Jump to content
Sign in to follow this  
Hrvoje

CINEMA 4D R20 OVERVIEW

Recommended Posts

13 hours ago, everfresh said:

not entirely sure of the scenario this might come in handy yet, but i'm sure i'll find a useful application for this. ;) thanks for the info.

can you tell if there will be a fix for PSDs (they don't work in animation as of now without baking everything)?

if that is true do we have volumetric skinning now in C4D? Meaning skin weights independent from mesh?

Share this post


Link to post
Share on other sites

In terms of features this is better than R19. I am glad they added OpenVDB in the way they did. I can see a good workflow from C4D to ZBrush. The fields system looks great and the fact that it can work as a polygon deformer should improve modeling workflows. The material nodes system is nice. I like the fact that added multiple ways to do things in the system and "small" things like create a input node from an output node's port.

 

As far as the UV tools, I know changes are coming and I don't UV via Cinema 4D right now. I look forward when they have better tools in the future.

Share this post


Link to post
Share on other sites
5 hours ago, hvanderwegen said:

...

In the meantime C4D is left with an aging renderer that just can't produce the same level of quality renders as ProRender and other modern renderers yet STILL gets the candy first, and not having material nodes for ProRender is stifling users from properly transitioning to ProRender. Which only solidifies the general feeling that an external render option is going to be preferable even over ProRender in C4D.

 

So weird. Perhaps I am beginning to understand why Newtek decided to rip out the old render engine: in with the new, out with the old. Look forward, don't look back. Sorry for my rant, but I was really looking forward to see ProRender's potential fulfilled in R20, but look at it now. No node-based materials? Instead, the old render engine got them? How does that make sense?

 

 

With all respect, I wouldn't call the Physical renderer aging. It it's not the fastest one, but the quality is very high. And in R20 it got progressive speed improvements.

 

I like using Physical, it's easy to achieve great results in various lighting situations. It was constantly updated in the past releases (Thin Film shader for example). Having it now extended with Nodes is amazing and the implementation seems rock solid from what was shown so far.

 

Yes, it would have been nice to have it for ProRender already in R20 but look at the other massive work that has been done in a short amount of time.

 

As I also have and (occasionally use) LW 2018 (and previous versions) and Cinema 4D R17-18 Broadcast / R19 Studio with Standard, Physical, Vray 3.6 and Cycles 4D, from my experience I'd say that those render engines are far ahead of LWs "new" render engine which is pretty slow, grainy and has several issues like fireflies in many lighting / material situations. I have to admit that I didn't spend much time with it yet, but the UI / workflow for good render settings is not intuitive.

 

It also broke compatibility with several important 3rd party plugins that were never updated. Dropping the old engine (that allowed much more experimenting and tuning) was a mistake of NewTek in my opinion. Even tough it can render VDB volumes it's quite limited in its use. The shading is not great for conductor surfaces, while Physical and Vray produce outstanding results with few settings. C4D is in a completely different league than LW in every possible aspect.

 

About Pro Render, yes it would have been nice to have Nodes already (and in XPresso) but 1) You can bake them 2) I have no doubt MAXON gets that right soon, 3) I guess most C4D users still use Standard / Physical renderer, 4) It's important to have a solid CPU render engine onboard and 5) There are great GPU engines available for C4D with nodal shading, so I don't see ProRender as a must have right now

Share this post


Link to post
Share on other sites
9 hours ago, RBarrett said:

As I said earlier, there were technical reasons for supporting Physical first. This was the best way to get the node editor in your hands as part of R20, so you could start using it and providing us feedback.

I think it may be connected with the fact, that SDK for using new nodes by 3rd party renderers isn't ready yet, and ProRender is 3rd party renderer, no matter how deep it integrated in cinema. 

I agree with @hvanderwegen, it looks wrong and nodes should be in every renderer from beginning. But I will choose good implementation, not the fast and buggy one. Yet, waiting another year for that is sad, hope MAXON can add this in mid cycle update, but don't hold my breath.

I love this update, not because I do fancy mograph stuff, but because I see where is cinema going in transition to new core and what level of features I can expect in the future. 

