Designing the design tools of C4D


#1

Matthäus Niedoba, who did a Siggraph talk just today, has written a cool article about the behind the scenes process that goes into putting together some C4D tools. Quite eye-opening - they used workflow flowcharts to compare steps that would be taken by the end user, to figure out the best approach. https://uxdesign.cc/designing-the-design-tools-of-cinema-4d-fab5e26859ae


#2

thanks.

It starts with an unnecessary exaggeration “almost zero learning” also i didn’t liked when he made those 2 images which appear to make a relation with 2D vectors and 3D polygons . But the description of what the modeller wants is quite good, just misses the needs solution to bevel/ fillet without that no model looks real.

Layers made the interface much less elegant, it seems like what was done to 3dsMax OpenSUBD, what about if the meatball object when clicked automatically showed the 2 necessary tabs in attribute manager? The user would need only to change between tabs so i don’t think it was a good example. Nevertheless, in a small thinking time i don’t find a better option for Volume Modeling than layers since it is quite important to show the relationships between parts and they can be a quite a big number, wonder when a volume object have 20-50 parts or more, layers would need collapsing and grouping so i think replicating the tree aesthetics of object manager in layers manager would maybe an idea.

Btw any idea if the volume layers can be copy pasted from object to object? or select a bunch of layers and copy and paste?
Another thing i am eager to test is if the boolean operations have same quality if i merge several objects lets say for a subtraction or one by one.


#3

Thats quite a interesting article. I wonder if this is Maxons new way of thinking, or one that has been in place for some time? I think in any case its always a work in progress as we can only keep adapting to improve based on analysis of the data collected. While reading that article I could not help but take notice of my inner voice shouting out “then why did they add volumes when they have other areas that are in greater need of attention?” My second line of thought is the methods in which symmetry is done for modelling is very counter intuitive for the artist which is why C4D breaks the mould in doing it different from other 3D applications.

Out of all the features of R20 volumes are of most interest to me, and is certainly something that at some point was welcome to have inside C4D as to if this was the right point in time is another thing. It goes without saying that C4D is very easy to use so they certainly got that right, but within that frame work while its nice to see more added, R20 leaves me feeling that future updates are more about adding new tech rather than back tracking to updating the old. Who are the users that they are listening to?

Communication is the key next year as R20 is leaving me feeling rather confused as to their priorities and who their core users are as to what percentage are they giving priority to because C4D plays the field at different markets, Games, Film, VFX, Mograph to mention a few, is my MSA going to continue pleasing mainly the Mograph user base?

I love the application, its ease of use, the enthusiastic vibe of the talented artists such as seen at Siggraph, but Im not seeing much on character rigging, animation advancements, new modeling tools that are in favour of getting better clean topology and edge flow with better tools for re topology, Uvs.

Dan


#4

Regarding prioritization of features, you have to keep in mind that not all features are created equal. Volumes were done by a relatively new developer who was able tro do his work based on very current Cinema 4D base technology. Other features are not this easily developed, they are much much older and come with a lot of technological baggage that has to be taken care of by an experienced core developer, a much scarcer ressource. The new core Maxon is working on for some time now is exactly about replacing those old structures with much more modern concepts, which can be used without the need of intimate core knowledge and the risks that come with modifying the old core. Over time ever more old Cinema 4D functionality will become easier to update, until at some point in time nothing will be left of the old core. We already took many steps on this road, but there are many more to come.


#5

My core is also getting older. As a Bodypaint customer, I’ve seen the years pass me by like seasons. This is the winter of our discontent. At this point I may be retired before the old core is fully modified.


#6

Thank you for your direct response, I appreciate that .

Having some incite into these things can make a huge difference even without the full picture of things. I like what I see with the Volumes, this has certainly caught my attention, and even although I cant comprehend the full significance of fields, I see in the core of its abilities some great innovative features are sure to come to Cinema 4D.

