Farsthary update channel

[Carlosan]

Hi Raul !

 

modifier stack = a must

 

Could you please take a look about this proposal ? ty

 

To add more primitives or surface objects with a non destructive workflow

 

to create polysurfaces: let draw strokes "on empty canvas", not limited to "over a volume"

[AbnRanger]

How about starting with these? They were requested 2yrs, have plenty of +1's in support of them, and we're still waiting. Would be a great start.

 

http://3d-coat.com/forum/index.php?showtopic=14525

 

I put up a number of feature request on Mantis a few months ago, that I'd like to ask for your support on...if you feel they would help speed up/enhance your workflow in the Retopo Room, as well. If you haven't already signed up to Mantis (for Bug Reporting/Feature Requests), please do so....regardless. It's the best way to keep a systematic record/log of them and Andrew uses them to prioritize. If you just place the bug report/feature request on the forums, the percentage of probability they will get addressed is much, much lower (obviously harder to remember and keep track of) than placing them on Mantis.

 

At any rate, please add a note with your "+1" and check the desired level of support for the feature at the bottom of the page. When Andrew sees there are a lot of users onboard with a request, he places a higher priority on it. Thanks everyone.

 

Bridge Selected Edges:

http://3d-coat.com/m...iew.php?id=1056

 

Shell Modifier/Tool

http://3d-coat.com/m...iew.php?id=1030

 

Hotkey Edge Loop/Ring Selection & Dot Loop/Ring

http://3d-coat.com/m...iew.php?id=1058

 

Soft/Gradient Selection

http://3d-coat.com/m...iew.php?id=1060

 

Hold SHIFT to extrude edges on the fly, while using TRANSFORM TOOL

http://3d-coat.com/m...iew.php?id=1057

 

On this last one, the EXTRUDE tool is not usable in most cases cause it goes haywire in areas where the the scale of the object is smaller than the one selected. Plus the SHIFT+Drag would also work simultaneously scaling or rotating. Very handy for building geometry around a corner (of a pipe or tube, for example), or closing a cavity with quad geometry (something CAPS tool doesn't do).

[AbnRanger]

The SHIFT key option to extrude on the fly (with the Transform tool active) is perhaps one of the easiest to do, but would be a HUGE benefit to the average user. You could do a LOT of modeling with just that small addition, alone. One major thing missing is, when you hit the ESC key to drop the Transform tool, it drops your selection. If you want to keep that selection, you have to manually make it all over again. There NEEDS to be a RESELECT (last) button for this purpose.

[AbnRanger]

You can see him use the SHIFT key to extrude on the fly, in this video. It works not only with MOVE, but when rotating or scaling.

 

[farsthary]

LOL thanks I will try to come up with the best workflow suited for 3DCoat but will try to take into account the best of the competence   Anyway it will not happen overnight because we have more urgent matters like making rock solid what we have, the rest will be relatively easy. 3DCoat exels at million poly models so low poly modeling will be blazing fast assuming we can somehow reuse the same experience and data structures there    but I'm very confident it will be really good.

Subdivision surfaces is another major feature that should follow and if depends on me I will use Industry standards for that, there's no point on reinvent the wheel

[farsthary]

@AbnRanger

I have the Bridged Selected Edges basically done, so will be easy o add, in fact it is in the foundation of the new RetopoCloseHoles tool I've shown recently.

 

regarding shell we do have a nice shell tool (I myself implemented the surface shell  and the volumetric with non overlapping features but are you referring to Retopo shell? (It can be added easily too :P)

[AbnRanger]

@AbnRanger

I have the Bridged Selected Edges basically done, so will be easy o add, in fact it is in the foundation of the new RetopoCloseHoles tool I've shown recently.

 

regarding shell we do have a nice shell tool (I myself implemented the surface shell  and the volumetric with non overlapping features but are you referring to Retopo shell? (It can be added easily too :P)

Yes, I was referring to a need for a Shell modifier in the Retopo room. Would be very helpful to work on all kinds of thin objects. Thanks. 

[philnolan3d]

It excels at high poly counts in the sculpt room. Not so much in the other rooms. Loading models into the paint room with even a few hundred polys can cause trouble.

 

LOL thanks I will try to come up with the best workflow suited for 3DCoat but will try to take into account the best of the competence   Anyway it will not happen overnight because we have more urgent matters like making rock solid what we have, the rest will be relatively easy. 3DCoat exels at million poly models so low poly modeling will be blazing fast assuming we can somehow reuse the same experience and data structures there    but I'm very confident it will be really good.

