Andrew Shpagin

V4.1 BETA (experimental 4.1.17D)

4,841 posts in this topic

Revolutionary features are nice, but 3DCoat really just needs to be a solid and refined tool for the areas of the pipeline it covers first. And if the number of unresolved mantis reports are anything to go by its got a long way to go.

Share this post


Link to post
Share on other sites

I'd be happy to see bug fixes, some UI cleanup and overall optimizing of what we have already, before Andrew adds any new major features. 3D-Coat, as it is, is more than enough for one developer to handle. Maybe an additional developer would be needed if he decides to add new functions like rigging.

Without a question. More programmers = better software.

I'm still wondering, how such a small team can manage to maintain code of such complicated software.

mind_blown.gif

Share this post


Link to post
Share on other sites

Yup, whenever I think of it, it makes my brain explode too. :D

Share this post


Link to post
Share on other sites

I'd be happy to see bug fixes, some UI cleanup and overall optimizing of what we have already, before Andrew adds any new major features. 3D-Coat, as it is, is more than enough for one developer to handle. Maybe an additional developer would be needed if he decides to add new functions like rigging.

I agree, the only difficulty there would be that to afford another developer, 3D Coat would either have to increase it's sales, use a different market tactic for increased profits or increase the price to purchase.

 

Out of these, my preference would be on increasing sales and the use of additional alternative marketing strategies.

 

such as : 

 

1. a subscription license alternative : smaller fee to use and it lasts till your paid subscription runs out unless you purchase extensions (this is being done in a lot of software packages now days.. However... most software packages make the mistake of making this the ONLY way to purchase their software which is a deal breaker for me though for a lot of people, a cheaper subscription method might be just what they are looking for..

 

2. Cut some license deals with other software manufacturers like Unity or UDK as 2 examples where by they provide a discount on their licenses to 3D Coat users and in exchange, 3D Coat would provide a custom applink to UDK or Unity to modify the mesh of characters or rig (if/when this is added) on the fly.  By providing direct connectivity like this, it will encourage more people to use the engines that have this support as well as encourage more users of those platforms to choose 3D Coat as their preferred asset creation suite.

 

3. Introduce a Sales Recruitment Discount scheme where by each sale that takes place via an existing member's personal link will push the purchase of 3D Coat down by a fraction of a percent or if they are feeling really generous by a few percent per sale... so if you can reach a lot of people then you can be marketing 3D Coat to others so they buy it before you do and when it comes to you buying it.. you can get it even cheaper so everyone wins.  It could also be used to make the cost of your next upgrade even cheaper as well.

 

4. Hold publicly announced competitions using platforms such as Steam to advertise for you : Steam gets to "give away" 1 3D Coat Pro license to the winner of the competition so that makes Steam look good to its users and they would likely promote this comp heavily as well.  Comp might be to create a replica of a certain 3D Scene in X amount of time documenting a summary of steps via a youtube video as proof that 3D Coat was the only software used exclusively to re-create the scene with a finished upload perhaps as well.

 

Each competition would rely on users taking full advantage of the awesome features 3D Coat offers so this would get even more people to move from bloated over priced software to 3D Coat as well.  Also, if the time-frame and complexity of the scene needed to be created (unless you were quite skilled with 3D Coat already).. would require more than the minimum trial period allows, this would encourage Educational License sales of the product at the cost of 1 copy of the pro license to 3D Coat and a lot of extra advertising for 3D Coat would be taking place as well.

 

These are just some rough ideas I have come up with again on the spot so would likely need some refining to work well but could really help boost sales and thus encourage further, faster development of 3D Coat for everyone.

Edited by Jax_Cavalera

Share this post


Link to post
Share on other sites

The one thing that has really got me excited (not counting the new Autopo coming because that is definitely an awesome thing don't get me wrong) however the thing that has me on the edge of my seat is the prospect that ...

 

Bone Rigging is on the possible features for a future release card set.. Being able to generate bone rigs within 3D Coat would make this a 1 stop shop for game devs, and being able to animate these rigs would be even more cool.

 

I'm picturing an Auto Rigging system where it does the following :

 

You have a box area where you get a curves like tool pretty much the same thing as the curves tool from the voxel room except that you lay down directional curved cylinder shapes.  Then when you have designated the start curve and all the extra curves out from that one curve, you click apply and it turns them into bones.. it's not too different to standard rigging but I think would be a lot faster.