 

Also I agree with many others that MAXON should be a little bit more open about their plans. I'm sure they watching and listening what people say about new release all over the internet, and, maybe, they will make some statement at siggraph to clarify all questions and doubts of community. We don't ask for complete roadmap with planed features and dates, though it would be nice, but rather acknowledgment of current situation, understanding of community needs and affirmation that it all in the work. 

Share this post


Link to post
Share on other sites
12 hours ago, jwiede said:

Looking at those I can't find a couple really useful ones, so had a few questions about users' capabilities w.r.t. nodal materials:

 

1.  Can users cast our own rays via nodes as needed, both for locality and boundary-checking as well as render-related pathing?

2.  Can users replicate complex material nodes using node networks (handling complex sampling scenarios, etc.)? 

3.  Could a user build an Rounded Edge shader themselves using nodes?

4.  Can users create node networks that generate "internal scattering" SSS-type effects / volumetric sampling?

 

Thanks!  And my apologies if some of these were already asked, still wading through all the replies in the thread.

  1. Single rays can be shot, where and what you do with them is mostly up to you.
    On the left the Context node delivers all current render/shading context data, the Ray node in the middle tells you what kind of ray is currently being processed. On the right side is the Trace Ray node that allows you to cast an own ray.
    Rays.thumb.jpg.ad8fe8dc3ab47d6c4c93e1dc0355a493.jpg
  2. I don't know what kind of sampling scenarios you have in mind. Additonaly to the Trace Ray node you also have Ambient Occlusion for sampling
    AO.thumb.jpg.b60a8e2ec2abd446e84e7a8b9854fb1f.jpg
  3. If i remember correctly there were examples for rounded edge shaders build by testers, however due to licensing restrictions we can't provide any of these.
  4. Current sampling methods only work for surfaces, not volumes, however you can  do some effects that take object thickness into account by casting rays inside of them. This is not a SSS replacement though but might be good for some interesting effects. I used this for X-Ray effects and similar.

I think you will be able to tell if this is interesting for you or not, once the Demo becomes available.

Share this post


Link to post
Share on other sites
7 hours ago, zeden said:

if that is true do we have volumetric skinning now in C4D? Meaning skin weights independent from mesh?

with the vdb mesher we can skin volumetrics, but i'm not sure what you mean by skinning weights independent from mesh. we can skin the weighted mesh  by putting the mesh into a volume builder and that then into a volume mesher and according to hrvoje we can modify weights with fields. and now that i had some time to think about it it could be quite powerful to be able to do that. i'm thinking of a possible super easy way to make some sort of precisely controllable correctional morphs with it. can't wait to dabble with it.

 

all in all i have to say the more infos i gained over the past 2 days r20 looks better and better. the volume mesher can actually produce cleaner results than i initially thought when the right filters are applied, fields is even more powerful than i could have ever imagined, the cad importer seems rock solid and very controllable, and even small things like the new gradient open up new possibilities. so thanks MAXON, you made this summer even hotter ;)

Share this post


Link to post
Share on other sites

After watching a few more videos on MoGraph Fields I now understand that fields have been expanded to effect way more things, in addition to being stackable, adding different ways to be triggered, and grow.  I'd go so far as to call them Triggers instead of Fields.  They are used to trigger changes with time.

 

It also seems the MAXON team did a good job with everything that was added.  It's all clean and well done.  I try to say away from CAD files because they can be so complex, with way too much detail, and a ton of clean up needed.  The many import options I hear with the CAD import hopefully makes it less of a nightmare. 

 

Some things do feel like they are missing.  MoGraph Fields field within a field within a field would be so much neater and easier to follow if the object browser was nodes.  Volumetric modeling would have been cooler if the sculpting tools could be used with it.  I can only guess there is some horrible core in there that does not allow easy copy pasting of one tool to other things.  If it was easy as it should be to copy paste a tool the steady stroke tool from sculpting should be easy to paste into bodypaint.  I'm not sure what is making development so slow, but there has to be something there.  MAXON should also buy the Cycles plugin while keeping Prorender on the way back burner.  If there are legal things in there just throw money at Insydium and say give Cycles plugin away for free.  With the passing of the great Cactus Dan those great character animation tools are not for sale anymore.  It seems like a hole that I had hoped MAXON would fill with R20.  In addition everyone loved the viewport improvements and it seems a shame whoever was working on that was taken off that task.  They were doing a great job.  It makes no sense to stop that.

