox "crown" going elsewhere. Andrew is plenty responsive to our needs. I like that he blends in the best tech out there. Maybe it is still not smooth yet, but I will keep supporting Andrew with my business.

[AbnRanger]

Again...Voxel Clay sculpting is nice for what it does, but it still presents a hindrance (extra work) for the most common workflow...importing models and sculpting detail. If that is what one wants to do, then MB or ZB offer a much more efficient, no-nonsense workflow. In MB or ZB, when you're done sculpting, you're done...in Voxels, once you're done sculpting, you're only halfway through the process.

Having the surface tools mirrored to the Sculpt Room (with similar brush speed) would allow users a full range of options and and remove that gap in the workflow.

[Javis]

Having the surface tools mirrored to the Sculpt Room (with similar brush speed) would allow users a full range of options and and remove that gap in the workflow.

I know I mentioned this before, but... New information on this, so I quote you again.

I brought this up with Andrew and also pointed out that thread where you first mentioned it. I truly think this has merit and encourage folks to speak up more about it. I personally agree with your thoughts on this as well... The Sculpt tab needs some love!

Anyway, when I mentioned the transference of the vox surface sculpting tools to the Sculpt tab, Andrew said that while they are different that it would be possible, only after Ptex of course. So... That said, I highly encourage anyone interested in this to contact Andrew on it, or bring it up lots and lots. In fact, I would post this in the 3.2 thread as well as the feature request area so that it has more visibility. If you wouldn't mind... I'd post it, but, I bug Andrew alot already, especially regarding the UV tools, heh.

[philnolan3d]

In MB or ZB, when you're done sculpting, you're done...in Voxels, once you're done sculpting, you're only halfway through the process.

I heard you say this before. With my orc bust it took me about an hour and a half, maybe 2 hours, to retopo the sculpture that I spent a good week detailing in voxels. I'd hardly call that half of the work. Same thing with the other bust I did.

[AbnRanger]

I heard you say this before. With my orc bust it took me about an hour and a half, maybe 2 hours, to retopo the sculpture that I spent a good week detailing in voxels. I'd hardly call that half of the work. Same thing with the other bust I did.

Most of the time a model will be brought in from another program, and will have UV's already done as well. Whether it takes an hour or several hours, it is STILL extra work that is added to the workflow just because of the nature of Voxels, compared to what what you have in ZB or MB. So what benefit you get with Voxels means you currently have a tradeoff...more steps/work required for EACH model you work on. So, it's not just an additional few hours one time only....but EVERYTIME you bring a model in. However, that problem is virtually eliminated if a user has equally good tools in the (Geometry) Sculpt Room. Having to re-model (retopo) and re-UV is a PITA anyway you slice it.

[ghib]

Most of the time a model will be brought in from another program, and will have UV's already done as well. Whether it takes an hour or several.....

A Sculpt mesh subdivides better if it's all quads with even distribution or with good pole placement, an Animation mesh quite often differs vastly and needs triangles in specific places so a decent workflow in MB or ZB might be to import a basic basemesh (not UV'd > Sculpt > retopo anyway. The triangles needed for good deformation in an Animation mesh tends to lead to BAD pinching that is nigh on impossible to get rid of in my opinion.

Retop overcomes this and is something that I see as unavoidable for the best quality work. However... time budget sometimes dictates that you can't always work this way which isn't ideal.

I actually prefer the method of bringing in an extremely simplistic base mesh (no longer than 5 - 10 mins work) Sculpt it up then Retop. This way you separate Form & Function, which have never been a great combination to the artistic mind.

p.s. If you're Uving your base mesh you're a mentalist

[AbnRanger]

A Sculpt mesh subdivides better if it's all quads with even distribution or with good pole placement, an Animation mesh quite often differs vastly and needs triangles in specific places so a decent workflow in MB or ZB might be to import a basic basemesh (not UV'd > Sculpt > retopo anyway. The triangles needed for good deformation in an Animation mesh tends to lead to BAD pinching that is nigh on impossible to get rid of in my opinion.

Retop overcomes this and is something that I see as unavoidable for the best quality work. However... time budget sometimes dictates that you can't always work this way which isn't ideal.