 

You could also toggle for each control node in the curve if it was to be "Curved" or "Straight".. for arms you may want curved and do 1 for the entire arm then it breaks this down into a number of straight bones when you click apply.. or it could simply be all Stiff Straight curves.. but being able to extrude new "bones" out like this would make it really quick.

 

if you could then select groups of curves and create a sphere or box and assign them to it as a grouping.. it could calculate out auto IK based on default Humanoid ik rigging setups.

 

I think the key that would make 3D Coat a smash hit though for rigging would be if you could Toggle IK / FK and the bones would remain in position.. but the controllers used to move the IK bones would turn off.  You could then save keyframes using just  FK bones but position them quickly using IK constraints and the exported file would only contain the rigged structures visible at the time of export so if IK was toggled Off.. your character would only have an FK rig and FK animations to go with it.

 

File export format would be .Blend because it's fully compatible with Blender and Unity (and most indie developers would likely use Unity to get a quick prototype built so they are going to mostly be the ones buying 3D Coat off Steam which makes them a good candidate target)  FBX could be the other file format for those with silver lined pockets and Autodesk products ;)

 

And yeah that to me would make 3D Coat a dream program.. I get a little confused going between Blender and 3D Coat due to them having diff interface controls .. not the end of the world just a matter of less import/export workflows and more of a 1 program for all thing.

 

anyways can't wait for this to come into 3D Coat even if it's not executed the way I described, I mean I only came up with that as I was writing this article so I'm sure with a bit more thought someone could improve on that much more.

 

btw.. what's wrong with the paint room at the moment?  it looks to work good for me.

I've stated before, it's best to get familiar with the tools 3D Coat already has, before requesting wild features it was never intended to have (bone rigging). It IS NOT an animation application. It is an application that handles the creation of content upstream from that stage in the pipeline. And as such, it doesn't need rigging, a full-blown render engine, animation, hair and fur system, etc. It's far better to refine and improve on what is already in the application. For example, true layers for sculpting, where you can blend different sculpt layers on the same object, like you can in Mudbox and ZBrush...including the ability to apply layer masks and dial up it's opacity. Granted, one can do that in the Paint room with depth maps, but it's not possible to do in the Voxel Room. So, if you want to do all your high detail work in LiveClay and sculpt wrinkles on one layer, pores on another, and blend them together interactively, you can't.

 

Since you already have a robust Pose tool in the Voxel Room, with an FFD cage and all, there is no real need for a bone rigging apparatus. You can also pose fairly well in the Tweak Room, using the different gradient selections and the Drag Points tool. I asked Andrew about possibly talking to Fori, the developer of pmG Messiah,  to see if they could enter into some form of partnership, perhaps. He said he might look into it, after he get's Auto-Retopo squared away. Maybe bundling Messiah with a version of 3D Coat (Maybe call it 3D Coat - Animate or FPS, for Full Pipeline Suite). This would satisfy those that keep asking for a rigging system, hair/fur, and a more robust renderer. Tighten the integration between the two, so it's practically seamless going back and forth (via the Applink, which will need some improvement)...have matching navigation and hotkey presets...even tweak the UI of Messiah a bit to look more like 3D Coat's. It's not that far from it, actually. As it, too, has a tabbed UI.

 

Messiah has a nice Interactive Renderer, much like Lightwave's VPR, Modo's, and FPrime....and it works with everything the main renderer does....including GI. It's based on Arnold, which has gotten a lot of attention, lately. What Messiah lacks, 3D Coat provides. And what 3D Coat lacks (in a full application sense), Messiah offers. Could be a perfect match, me thinks. Why? Because, I think Messiah's market has pretty much dried up. No one is looking for a standalone Character Animation application. LW recently improved their CA system with Genoma, and Modo added CA tools in 601. Apart from a partnership like this, I am afraid it will be EOL'ed soon....shame. It's really powerful at what it does, and greatly under-appreciated.

Share this post


Link to post
Share on other sites

Messiah looks good I agree, though if I had the choice between Messiah and iClone.. I would choose iClone hands down every time.  So a partnership between an external 3rd party is a great option, I mean we have this now with Blender and the Applink setup.. the only downside would be that.. well.. Blender has that interface which is so counter intuitive to the average user it really pushes people away.