Coming from the perspective of a character designer, and some what generalist in all related areas id like to see some progress in some of the other areas with each release as Im not a big studio, I dont use Cinema 4D for my main income, but also use Cinema 4D as a high end artist tool for my personal work which Im please to be in a position to own back in the day when my financial circumstances where much better. At the same time being a Studio owner in order to have these character tools I have to keep up with higher MSA cost with an extra £100 added this year. I understand its hard to please everyone, but Cinema 4D is aimed and advertised at a very vast array of users, and I feel Im within small minority group.

Dan


#7

I enjoyed that article. Even though it kind of glossed over the technical things (the guy is a designer, not a coder, so I can’t blame him for that).
Writing code is easy. But figuring out the best way to approach and tackle a specific task is where it gets tough for me. And it’s always hard to find information about how people approached a task.
I really appreciate it any time someone takes the time to explain the thought process that went through their head when they made something.
That really does help me a lot in my own personal projects.

-Scott


#8

Has anybody from Maxon thought to make not linear ui for volumes, exactly node based?


#9

I was just thinking the same thing earlier today. Don’t know if nodes would be over-kill… Perhaps layers is the default but one can access a nodes window for more power. Version 22 perhaps…


#10

Yes.
Or clicking on volumes and drop down to graph ui like xpresso tag. Different nodes for manipulation of vdb(clip, add, sdf…), math-image nodes, texture(noises or images) nodes…


#11

In My View R20 has provided definitive clarification of the
users whom Maxon considers their target Demographic.
Motion graphics Architectural &product Visualization
and Engineering and the “3D generalis” who wants an easy to use general 3DCC package.
Not a bad lot for providing a decent revenue stream IMHO.

And you likely will not any time soon IMHO
We character animators have long since moved on the other solutions ( MAYA MOBU even Iclone Pro pipeline)

At this point I am having a hard time imagining Maxon implementing new Character animation tools, in a single release , so advanced that
they would suddenly supplant Autodesk etc. in anyone’s pipeline for complex character work.
For me personally Cinema4D,still remains a good Final gathering/lighting&rendering hub for my Characters Animated/lipsynced/cloth simmed elsewhere and inserted via Pre-animated/point cached meshes.


#12

Rick Barrett suggested, over at the Cafe, that the ground was being prepared for updates for all those things you listed. Not just mograph stuff. Some of what gets updated when seems to be based on technical things stemming from the rewrite, not Maxon being picky or stubborn or ignorant or choosing to ignore non-mograph customers. He didn’t give a timeframe. Bjorn’s comment above is really interesting though. It suggests, as the core is refreshed, the pace of the refreshed features will eventually speed up. They had to clear the big hump first.

More pointedly, if everything is being brought across to the new core, everything will eventually be looked at, perhaps afresh, perhaps reconsidered, improved, renewed. I can’t see them spending several years to get a new core up and running just to then idle along with abstract mograph updates and not much else. They have a cool, powerful new core. I’ll be surprised if they don’t eventually make really good use of it.


#13

Well but what srek says above is that the core is still far from being rebuild if volumes is one of the things that could be done for this release because it does not need old core.
Seems R20 was still very limited in development by the old core. Otherwise they could have put more developers that don’t need to know old core parts.


#14

Seconded. I think the last couple releases have definitively shown Motion Graphics continues to be a priority / that they intend to keep their lead in this area, perhaps at the expense of some more “niche” workflows (niche for them not in a general sense). This is logical IMO. Wherever your bread and butter is, you make sure you don’t fall behind there first, then you work the weak edges of the application piece by piece.

Also agree ArchVis/CAD will continue to get some love but I doubt that every release will show a focus in this area; it isn’t necessary, strictly speaking. More likely the advancements that will impact this area will be tied to the rendering technologies in the app and therefore not strictly ArchVis/CAD in nature. This should leave some room for MAXON to address some of the more glaring weaknesses over the next couple releases.

Hopefully we’ve turned a corner / will see the pace of develepment in areas that traditionally lag behind, speed up. R20 was a pretty substantial update and across a fairly wide area. While I’m learning third party solutions, I’m also rooting for big UV workflow improvements for some of you in R21, along with “rounding out” their new workflow tech — SSS support in nodal materials + whatever else they didn’t have time for on a broad level, and support for said materials in ProRender. That and we can bank on 2-3 modeling tools being added to the new core, getting workflow enhancements every release IMO.