I actually prefer the method of bringing in an extremely simplistic base mesh (no longer than 5 - 10 mins work) Sculpt it up then Retop. This way you separate Form & Function, which have never been a great combination to the artistic mind.

p.s. If you're Uving your base mesh you're a mentalist

You need to have UV's already done BEFORE going into MB or ZB to bake to normal/displacement maps. Often times a studio may take a base model in their library, that has medium-level poly count, and take it into ZB or MB to sculpt in detail. The issue is to not force users to HAVE to go to 1) Voxel 2) Retopo 3) UV 4) Bake Normal/Displacement maps when all you want is to 1) sculpt/detail the model you have and 2) bake the detail to a map.

[ghib]

Do people not bake using arbitrary meshes anymore? As far as I know you can do this in Mudbox but Zbrush is a different matter. Not sure what the latest baking in 3.5 is like as I do all my baking using Xnormal these days as I work in the game industry I haven't had much need to use Displacement maps.

Most major apps allow you to bake between arbitrary meshes these days. There are of course advantages and disadvantages to both methods but doing it the separate way gives you a bit more flexibility, especially if topology needs to change drastically.

If you think about it in this way.. You have a library of stock models that you can take to ZB/MB to sculpt. You can get started straight away. You then BUILD the model (by Retopo) so you're only building the model once anyway. It's not like you're redoing work and this way you're modelling 'over' your form.

Doing things the way you describe isn't necessarily bad.. it's just old skool (I jest)

[Nemoid_]

well in many pipelines you still have modelers and sculptors with ZB , Mudbox or whatever.

In some case artists do both things.

in the first case, its true. they give you a mid poly mesh and you have to sculpt/detail it.

UV map can be provided at start or even later

(don't forget ZB allows artists to paint models even without UVs and add it later and bake maps)

however the most commont pipeline is mid poly mesh with unwrapped UVs, to detail/paint and bake maps. so, you have to be allowed to import model, sculpt /paint and bake.

other processes, like retopo, mesh creation within ZB or other, are great exceptions even if they're adopted more and more when character artist is called to provide the whole work, sometimes from concept to animation ready character.so every artist in this case chooses the proces he's confortable with.

Simply, 3D Coat has to be good for these needs.

[Amber]

Voxel Clay sculpting is a great alternative and avoids some issues ZB and Mudbox users face with Geometry sculpting...IF you start there. However, if you have MB or ZB and you have a low-mid poly model to work with, you're best bet is to use either one of those instead of 3DC...as that workflow is far more efficient and faster than importing a model into 3DC.

QFA

AbnRanger is spot on, Voxels have already steep learing curve until it's possible to seamlessly import/export low-poly uvmapped objects and bake dis maps I'm affraid usage of voxels is quite limited for most people so why bother learning it.

I also agree with what you said 3DC needs some speed improvements brushes and voxels are sometimes way too slow.

I'd like to see new features in retopology and subdivision support there.

... and better manual written in proper English.It's ordeal to read it like it's now, feels like technical book written in akward English.

[Marc Wakefield]

I honestly do not think the Voxel learning curve is in any way steep at all. Far from it...

As far as I am concerned, you can just jump in and start sculpting without any worries, constraints or considerations. It is as easy and natural for me as using a ball of clay.

You can also create a bunch of parts (body parts, props, etc) which you can add to any sculpture as a primitive / merge object which greatly speeds up model creation. In terms of speed and artistic freedom, it is awesome!

Retopo is easy and way quicker than creating a mid-poly model from scratch. UV's are also pretty easy in 3D Coat (with ptex considerably so).

You can bake the high poly texture to the low poly model with local ao and spend a bit of time detailing and painting colour and spec before exporting the whole kit and kaboodle. shimples

The voxel and sculpt rooms could do with being merged / mirrored and the surface tools could do with a bit of optimization. The ability to bake lightmaps would be good and a bit more speed wouldn't go amiss but I am using xp 32 non cuda so upgrading my laptop would fix that. (although I do get 100 fps + quite regularly on my old Dell xps m1710)

IMHO 3D Coat is a fantastic, extremely useful and relevant tool and voxel's are the future so stop whining! If you love mb & zb so much, buy em and go away.

I don't mean to come across as a fanboy but It almost seems like some people on here are making 3D Coat out to be a second rate lame useless donkey and it is so, so far from that.

A lot of work has gone into 3D Coat and it is updated more frequently than some people update their under crackers. It may not be perfect but I am confident it will be soon.

Peace to all!