 

I wouldn't exactly call the ability to rig bones a Wild request, to me it seems more practical to work within the confines of a single application where all the interface controls remain uniform than it is to navigate away into another software package with different support staff, different control setups and a different community / vibe as well as a different compatibility etc..  Right now anyone can do and has to do these tasks in external applications.  Back in the day people couldn't do dynamic tessellation modelling inside of 3D Coat .. over time it's expanded with functionality and has been rewarded with a stronger user-base for this.

 

I think more than anything else, people want to see these features included and THEN refined.  Also I think people DO NOT want a feature that is broken or doesn't perform at all so features like bone rigging aren't something that anyone would really expect half done but at least included.. we would want it so that it is a usable feature for it's purpose and function before it's added to the software package.

 

I agree there are other areas that need work before new features like bone rigging and animation are introduced.  That being said, the Pose tool is not a substitute for the ability to pose a rig with an IK bone rig or even an FK rig.  I know that's pretty obvious based on how limited and irreversible any adjustments are using the pose tool compared to any form of rigged bone infrastructure.  The time it costs to try and modify poses with the pose tool for some form of still frame animation is not something I would be using it for.. kinda like using a spoon to dig out a swimming pool instead of an excavator.

 

The pose tool like every feature does have it's place, and without it.. getting a mesh into the correct pose so it can be rigged up would not be an easy task to achieve but for frequent pose shifts.. you would be better off posing the mesh once then rigging it and then posing with a rig.

 

I'm not really following the importance of needing to blend multiple layers, perhaps I just have not had any real need for this kind of feature.  If I had multiple mesh elements I wanted to create, can't I simply have them as individual objects and mold them so they sit within each other or wrap around each other easily and perform CSG Subtract functions to ensure no mesh clipping occurs between object surfaces.  Is there an example of a situation where you could only achieve the desired result using the blending of multiple layers?

Edited by Jax_Cavalera

Share this post


Link to post
Share on other sites

....I'm not really following the importance of needing to blend multiple layers, perhaps I just have not had any real need for this kind of feature.  If I had multiple mesh elements I wanted to create, can't I simply have them as individual objects and mold them so they sit within each other or wrap around each other easily and perform CSG Subtract functions to ensure no mesh clipping occurs between object surfaces.  Is there an example of a situation where you could only achieve the desired result using the blending of multiple layers?

 

ZBrush Sculpt layers:

http://www.youtube.com/watch?v=nmqyLbVHnlM

 

http://www.youtube.com/watch?v=4Pz5P_VzgdU

 

http://www.youtube.com/watch?v=b7YnsAyDRaA

 

Mudbox Sculpt layers:

http://www.youtube.com/watch?v=BnSZKZjsLKE

Edited by AbnRanger

Share this post


Link to post
Share on other sites

Ahh gotcha so it's basically Zbrush's way of creating layered normals/bumpmaps... Except that instead of using virtual depth it uses 3d mesh layers.

 

I am pretty sure I've seen the ability in 3D Coat to perform CSG subtracts to create the same visual look as achieved... the only problem being that it's not reversable by deleting an object as it's a permanent change it makes.  Yeah I think something like this could be neat to have... is it necessary to create detail .. probably not.  I mean the level of detail you could create with it is unlimited ... as in you could go from just a small scratch mark on a surface.. to adding an entirely new limb on another layer.. but if you were adding a new limb.. why not just use a new object layer?  and if it's just a small scratch.. why not use a normals or bumpmap layer?

Share this post


Link to post
Share on other sites

Ahh gotcha so it's basically Zbrush's way of creating layered normals/bumpmaps... Except that instead of using virtual depth it uses 3d mesh layers.

 

I am pretty sure I've seen the ability in 3D Coat to perform CSG subtracts to create the same visual look as achieved... the only problem being that it's not reversable by deleting an object as it's a permanent change it makes.  Yeah I think something like this could be neat to have... is it necessary to create detail .. probably not.  I mean the level of detail you could create with it is unlimited ... as in you could go from just a small scratch mark on a surface.. to adding an entirely new limb on another layer.. but if you were adding a new limb.. why not just use a new object layer?  and if it's just a small scratch.. why not use a normals or bumpmap layer?

You can sculpt this way, using depth maps in the Paint Room, as I mentioned previously. However, if you WANT this kind of fexibility when sculpting in the Voxel Room, you're out of luck...currently. THIS is one reason why I keep harping on the need for Andrew to tune out requests for things 3D Coat was never designed/intended to do....such as bone rigging. It's out of 3D Coat's element. Would it be cool? Maybe. So might a hair and fur system...but again, this is getting away from 3D Coat's mission. To model/sculpt and texture paint.