Subdivision surfaces is another major feature that should follow and if depends on me I will use Industry standards for that, there's no point on reinvent the wheel

 

[TimmyZDesign]

Subdivision surfaces is another major feature that should follow and if depends on me I will use Industry standards for that, there's no point on reinvent the wheel

Yes, OpenSubdiv is the new industry standard method for subdivision surfaces now. Maya, 3DS Max, Modo, and many others have adopted it. The main advantage of OpenSubdiv is speed. It would be nice to make OpenSubdiv models in 3D-Coat, and know they will subdivide and crease the exact same way in other industry standard software.

Anyways, so much good news coming from you recently! Great to hear about the Retopo Room and other improvements coming. Thanks again!

[geo_n]

One thing that is crucial is the basic foundation of polymodelling. What is it?

 

Selection and deselection.

 

Probably lightwave got it right the first time and modo copied it, 3dmax copied it with brush mode selection and some other appz now have the same method.

 

When selecting polygons you just swoosh your mouse with lmb over some polys and it select them. Release lmb and swoosh your mouse over the same polys and they get deselected.

Rmb has lasso mode selection with the same behavior. Lasso select over polys to select polys inside it. Repeat the lasso selection and it deselects polys inside it.

 

And If nothing is selected everything is selected. Meaning when doing s,m,r (scale,move,rotate) everything is affected if nothing is selected.

And there's a button to deselect all at once. In my case I mapped it to "q" for all appz I use.

 

Its funny how max really sucks for selecting polys way back with only window selection and click click click poly select method. Its terrible. The foundation and navigation was not good and hard for tablet users.

 

Swoosh your mouse with lmb to select and deselect polys is also good for tablet users. I was tap tap tapping with tablet in max to select polys and hated it before and would always feel good going back to lw or modo.

 

Edited June 15, 2015 by geo_n

[AbnRanger]

Yes, OpenSubdiv is the new industry standard method for subdivision surfaces now. Maya, 3DS Max, Modo, and many others have adopted it. The main advantage of OpenSubdiv is speed. It would be nice to make OpenSubdiv models in 3D-Coat, and know they will subdivide and crease the exact same way in other industry standard software.

Anyways, so much good news coming from you recently! Great to hear about the Retopo Room and other improvements coming. Thanks again!

For OpenSubdiv to be most effectively employed, in 3D Coat, it would be best to merge the Tweak Room with the Retopo Room, and call it the POLY EDIT room or something to that effect. That way you could do all your low poly editing/modeling/retopo in one workspace. We need to remove/reduce the gaps between the rooms as much as possible...so that it's easy to get assets from one room to the other.

 

Need to add a mouth socket to your low poly model in the Paint room, or just clean up the mesh, where you've noticed a small tear/opening here or there...it's way too much trouble, currently, to do that in 3D Coat. You have to import the mesh into the Retopo Room > make your edits > Export outside of 3D Coat > go to the Paint Room > FILE Menu > IMPORT > REPLACE GEOMETRY. That's the long way around for what should be a simple process.

 

There is a quick & handy way to get that mesh into the Retopo Room (Retopo Menu > Use Visible Paint Object as Retopo), but no way to easily get it back into the Paint Room. Not without having to EXPORT out of the app and then IMPORT it back in. That's the ugly way of having to do that. No user should have to send an asset, that already exists in the app, OUTSIDE the app, and then have to IMPORT it back in. It's about as silly as going out your front door and then entering your kitchen through the back door...just to make a fat sammich!!! You shouldn't have to leave the building just to go into another room....but that's the way 3D Coat has been doing it...mostly, for years.

 

I talked to Andrew about this multiple times, and he agreed....but nothing has ever been done about it.

[AbnRanger]

In the interim, what needs to happen with a mesh one brought from the Paint Room to edit (and they want to get it back into the Paint Room, without having to bake everything all over again and start from scratch), is to provide an option to REPLACE GEOMETRY in Paint Room > select from a list of Paint objects currently in the Paint Room and have a second box for the user to select the Retopo Mesh Layers they want to use to replace them with.

 

That, or better yet....just expose Paint Room Objects to the tools in the Retopo Room, directly, so there is no sending meshes back and forth

[AbnRanger]

....maybe add an Edit Mesh mode (button at the bottom of the Tool Panel?) for Paint Room Meshes, where 3D Coat temporarily locks the Retopo mesh layers/objects, while exposing the Paint Room meshes to the Retopo Room tools. This would be similar to how 3D Coat currently exposes the Paint Room tools to objects in the Sculpt Room. The Sculpt room just "borrows" the tools, so to speak. By doing this and merging the Tweak Room tools with the Retopo Room, you could subdivide Paint room meshes...which would make the Tweak tools work MUCH more effectively. Right now, performance in the Tweak Room is dreadfully slow.

