Jump to content

R25 Expectations


Guest Igor

Recommended Posts

On 7/6/2021 at 7:42 PM, Matches said:

While on the subject of symmetry, symmetrical painting in the very neglected BodyPaint3d would be nice, as well as some general updates and baking options, to make it more user friendly and modern to be competitive with Substance and Mari.

 

I also developed an entire system for symmetrical painting that works using a Sculpting style workflow on top of the existing Body Paint layers system. I gave this away for free for a while too but have recently pulled it all down. Nobody really wants it (although it had over 3000 downloads when I made it pay what you want, but only earned about $7 in donations). Everyone has moved onto other applications. The main issue that many people face now is that they are using 3rd Party renderers in C4D. And you can't paint live on a texture and have it update in that renderer. People using Mari and Substance are disconnected enough that this does not bother them. But inside C4D people expect, or want to, see the final result while they are painting. So no one really wants to setup Standard Materials in C4D just so they can texture paint anymore. And in a world of procedural materials and textures I really don't know many people that even paint anymore at all. If you know people who want these tools then let me know.

 

You could paint using Stamps, Stencils with Symmetry right in the Viewport. These tools have been around since R17.

 

https://www.plugins4d.com/Product/PaintBrushes

 

I am curious to know what people actually want from BodyPaint. It sounds like it is just a new look to make it seem fresh. Internally it is a solid system. But these days do people want to paint on textures or do they want to paint masks like what Substance does? 

 

How complex are the models people want to paint? Mari is dedicated and can handle huge models and a large number and large sized, textures. Which is why it has a custom caching system. Can you, or should you, even attempt to compete with Mari? A dedicated tool for the job will always win for painting complex models in a studio setup. And with the subscription model now they don’t even sell BodyPaint3D as a separate application, so even harder to convince a studio to purchase a Swiss Army knife tool like C4D when all they want is a dedicated painting tool.

 

 

Link to comment
On 7/6/2021 at 9:42 AM, Matches said:

Even though I'm a fan of C4D's symmetry generator, symmetrical modeling would be a welcome addition and a very useful option.

 

I had recently programmed a plugin for symmetrical modelling (within the boundaries that the API currently offers), but it sold a mere four times. The project told me two things: 1. The current API makes it fairly difficult to put a symmetry module on top of the core system, and 2. While people are often calling for symmetry in modeling, the actual need is very, very low.

 

I just guess the symmetry object is the top of what people actually require, and anything else would be a spritz of sweet cream on top of it. I understand why Maxon isn't in a hurry to implement that.

Link to comment
53 minutes ago, Cairyn said:

 

I had recently programmed a plugin for symmetrical modelling (within the boundaries that the API currently offers), but it sold a mere four times. The project told me two things: 1. The current API makes it fairly difficult to put a symmetry module on top of the core system, and 2. While people are often calling for symmetry in modeling, the actual need is very, very low.

 

I just guess the symmetry object is the top of what people actually require, and anything else would be a spritz of sweet cream on top of it. I understand why Maxon isn't in a hurry to implement that.

The basic functionality that C4D is missing is the ability to have part of a model symmetrical and a part not. For instance when making a head you could like to have the mouth asymmetrical while the rest stays symmetric. It's about workflow. Now we are in a situation where you can have a fully symmetrical model with the symmetry object. But once you make it editable to be able to make an asymmetrical change you loose the ability to still change the parts that you want to keep symmetrical. It's how symmetry works in the sculpting system. It's something you can turn on or off depending on your need. It shouldn't even require similar topology on both sides but make the symmetric changes based on the distance from the plane of symmetry.

 

This is something that I'm sure many people would find very useful, which is why it's returning for over ten years now in these feature wish lists.