1 person likes this

Share this post


Link to post
Share on other sites

I thought originally 3D Coat was primarily a texturing tool and was never originally a modelling / sculpting software package but over time has been developing towards providing that 1-stop-shop for all things 3D.  I've got no problems either with adding in the layered detail feature I was more saying that if you can already achieve this result using existing tools then is this feature more important than adding in brand new functionality that will bring 3D Coat closer to becoming a true 1-stop-shop?

 

Perhaps it would be good to hear from the developers what their current goals are for 3D Coat's future.  To hear if they are wanting to keep it locked in as a 3D model/sculpt and texture software package or if they are wanting to extend it's functionality out beyond this to reach a larger user demographic.  At least then we as the general public of users would know what areas are a waste of time to discuss for future 3D Coat features and which areas are totally on the cards.

Share this post


Link to post
Share on other sites

Without a question. More programmers = better software.

I'm still wondering, how such a small team can manage to maintain code of such complicated software.

 

 

That's actually quite questionable. There's an old expression, "too many cooks in the kitchen spoils the broth".

1 person likes this

Share this post


Link to post
Share on other sites

That's actually quite questionable. There's an old expression, "too many cooks in the kitchen spoils the broth".

Yeah...and Raul is either back or due to be back. So, that will help a good deal.

Share this post


Link to post
Share on other sites

Too many programmers can also = disaster.  MS Office was based on VBA and years ago, I know that you had to use one time and date function in Excel, and a different one to achieve the same thing in Access.  Many examples like this appeared such that it was easy to tell they had different teams working on the programs and it became a pain for one person to know VBA for one and the other.  The point is, consistency becomes harder with more people and can lead to an even more complex situation for bug hunting.  For my part, the only reason I would want Andrew to have someone else on board is to have them as a backup.

Share this post


Link to post
Share on other sites

I think that a poorly managed project will always fall to pieces when the number of active participants increases.  The key to utilizing the power of numbers is good management and infrastructure.  Not doing something for fear of what may happen if it's executed poorly doesn't strike me as the right attitude to have if we do have confidence in Andrew and the staff that are developing 3D Coat.

Share this post


Link to post
Share on other sites

+1 Jax_Cavalera. It is kind of sad, that while blender institution hired new developer and they are all squashing lots of bugs in past weeks, in 3dc there were 0 bug fixes and there will be probably new bugs with intoduction of autoretopo. I expected bit more stability from paid software.

1 person likes this

Share this post


Link to post
Share on other sites

You can sculpt this way, using depth maps in the Paint Room, as I mentioned previously. However, if you WANT this kind of fexibility when sculpting in the Voxel Room, you're out of luck...currently. THIS is one reason why I keep harping on the need for Andrew to tune out requests for things 3D Coat was never designed/intended to do....such as bone rigging. It's out of 3D Coat's element. Would it be cool? Maybe. So might a hair and fur system...but again, this is getting away from 3D Coat's mission. To model/sculpt and texture paint.

Agreed. Adding rigging or other such features won't sell more licenses of 3DC. I think 3DC needs to stick to sculpting/painting. Within that feature set, there is still a lot of room for improvement. One area that 3DC can outshine ZBrush in, is in the UI. One reason I also use Cinema 4D is because it's very (artist) user friendly. Anyone with little technical 3D knowledge can jump into C4D very quickly. 3DC can capture that space in the sculpting/painting app market. 3DC is so close to getting there already.  

Share this post


Link to post
Share on other sites

all remains in the code lines

 

if 3DC design is modular... to add new team developers to expand areas its easy

if not... if the design is not modular and every change in one code line affect all the structure... well... its complicated.

 

we need the sdk.

we need a better angelscript implementation

we need phyton.

Share this post


Link to post
Share on other sites

+1 Jax_Cavalera. It is kind of sad, that while blender institution hired new developer and they are all squashing lots of bugs in past weeks, in 3dc there were 0 bug fixes and there will be probably new bugs with intoduction of autoretopo. I expected bit more stability from paid software.

 

How do you know there were no bugs fixed?

Share this post


Link to post
Share on other sites

Updated to 4.0.07 [bETA!]

Changes:

1) Essentially updated quadrangulation.

- quads only in very most cases

- adaptive density, much better features capturing (like fingers)

- users strokes will create exact edges if tangent post-smoothing is turned off

- hardsurface retopology. But it is a bit experimental.