 

If OpenSubdiv is used as the Subdivision algorithm in the Retopo Room, it could make the Tweak tools for sculpting MUCH more lively. OpenSubdiv has GPU acceleration code (including CUDA and OpenCL) ready made

[ghib]

Please use NVil as inspiration for poly modelling, or better yet just hire Kun (Istonia) to make it for you.  He's amazing and about as prolific as Andrew when it comes to iterating. http://voidworld.cmcproductions.co.uk/index.php

[Javis]

No please not modo as inspiration for modelling. Too clicky, no mod stack, not flexible. This is from modo 301 to 801 user who is still waiting for modo to catch up to max.

 

It's not that clicky at all. And besides, don't expect a mod or history stack in 3DC. That's a fat request if there ever was one!

[philnolan3d]

Yeah, I was thinking a history attack may require a lot of major programming.

[gbball]

Please use NVil as inspiration for poly modelling, or better yet just hire Kun (Istonia) to make it for you.  He's amazing and about as prolific as Andrew when it comes to iterating. http://voidworld.cmcproductions.co.uk/index.php

 

Agreed,

 

I would love to see Kun join the team.

[ajz3d]

It's not that clicky at all. And besides, don't expect a mod or history stack in 3DC. That's a fat request if there ever was one!

Why? Implementing modifier stacks doesn't sound too complicated.

After all, what is a stack (in the meaning of a geometry modifier stack)? It's a certain state of geometry. You can store it as .obj or some other more comfortable and faster-to-access (custom) format inside the .3b file, or hell, even as a hidden retopoGroup (as a hack). Each [n] item will operate on [n-1] item as it was the original geometry, and [n] item could be stored inside .3b file in the same format. So [n+1] (another level on the stack) would operate on [n] item. Etc, etc.

True, we don't know if 3D-Coat's code allows for this without major modifications, but if it's well written, and I'm sure it is, this shouldn't be a problem I think.

[ajz3d]

Agreed,

 

I would love to see Kun join the team.

3D-Coat+Nvil =

[Javis]

Sure, hire a few more developers and it might not be too trivial. But with the current resources, I doubt it's realistic.

[ajz3d]

Sure, hire a few more developers and it might not be too trivial. But with the current resources, I doubt it's realistic.

We only wish 3D-Coat the best, Javis.

[gbball]

Sure, hire a few more developers and it might not be too trivial. But with the current resources, I doubt it's realistic.

 

You never know what could be arranged.  I think Nvil is still relatively obscure I think.  I don't know that the developer is even working on it full time.  It complements 3d Coat beautifully though. 

 

Like you said, I'm not sure that it's possible since the codebase is c# rather than c++.  But Alias and Wavefront merged years ago to help make Maya an industry leader, so I figure anything is possible.

 

Anyways, this is probably not the place to discuss it, since it's Raul's thread.

 

All that being said.  I think 3d coat has an opportunity to do something better than standard polymesh modeling.  Something that would gracefully degrade to regular polygonal modeling, but within 3d coat could be so much more.  I would love to work with the team to implement my idea which I've alluded to before.

 

Perhaps it's something we can connect on this summer Raul as I will finally have some time come July.

[Javis]

We only wish 3D-Coat the best, Javis.

 

 

Absolutely. I'm just saying don't get your hopes up too high. It would be cool, just not very realistic.

You never know what could be arranged.  I think Nvil is still relatively obscure I think.  I don't know that the developer is even working on it full time.  It complements 3d Coat beautifully though. 

 

Like you said, I'm not sure that it's possible since the codebase is c# rather than c++.  But Alias and Wavefront merged years ago to help make Maya an industry leader, so I figure anything is possible.

 

Anyways, this is probably not the place to discuss it, since it's Raul's thread.

 

All that being said.  I think 3d coat has an opportunity to do something better than standard polymesh modeling.  Something that would gracefully degrade to regular polygonal modeling, but within 3d coat could be so much more.  I would love to work with the team to implement my idea which I've alluded to before.

 

Perhaps it's something we can connect on this summer Raul as I will finally have some time come July.

 

 

Maybe so! Good things come in small packages. Or so I hear.

[geo_n]

It's not that clicky at all. And besides, don't expect a mod or history stack in 3DC. That's a fat request if there ever was one!

 

I didn't say mod stack for 3dc.

I said it for modo for not having non-destructive workflow after all these years and with the backing of the foundry and all its resources...nada. Meshfusion getting there but with its meshy mesh output, no thanks.

 

Andrew is doing well on his own but I don't see a single programmer able to pull off non-destructive workflow.

 

Anyway I'm ok with 3dc copying the fundamentals(selection/deselect, lasso, falloffs, action center) of lightwave modeller which modo then copied. it is simple and straightforward.