Link to comment
35 minutes ago, janheylen said:

The basic functionality that C4D is missing is the ability to have part of a model symmetrical and a part not. For instance when making a head you could like to have the mouth asymmetrical while the rest stays symmetric. It's about workflow. Now we are in a situation where you can have a fully symmetrical model with the symmetry object. But once you make it editable to be able to make an asymmetrical change you loose the ability to still change the parts that you want to keep symmetrical. It's how symmetry works in the sculpting system. It's something you can turn on or off depending on your need. It shouldn't even require similar topology on both sides but make the symmetric changes based on the distance from the plane of symmetry.

 

This is something that I'm sure many people would find very useful, which is why it's returning for over ten years now in these feature wish lists.

 

Yes, that's something I was missing too. CollieSymmetryHelper indeed allows for partial symmetry - you can exclude points from being mirrored.

 

However, I went for a point-wise symmetry which demands a model where the symmetric parts are symmetric to begin with (to create the relationship between "right" and "left" points), and which makes it difficult to add or remove points, as the mapping gets destroyed by such an operation. (C4D used to have internal messages that allowed to repair this mapping, but sadly the new modeling core is no longer doing that - I discussed with C4D developers already.) That's about the best you can do with the current API and some limited logic.

 

Symmetry independent of topology is a very difficult topic indeed. If you do not anchor a point to a "symmetric partner" (a point on the other side, or a virtual position made up from tri surfaces or point groups), you can only map it dynamically to the "other side's" surface. This invokes the risk that this surface is ambiguous, especially in models with complex manifolds. Not to speak of disruptions of that topology, like holes. May lead to frustration in the user if the moved point suddenly causes broken topology on the symmetric side 😿

 

Most likely (I didn't spend too much time on the thought) you can solve this issue if you split the symmetry algorithm into a "map surface" part before the change, and a "deform surface symmetrically" part after. This would give you the additional information what position on a virtual surface with non-identical topology a point would map to, and allow you to apply point movements to the symmetrical surface in a meaningful way. (You'd still need to clarify what the radius of influence is, e.g. if the polygon size / resolution on one side is vastly different from the other.)

 

I guess the sculpting symmetry doesn't care for topology because the symmetry tools are applied symmetrically (guessing here), instead of restoring the symmetry after the change. This is a completely different approach, which can only work if all the tools are adapted to it.

Link to comment

If I may, these threads always follow the theme of C4D staying in the forefront of the technology on pretty much EVERYTHING.  We want it to be the best or at least rated in the top 3 of every possible element of digital content creation: modeling, sculpting, painting, texturing, rendering, EFX, dynamics...the list just goes on and on.

 

And for what Maxon charges, that is not an unfair request.  But history has shown that Maxon can NOT keep C4D current in every area. Maxon knows it so they have started to acquire or license other technology:  Redshift and Ministry of Flats to name a few.  Or, they have developers spin off and create outstanding plugins like X-Particles.

 

One question though: Why do we keep thinking that every aspect of DCC software needs to be developed in house?  Far more critical and larger applications than C4D are not 100% developed in-house.  Whole sections of networking IOS  software are outsourced so why can't Maxon follow that model as well?  The jewel in Maxon's crown should be their architecture, the UX design standards and quality requirements.  And then rather than have people writing the code, they have people writing the standards that contract software developers need to follow.  This should be happening already but honestly it doesn't feel that way given how certain aspects of the program are aging so badly.  This may also be why Blender is so successful.  Everything is outsourced. They have a whole community of developers working to grow that program. 

 

If this is NOT happening already or is against Maxon's culture or goes against their standards on safe guarding their intellectual property, then what I would like to see is a greater and deeper partnership with these third party plugin developers such that their plugins follow the same internal logical consistency as the rest of C4D.  Everything works together so you barely even know you are crossing from C4D domain into a plugin domain.  Load the plugin and the tools  that go with that plugin just appear in the standard tool pallet along with the native C4D tools.  All the plugins work as additional modifiers that can be stacked in the Object Manager rather than have their own separate interface.  Some do that already, but to be a Maxon certified plugin (do they still even have that designation?) you need to follow these standard UI rules.