- good symmetry support. Symmetry improves quality drastically.

- no more spirals. Edgeloops are created in almost all cases.

2) Clone/Cut&Clone problem fixed (reported by philnolan)

To be honest it was very very hard to do all this. Sometimes I felt - why I started this and does it really worth so huge efforts? But dropping at the middle is not for me. I was silent on forums, but it was time of very hard work, during vacation too. I was never working so hard like now even if my twitter was a bit silent. I had a lot of support requests too, so it was far not easy time especially due to I was not able to upload something there for so long time. Also I had personal issues - my 92 years grandma that was living with us died.

The quadrangulation definitely improved greatly, but I still feel no stability. Sometimes it produces very good result but next time on same model - not as good. Still need to work over it. I feel that other areas of 3D-Coat require my attention too. And I understand if someone frustrated due to lack of big progress in paint room. But please understand me too. This session was not easy to me.

3 people like this

Share this post


Link to post
Share on other sites

TY !

 

Pixologic take 2 years to develop a new autoretopology system

 

And they have a team of 25 dev working full time

 

you are only one

 

man... you do a LOT of work... ty for your effort

 

may be... to publish the new v4 SDK will help ?

 

-------------------------------

 

Support for you and your family

1 person likes this

Share this post


Link to post
Share on other sites

Sorry for you grandma, I lost my grandpa this week too.

Share this post


Link to post
Share on other sites

Updated to 4.0.07 [bETA!]Changes:1) Essentially updated quadrangulation.- quads only in very most cases- adaptive density, much better features capturing (like fingers)- users strokes will create exact edges if tangent post-smoothing is turned off- hardsurface retopology. But it is a bit experimental.- good symmetry support. Symmetry improves quality drastically.- no more spirals. Edgeloops are created in almost all cases.2) Clone/Cut&Clone problem fixed (reported by philnolan)To be honest it was very very hard to do all this. Sometimes I felt - why I started this and does it really worth so huge efforts? But dropping at the middle is not for me. I was silent on forums, but it was time of very hard work, during vacation too. I was never working so hard like now even if my twitter was a bit silent. I had a lot of support requests too, so it was far not easy time especially due to I was not able to upload something there for so long time. Also I had personal issues - my 92 years grandma that was living with us died.The quadrangulation definitely improved greatly, but I still feel no stability. Sometimes it produces very good result but next time on same model - not as good. Still need to work over it. I feel that other areas of 3D-Coat require my attention too. And I understand if someone frustrated due to lack of big progress in paint room. But please understand me too. This session was not easy to me.

I'm kinda private person (I don't like talking about personal stuff like that on the Internet as meaning is lost) but my condolences, I lost my last grandparent a few weeks ago too. Take care.

Share this post


Link to post
Share on other sites

Updated to 4.0.07 [bETA!]

Changes:

1) Essentially updated quadrangulation.

- quads only in very most cases

- adaptive density, much better features capturing (like fingers)

- users strokes will create exact edges if tangent post-smoothing is turned off

- hardsurface retopology. But it is a bit experimental.

- good symmetry support. Symmetry improves quality drastically.

- no more spirals. Edgeloops are created in almost all cases.

2) Clone/Cut&Clone problem fixed (reported by philnolan)

To be honest it was very very hard to do all this. Sometimes I felt - why I started this and does it really worth so huge efforts? But dropping at the middle is not for me. I was silent on forums, but it was time of very hard work, during vacation too. I was never working so hard like now even if my twitter was a bit silent. I had a lot of support requests too, so it was far not easy time especially due to I was not able to upload something there for so long time. Also I had personal issues - my 92 years grandma that was living with us died.

The quadrangulation definitely improved greatly, but I still feel no stability. Sometimes it produces very good result but next time on same model - not as good. Still need to work over it. I feel that other areas of 3D-Coat require my attention too. And I understand if someone frustrated due to lack of big progress in paint room. But please understand me too. This session was not easy to me.

Thanks a bunch for the update, Andrew...and all the hard work involved in it. I don't think anyone was trying to rush...probably just thought the images of the progress, on Twitter, meant that a build was imminent. By the way, sorry to hear about your Grandma. Lost two grandparents a few years back, so I understand the loss.

Share this post


Link to post
Share on other sites

We all really appreciate your hard work Andrew. I'm sure it will help a lot of people. Sorry to hear about your Grandma. I lost mine a few years ago myself. I'm out all day today but I'll download it and try it out as soon as I get home.

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.