 

Now let's think back to the days of R10 where the new list looked like follows and included integrating bodypaint into C4D.

http://c4d.cn/help/c4dcn/cn/html/11990.html

This was all after releasing mograph in R9.5.  These were some exciting times.

 

C4D is a great program, but that does not excuse such slow updates for the last few years.  I've been hoping for so many years now that they will go back to the days of R10 and have at least one huge release again.  The founder are still said to give advice, but the past few years of slow C4D updates seems to show the founders gave up years ago.  C4D has a ton of catching up to do and instead of catching up each year they are falling further behind.  It seems like a sinking ship.  Having spent so much time with C4D I wanted it to succeed so bad, but year after year they tell me it's not happening.  There comes a time where people on a boat need to decide when to abandon ship, jump in the water, and swim.  Some might say it is just the current making the ship dip.  The engine is fixed now they say and we can speed to wherever we want.  There is no chance of sinking they say.  It seems like I can barely make out an Adobe iceberg on the horizon.  People assure me it's my imagination, but my feelings are telling me to start swimming.

 

Treading water the swim is not as hard as I thought it would be.  The concepts I learned while using C4D is letting me sail through the water.  Video tutorials are throwing out life jackets.  One of the best things of all is I don't have to hope the ship is not sinking.  In the distance I can see a solid Blender Houdini island that is only growing bigger by the day.  Things will be different on the island.  Some things will be tougher, but hopefully things will be better.  The swim is a long one.  Maybe one day after I'm on the island I'll happily see an old C4D boat wash up on the island.  This would be a good day.  I'd welcome the old friend and make that washed up boat into a little house on my island.  I hope that day comes.  For now I swim.

 

Such a corny analogy.  I wanted to gather my thoughts and it ended up like this.  I was sad writing this.  I can't help seeing what I see though.

 

In reality I'll probably keep C4D for another year and update next year if they have a good deal with a great update.  I am spending serious time with Blender and Houdini.  Blender and Houdini are great tools to know even if C4D doesn't sink.  The physics, character, retopo, and UV tools are great in Houdini.  Modeling is not so bad so far either.  Not sure if I even need Blender, but I do like my free Cycles and Eevee is great.  Haven't learned how to fully recreate all MoGraph situations in Houdini, but I think it's possible.

 

Last, to show a little of what I talk about, here is what Houdini 16.5 added so people can compare to the C4D R20 update.

 

Share this post


Link to post
Share on other sites
  • Topic Author
  • Here is a quickie of fields in action with bevel deformer via selection tags

     

    Bevel_fields.mp4


    Free C4D trainings from MAXON professional on our youtube channel

     

    https://www.youtube.com/c4dcafe

    Share this post


    Link to post
    Share on other sites

    Regarding the node system and radeon pro render. I think MAXON made the right choice here. AR and physical account for >50% of the engines used by c4d users, pro render doesn't even hit half a percent, making it work with the least used render engine c4d has first would have been a huge mistake (and before you say it, no, a node editor for pro render is pretty low on the list of reasons why you might not use it, pro render is still heavily being developed, it will take some more time before it is good enough to be considered the defacto c4d render engine.)

     

    The people working on c4d's pro render integration were already working flat out to get the motion blur, sss and out of core memory system done for R20, there simply wasn't any scope for them to switch the entire material system in a single release on top of the other changes. Finally, the best route is to make one single unified SDK for all third party engines to use so they can make use of c4d's node interface. That part won't be ready for some time, but when it is, pro render will likely use it just as all the other third party render engines would.

     

    Nodes for pro render will come in time, but first they have bigger fish to fry.