 

Also, and this a big one, there is one single NODE editor for anything requiring a nodal interface: Xpresso, material nodes, scene nodes, Redshift nodes,  Plus tags that allow you to mix and match shaders, lights, cameras, UV's, weight maps, etc created by various 3rd party developers with the render engine of your choice.  Everything is interoperable and exchangeable.  The core C4D application becomes the tonic water in your software cocktail.  All the separate modules just flow together into one well designed cohesive application.  Maybe the new core and scene nodes are the first step to make this happen...but honestly, I can't see it going that way.

 

Finally, coupled with all this there is a huge price drop for the main application.  If we have to buy plugins to round out the application then you can't be charging us full price. 

 

Honestly, that is why I hate subscriptions.  You have to pay yearly just to use a program that history has shown can't keep current with the rapid pace of technological change in all areas.  For some, the wait for those changes will be longer than others but they keep paying.  Something's got to give...either how Maxon manages their development or how they manage their upgrade policy...but somethings got to change because Blender really does look more attractive with each release.

 

Dave

 

 

Sorry...but I simply do not have enough faith to be an atheist.

Link to comment
8 hours ago, kbar said:

 

Could you explain the "Surface" option and how that applies to sculpting. I might not be getting exactly what it is you are after. Since the sculpting brushes are surface based. They don't just take what is under the brush preview. It looks at the connected polygons on the surface of the model and if they are with the radius of the brush then the vertices are affected. There is also the "Surface Distance" option on the grab brush which works a little different as well, it walks the surface of the connected edges and calculates the amount of distance travelled from the centre hit point. 

 

Also if you want a duplicate layers option I added it to my own sculpting brush tools if you need it: https://www.plugins4d.com/Product/SculptProjectionBrush

 

I have asked people a few times what they would want from the Sculpting system as far as additions go. But I don't hear anything at all other than wanting auto retopology and meshing tools. Which aren't actually related to the sculpting tools.

 

And since R23 there is a new remeshing algorithm built in called "Remesh". Which got improved further in S24 with more options.

 

Thanks kbar, I checked out what you mentioned regarding the "Surface Distance", and that is indeed what I was looking for. I must have missed this, since this is option is only available on the "Grab" brush and not the other ones for some reason. Regardless I'm glad to know about this function and I'm grateful you pointed it out .

 

Concerning what you mentioned about the sculpting layers, while I appreciate that you've expanded C4D's sculpting toolset, I still don't think it answers why C4D itself doesn't ship with these features, especially since the sculpting module was introduced way back in R14 and has remained virtually untouched since then.

I believe that remeshing is an expected feature for modern digital sculpting, and really should be included in any self respecting software that wants to stay competitive and advertises a fully functional sculpting workflow, as opposed to an addendum toolset to modeling.

 

I've experimented with the "Remesh" and found the results to be very unimpressive, especially compared to the auto-topo solutions from Zbrush, 3dcoat, mudbox, the Exoside plugin, and even blender. My point is not to bash Maxon's efforts, but rather to point out that they've allowed themselves to fall way behind the trend, and if they want to catch up and stay competitive then they've got to step it up, especially considering that they introduced sculpting almost ten years ago.
 

8 hours ago, kbar said:

 

I also developed an entire system for symmetrical painting that works using a Sculpting style workflow on top of the existing Body Paint layers system. I gave this away for free for a while too but have recently pulled it all down. Nobody really wants it (although it had over 3000 downloads when I made it pay what you want, but only earned about $7 in donations). Everyone has moved onto other applications. The main issue that many people face now is that they are using 3rd Party renderers in C4D. And you can't paint live on a texture and have it update in that renderer. People using Mari and Substance are disconnected enough that this does not bother them. But inside C4D people expect, or want to, see the final result while they are painting. So no one really wants to setup Standard Materials in C4D just so they can texture paint anymore. And in a world of procedural materials and textures I really don't know many people that even paint anymore at all. If you know people who want these tools then let me know.

 

You could paint using Stamps, Stencils with Symmetry right in the Viewport. These tools have been around since R17.