[AbnRanger]

The Retopo Room is already pretty close to being a full-fledged poly-modeler, and I think Andrew tasked Raul to work on the modeling tools...as he has already begun to. I don't see the logic in hiring an outside developer to do his assigned task for him. Let's be patient and see what Raul has in store. We might end up with a much improved Auto-Retopo or quad-mesh generator, to boot, as I thought it was on the "To Do" list. I like the current Transform Gizmo, but it needs (to add) the ability to move and scale along 2 axis' simultaneously...like most 3D apps have. It's a must, IMHO, because the user will need to do that frequently. Soft/Gradient Selection is also an absolute MUST for any poly modeling app. as it will be needed frequently, as well.

 

As for the prospect of a history stack.....Andrew already started adding History to the Primitives tool (needs some refinement, IMHO, but it's a start), and he said it could lead to history stacks in other tools (Pose Tool next?). Like Geo_n said, selection efficiency in 3D Coat needs to be top notch (ie., selection sets drop list rather than saving to a file). That will take it pretty far. But before we start unloading all of our lists of requests...how about Priority No.1 being absolute TOOL CONSISTENCY, throughout the app? Allow me provide a few examples. Selecting loops for UV seam selection already exists. So, why is there not an identical method for standard edge-loop selection? I don't get that inconsistency. You are using edges to make a seam selection (when in UV mode). It makes no sense, whatsoever, to leave that functionality out of the SELECT tool. You have a SELECT PATH and UV PATH tool that are identical. Speaking of the Select Path tool....once you make a selection, if you want to RELAX that selected path....you have to switch back to the SELECT tool to do it. Why?

 

This highlights a long-standing problem with 3D Coat, most new features are rushed and sometimes left unfinished. Myself and others have been asking Andrew for a full-fledged procedural noise generator for 3D Coat (both in the Sculpt and Paint Room). He starts off with the Noise tool, but never adds but one noise type and then leaves it untouched to move on to PBR or other tasks. I even gave him links...even the files themselves, to a free open-source library of procedural noise maps (Bercon Maps)...but Andrew won't bother adding them. I don't understand it. This is one of the most powerful tools in ZBrush, and 3D Coat users have been deprived of similar functionality for years...even though it's been requested numerous times. Why make a Noise generator at all, if you aren't going to supply a host of noise types? Why have just a handful of Procedural noise patterns in the Fill tool, when you add many more patterns (via free libraries of those patterns).

 

There are just loads of inconsistencies and unfinished tools (many with not so much as an icon or even a simple tool-tip) like this throughout 3D Coat, that require users to request and request and request and request over and over and over...for years to get them corrected. This really needs to be addressed. When working on a tool that has a counterpart in other workspaces, GREAT care should be taken to ensure that they are 100% identical in functionality and features. The Transform gizmos have been all over the place in inconsistency...and remains so, with the transform gizmos in the Tweak Room. Nothing like the Gizmos elsewhere in the app. What does that tell new users? That Pilgway can't be bothered making their tools consistent across the entire app? They can spend almost a year on PBR, but cannot waste their time making like tools match? It's frustrating, from a user's perspective, because if nothing else, it only steepens the learning curve and introduces confusion.

 

Another inconsistency that I've been begging Andrew to fix is FreeForm Lattices in the Sculpt Room. With the Primitives tool active, you can select a model from the models pallet and set an arbitrary number of control points on the lattice. Not so with the FFD lattices in the Pose tool. You are limited to just a handful of presets. Consistency = efficiency + easier learning curve.

[Carlosan]

Low poly extrude and scale prototype

Posted on June 17, 2015 by Farsthary

This is quick alpha test for low poly modeling tools in 3DCoat.

Still there’s a lot of room for improvement from shader to UI, also in selection since there’s currently no selection mechanism rather than freeze values. Is something must be reworked in the future.

 

[Michaelgdrs]

COMMMEEEEEE oOOOOONNNNNN 

[AbnRanger]

Raul, is there any way to develop a proper Beveling tool in the Sculpt Room. That EXTRUDE tool looks to be very close, but bevels aren't necessarily extruded. Using the Pose tool and E-Panel Splines does work to some degree, but not always. On irregular shapes, it can be really difficult to make it work. That's why a dedicated bevel tool would be very welcome.

[Denis]

Raul, is there any way to develop a proper Beveling tool in the Sculpt Room. That EXTRUDE tool looks to be very close, but bevels aren't necessarily extruded. Using the Pose tool and E-Panel Splines does work to some degree, but not always. On irregular shapes, it can be really difficult to make it work. That's why a dedicated bevel tool would be very welcome.

+ 1

[TimmyZDesign]

Raul, is there any way to develop a proper Beveling tool in the Sculpt Room. That EXTRUDE tool looks to be very close, but bevels aren't necessarily extruded. Using the Pose tool and E-Panel Splines does work to some degree, but not always. On irregular shapes, it can be really difficult to make it work. That's why a dedicated bevel tool would be very welcome.

+1