I have a feeling those of use who have put main viewport performance / object management as a priority may end up not seeing one big update but a series of smaller ones like the instances improvement this time around.


#15

Maxon is now fully Nemetschek so it is natural they will push C4D also for architecture and visualization as second market after Mograph. And that market needs human characters (walking, doing things) so they cannot forget character animation.
I still have severe doubts about Prorender viability now with raytracing(and new Nvidias RTX ) the big talk in Siggraph with real time rendering and Unreal increasingly employed for rendering in Viz market. Right now seems a big investment for not a lot of gain.
As long as C4D is open to other plugins that can get that technology.


#16

Sorry, you’re not really know market. Many Allplan and Vectorworks users do not like CineRender(Renderworks) and want maxwell-like(allplan dev team made own path tracer) or vray for vectorworks.


#17

We can already insert human Characters performing everything from simple walk cycles in an Arch vis shots to Full acting performances with cloth sims,Ragdoll and lipsych in animated films ,Via Alembic,MDD or FBX.
Just as we can get fluids from Xparticles or Reaflow etc.

Indeed its bloody convenient to have such features be native and integrated in one environment.
However I personally am not seeing much incentive for Maxon to focus on these areas this" late in the game"
as the people who have invested in these external solutions for CA are the ones most likely to drop out of MSA eventually IMHO.


#18

There are CA coders/TDs who have worked for biggies like Weta, ILM, DreamWorks, Pixar and so on.

There are others who have implemented various advanced CA functionality at university or for their PhD thesis.

These people know how to implement pretty freaking advanced Character Animation functionality in C/C++ code.

Maxon would just need to hire 1 or 2 of these people for a while, and CA in Cinema4D could advance in leaps and bounds in a year or two.

It seems to me though that Nemetschek - a CAD giant - does not see much financial value in advanced CA tools and similar.

This is a great shame as C4D has the right UI/UX design to be used for feature animation, film VFX and so forth.

If C4D had advanced CA, advanced multi-physics and so forth, I’d be an active user right now.

As it stands, C4D has become a software that is aimed at CAD/Archviz/Broadcast only - if you are after a Maya alternative with powerful CA and physics, you’re bound to keep waiting and waiting…


#19

RTX is just a programming interface for allowing use of new hybrid raytracing techniques in DirectX 12 games and so forth.

Its not some kind of “Amazing World Standard For Raytracing” that somehow will beat every other competitor the minute it is released.

AMD are just as good as Nvidia technically.

So the AMD version of realtime raytracing will probably be no worse than Nvidia’s - Nvidia are not the be all and end all of 3D raytracing.

Also remember that Nvidia didn’t invent raytracing or anything - they are just hardware accelerating a very old and very well explored rendering technique it with their new GPUs.

In 5 years from now, there may be solutions form Nvidia, AMD, Intel, Imagination and possibly some others.

They will all run at approximately the same efficiency.


#20

If the new coder was concentrating on mograph stuff because it was easier, it suggests the other coders are concentrating on harder stuff that maybe isn’t mograph. C4D had folks spending time on updating character stuff in the past, no reason it can’t do so again. I also can’t see any time spent improving dynamics, particles, character, pyro, rendering, VFX, paint, take your pick, making them sell less copies.

Any users that have ‘moved on’ might be one group. There are new users each year, all asking the same questions (now on Reddit and social media perhaps more than forums) - which app should I learn? Which app is good for character, VFX etc? I’ve seen a hundred variations of the same answer, ‘C4D is best for mograph, but if you want blah blah blah, you better pick blah blah blah, as C4D is far behind it’.

If that default answer ever shifts to, hey, C4D is a great all rounder, it can do a much improved job in those other areas, I can’t see it hurting things.

Of course this is optimistic speculation, but as recently as this year (or maybe even this month, I didn’t check) the various NAB and Siggraph sessions certainly show a wide variety of techniques and approaches, including character and VFX stuff here and there. Did they spend all that time building motion and object trackers just to have shiny spheres rolling down the street? I dunno.