 

https://www.plugins4d.com/Product/PaintBrushes

 

I am curious to know what people actually want from BodyPaint. It sounds like it is just a new look to make it seem fresh. Internally it is a solid system. But these days do people want to paint on textures or do they want to paint masks like what Substance does? 

 

How complex are the models people want to paint? Mari is dedicated and can handle huge models and a large number and large sized, textures. Which is why it has a custom caching system. Can you, or should you, even attempt to compete with Mari? A dedicated tool for the job will always win for painting complex models in a studio setup. And with the subscription model now they don’t even sell BodyPaint3D as a separate application, so even harder to convince a studio to purchase a Swiss Army knife tool like C4D when all they want is a dedicated painting tool.

 

 

 

As far as Bodypaint is concerned, I respectfully reject the notion that C4D users are disinterested in Bodypaint and it's future development. Your experience with your "Symmetry" addition and it's dramatic change in downloads ratio, should tell you that Bodypaint users are not disinterested, but rather unwilling to pay any extra to gain basic features that are standard in other packages and should come with no additional fee.

Also, the idea that Bodypaint was abandoned because users expect live feedback confuses me. Many years ago, Bodypaint was a prominent staple in professional pipelines and had a strong foothold in the industry; Why would it be traded for other programs that are identically deficient?

The fact is that Bodypaint has been neglected for years, and I remember 3dKiwi ranting about this years ago and still nothing's changed even up until now. In fact, his Bodypaint tutorial course from about (over) ten years ago is probably still completely accurate.

Substance is a relatively new software, but I remember even a few years ago, when hand-painted textures were a bigger market, 3dCoat was much more prolific than Bodypaint, and offered a much more feature-rich solution for painting. Even today 3dCoat continues to stay competitive with the integration of a BPR workflow, smart materials, and masks etc.

Blender has also become a much better solution for painting and general workflow, is constantly being updated and developed, and for many is the Swiss Army knife tool that a lot of studios are adopting for the majority of their work. And is also free.

 

I want to reiterate that my intention is not to "hate" on Maxon, but rather to voice my criticism and hope that they decide to up their game. Competition is what motivates the industry, and those program devs that are content to stay in their respective lanes or get too full of themselves, will fall by the wayside to other packages that are eager to dominate and become number one; Zbrush, Houdini, and Blender are good examples of upward trending, highly competitive programs.
 

Link to comment

Hi

 

Since Maxon is upgrading animation tools and timeline capabilities in the previous versions i really would like to see a better animation layer system.

The Motion System in c4d has a huge problem that makes it unusable for me. Since its based on a Motions System Tag, it requires the hierarchy to remain unaltered through the animation. This is a big problem for riggers that need to constantly add stuff. 

 

Also a system for painting vertex weights with layers in order to be less destructive.

 

Also how about some new sculpting features? Now that maxon owns a sculpting tool how about sharing some of those nice features and better performance.

 

cheers

Link to comment
1 hour ago, Matches said:

I've experimented with the "Remesh" and found the results to be very unimpressive, especially compared to the auto-topo solutions from Zbrush, 3dcoat, mudbox, the Exoside plugin, and even blender.

The topology of C4D's Remesher is pretty bad. I can recommend Exoside's Quadmesher plugin. I use it mostly to remesh Volume and Boolean models and for this it gives pretty good results. I'm happy to have a good plugin that provides this functionality but I feel it's an essential feature that Maxon should provide. Sculpting, Volumes and Booleans are three artist friendly ways of generating models and they all need autoretopo to live up to their potential.

 

Could an autoretopo algorithm for Volume meshes not be trained using machine learning?

Start with a database of models with perfect quad topology.

Remesh the models using the Volume Builder and Mesher.

Train a neural network by using the remeshed models as input and comparing the output with the original models.

Shouldn't this be possible?

 

Link to comment
2 minutes ago, janheylen said:

Could an autoretopo algorithm for Volume meshes not be trained using machine learning?

Start with a database of models with perfect quad topology.

Remesh the models using the Volume Builder and Mesher.