    Share this post


    Link to post
    Share on other sites
    Guest mbyrne

    Can someone shed some more light on the direction in fields? It says "direction for certain effectors" - what does that mean?

     

    After watching few videos I am happy with release. Yes too bad about BP being dead in water but I like that they are going "all in" on Nodal. Talked to Houdini guy here and he says that some nodes in screenshots don't make sense for materials at all and are better suited for general purpose? Why is that in there?

     

    @RBarrett @HSrdelic@srek

     

    MAXON peeps, are we looking at 1st gen node stuff with more to come? Why can't we get a straight answer? I don't mind waiting but would love to know that my investment is not heading into Lightwave direction. Is C4D going nodal?

     

    Share this post


    Link to post
    Share on other sites
  • Topic Author
  • 3 minutes ago, mbyrne said:

    Can someone shed some more light on the direction in fields? It says "direction for certain effectors" - what does that mean?

     

    That is an interesting one. Basically you can set clone directions, or "whey they are pointing too" with fields :)

    You can now have nice direction distribution etc. For example, you can make fish school way easier now or similar effects

    6 minutes ago, mbyrne said:

    MAXON peeps, are we looking at 1st gen node stuff with more to come? Why can't we get a straight answer? I don't mind waiting but would love to know that my investment is not heading into Lightwave direction. Is C4D going nodal?

    Well, node materials is completely new development and in reality, first major thing provided by new core. I don't know the answer to your question, since I don't make decisions, but as far as I know, new core is certainly capable of supporting complete node based system of any kind.

     


    Free C4D trainings from MAXON professional on our youtube channel

     

    https://www.youtube.com/c4dcafe

    Share this post


    Link to post
    Share on other sites

    Okay....I have been watching a few more videos on R20 and it is beginning to sink in that this is a pretty significant release.  Also, you need to look at R20 as follows: the sum of the parts is so much greater than the whole.  If we only think about fields, material nodes, volume modeling in the traditional sense, then you are understating what those features mean to R20 because their implementation by MAXON throughout the entire architecture of the program  just seems pretty brilliant and well thought out.  If this is what the new core gets us in the future for other features, then I need to learn a greater level of patience - which will be hard but I am confident it will be worth the wait.

     

    I have also been thinking about the delays in Bodypaint.  UV management is at the core of the program as it touches modeling, texturing and rendering.   As such, I honestly think that it will be last on the release pipeline not because it is low priority by MAXON but because it has to be with respect to all the other features they need to implement.  If new features are going to have the same level of integration that we are seeing in R20 due to the new core, then it just makes sense for Bodypaint to be close to the last thing to be implemented because UV editing touches so many other areas.  Think of it this way: do you put up the curtains in a home while the building is still being framed?  Same idea.  There is a logical order to how things need to get done and MAXON is definitely logical.   

     

    It is a no win situation for MAXON given their current culture because they can't explain why the most anticipated feature update is going to be at the tail end of their implementation pipeline unless they divulge what is coming before it.     They also can't say it is a high priority --- because even though it may be a high priority in their development workload it will still need to be implemented after most everything else they have planned.  So statements about priority will seem pretty hollow as we are kept waiting.

     

    That is just the way it is.  Thus the eternal tap dance on the part of the MAXON employees.  They want to scream out that it will be worth the wait but even statements like those beg more questions on when? and why? and how come not now? because to answer any of those questions gets into the order in which features and capabilities need to be released and that is an area MAXON will not go.  So they stay quiet. 

     

    Maybe that will change with the new CEO and we can only hope. 

     

    Dave

     

     

    Share this post


    Link to post
    Share on other sites

    Join the conversation

    You can post now and register later. If you have an account, sign in now to post with your account.

    Guest
    Reply to this topic...

    ×   Pasted as rich text.   Paste as plain text instead

      Only 75 emoji are allowed.

    ×   Your link has been automatically embedded.   Display as a link instead

    ×   Your previous content has been restored.   Clear editor

    ×   You cannot paste images directly. Upload or insert images from URL.

    Sign in to follow this  

    • Recently Browsing   0 members

      No registered users viewing this page.

    ×
    ×
    • Create New...