Train a neural network by using the remeshed models as input and comparing the output with the original models.

Shouldn't this be possible?

 

There basically two possibilities because I 100% guarantee you that everybody is thinking about AI stuff at the moment that either:

a) It works but needs more time in the oven

b) Doesn't work with current NN tech for whatever reason

Visit my Portfolio on Artstation.com & Follow me on Instagram

Link to comment
6 minutes ago, DasFrodo said:

There basically two possibilities because I 100% guarantee you that everybody is thinking about AI stuff at the moment that either:

a) It works but needs more time in the oven

b) Doesn't work with current NN tech for whatever reason

A general purpose retopo algorithm might be still too hard. No doubt you are right that many companies working in 3D are trying to solve this.

But an algorithm built specifically for Volume Meshes seems like a much smaller problem to solve. The topology that's generated by the Volume mesher is clean and uniform. The topology of a good quad model we would like as output can also be based on some general rules. 

Link to comment
15 minutes ago, hikarubr said:

I disagree. The fact is that symmetry is a basic feature that should be in C4D. It's so basic (every other general 3D app has proper symmetry) that people are not willing to pay for it. It's like the Magic Solo plugin from Nitroman - way better than Maxon implementation and a lot of people used before there was an alternative from Maxon. but the plugin is free - I doubt people would pay much for this. Nitroman usually charges for his other plugins that are more complex. His simple plugins are all free.

 

I keep hearing that line (Matches above said essentially the same) and people may actually think that way, but what is the result? "Maxon should do that, I will scoff at anything else" is not an argument at all. There is no connection between external plugin programmers (for any kind of functionality) and Maxon, so Maxon is not affected by any buying decision for third party tools. If I do not buy, say, a new layered motion system (as jbatista mentioned; I could use that as well 🦊) because I expect it from Maxon, will it be available in a regular release any sooner for my refusal to use an external product? Hardly. It will be available whenever Maxon thinks fit. Maybe never. (Maybe tomorrow. Who knows?)

 

What will be the result for the third party programmer? They will either withdraw from the market (if they intended to make plugins their livelihood), or they will simply keep their plugins for themselves as competitive advantage (if they earn money from project work and their plugins are actually just a by-product). In either case they will not bother publishing further plugins. This is a deadly behavior for the plugin market.

 

(If the users are lucky, the programmer will be successful with other stuff and put out the unsuccessful plugins for free as advertising for their more costly product. But that requires a large inventory of plugins and an overall healthy income, which in turn requires a sound plugin market. I seem to remember several plugins disappearing with the recent C4D core change, and several plugin developers just throwing the towel, so how sound is the plugin market?)

 

Mind you, I am totally with you: Symmetry for box modeling should be in C4D as a basic feature. I paid quite a bit of money for Cinema over the years, yet every year I look at removed features, missing updates, and lagging functionality. (Don't get me started about fur...) Nevertheless, what is one supposed to do? Either buy a replacement, or find a workaround, or live with the deficits, or program it yourself, or change the main application for greener pastures.

Link to comment

As far as basic competitive feature sets go, I hope R25 will finally have a decent market competitive built-in renderer. They started work on implementing ProRender, got rid of that, and now users are left hanging without a decent modern fast GPU render solution replacement.

 

It is rather surprising that C4D is the only DCC that requires its users to invest in a third-party/external render solution to gain access to a modern up-to-date (fast) renderer.  (well, there *is* Centileo, which is free.)

 

Maxon should really just integrate Redshift in this upcoming version (based on history, management may decide to do exactly this, but increase the base package rental fee 😜 )

Link to comment
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • LATEST ACTIVITIES

    1. 0

      Nodes Modifier result isn't updated anymore

    2. 220

      Scene Nodes | Capsules file pit

    3. 3

      Flipped clones in multi-segments curve

    4. 2

      How to create continuous UV texture for irregular wall like shapes ?

    5. 52

      Looking forward to the next C4D release

×
×
  • Create New...

Copyright Core 4D © 2023 Powered by Invision Community