So, C4D… where next?


I will believe in Maxon again when:
1-Redshift gets ported to AMD GPU’s and Metal.
2-Redshift gets integrated in C4D as the default render engine FREE OF CHARGE


Oh and please, switch the viewport colors back to their pre R21 state as default.
Modelling has become a NIGHTMARE. I can barely see the edges and ngon lines and now need to add a dark material on my mesh. More clicks for nothing!
Who the hell at Maxon thought this would be an improvement?

You brought the offline help back. Thanks a million times.
Now please do the same with the original viewport color scheme.


I’d like to see Redshift included free of charge as well.

  1. You know the Metal port is being worked on mostly by Apple as it happens. I don’t think they’ve got much enthusiasm for supporting AMD on the PC.

  2. Maxon/Nemetchek paid a lot of money for Redshift, they want their return. ProRender is the free renderer in C4D.

Redshift’s USP was speed, technology is advancing very quickly especially with AI noise filtering is now robbing Redshift of its advantage. After speed Redshift is a very ordinary renderer. Arnold on 32 and now 64 core CPUs will give you a similar realtime performance as Redshift but at a significantly better quality. The next-gen GPUs will bring GPU render times down such that the difference between Redshift and other renderers like ProRender and Cycles is negligible.

Next-Gen AMD GPUs could be pushing 17-20 TFLOPs with hardware ray tracing which means ProRender becomes relevant. It’s free…

Redshift is a GPU Only renderer which may prove to be its biggest Achilles heel. Arnold, ProRender and Cycles can use the CPU too.


Meanwhile, redshift is a functional Production-ready renderer.
Pro render is not.
And physicial is almost unusable as soon as you use a PBR workflow unless you have a 16-core machine minimum.


I’d use Arnold every single day over Redshift. The quality of image and flexibility of the node system in Arnold aren’t just in a different ballpark they’re in a different league. Arnold works out much cheaper to use if you take advantage of their short term rentals.

Redshift always looks like a game engine without masses of work on materials. The VFX industry learned that hardware is cheap so just use unbiased renderers. Redshift have promised features they’ve not been able to deliver for two years, they talk a good game but their delivery on promises is weak. If they give a demo at the nVidia GTC gig this year it’ll be the 3rd straight year they’ll have promised Toon Rendering and random walk SSS and I still won’t be holding my breath. Development has slowed to a crawl since the Maxon acquisition.

Physical always has been a joke so you won’t have any argument from me. Maxon can’t keep anyone long enough to maintain it.

I’m going to defend ProRender 2.0 because I have used it in production with Intel’s denoiser and the render times and quality were absolutely fine. PR2.0 has come a long way from that massive disappointment that Maxon shipped in C4D. I expect ProRender to come of age with AMD’s new GPUs and ray tracing hardware.

Of all the renderers it’s the only renderer that can do rasterised with ray tracing, biased rendering and full pass tracing. This flexibility will make this renderer a very useful production tool.


Very interesting. It’s actually the first time that I hear someone using PR in production. I thought the engine was just ok for still images, not animation.


Im not following your logic. how next gen GPUs or Nvidia denoiser are robbing Redshift of its advantage if Redshift can also use them to speed up its rendering power? Also, if I spend 4K in a 64 cpu core render, yes I expect it to be really fast, but how many high end video cards can 4K buy me?


Redshift barely got any speedup when they used RTX because they had a very capable ray tracing system in CUDA.

ProRender does not yet use RTX and there’s no AMD ray tracing hardware to use. ProRender will probably get a significant boost from RT hardware, Cycles and E-Cycles have also benefitted greatly from RTX with nVidia and the OpenCL version will likely get a huge boost too when AMD release the new GPUs.

These other renderers are getting much bigger gains than Redshift from RTX as so are closing the gap. E-Cycles in my experience is only marginally slower than Redshift. Redshift will get faster on new hardware too but overall the gap has closed even more to the point the difference is not worth worrying about.

Had we stayed using C4D we would be using Arnold now. But we would also be using a renderfarm so I would rather buy a 32c or 64c CPU for test renders/lighting etc rather than expensive GPUs. The CPUs would also be used in the rest of the content creation pipeline so they’d get used on multiple tasks whereas highend GPUs basically have one task only, rendering.

Even if you had 4k worth of nVidia GPUs you’d still need more to do real world projects. I personally favour having enough render power locally for previews/lighting and then send to a reliable renderfarm.

You can put 4k worth of GPUs in a Redshift renderbox and it’ll still look like gash compared to Arnold no matter how fast Redshift renders.


Thank you for the detailed answer.



OMG thanks!!!:):):):):):slight_smile:
I don’t want to sound like the grumpy guy but why did Maxon change the color?


Yeah, the new color is unworkable… They should revert that asap…


Oh and please, switch the viewport colors back to their pre R21 state as default.

my eyes are protesting.


Most interesting thread. Been around 3D a long time ( and I have to say I’ve seen this same story unfold many times before. Most recently with Newtek, e-on and Trimble (owner of SketchUp). A couple of years ago, I used SketchUp for creating an historically accurate Alamo for a large scale commercial AR project. It was only barely up to the task.

I had used Lightwave as well-- and years earlier e-on software. All companies ended up vectoring in the same direction Maxon appears to be headed. SketchUp is now a joke and limited by it’s architecture–same is true with Lightwave (limited by architecture). The same 3D die hards are the last in their respective forums proclaiming it’s virtues. The rest of us have left.

As I looked around, someone mentioned Blender. I, like many of you, thought to myself, “never would I use an open source solution-- especially NOT Blender.” Then I got turned on to masterxeon1001 and his awesome hard surface modeling workflow. I called a buddy of mine who works in Hollywood and he said HardOps was only available in Blender.

So, with great reluctance, I decided to approach Blender with an open mind-- knowing learning any complicated 3D software would take patience and time.

I was fortunate to make friends with masterxeon1001 aka Jerry Perkins, and he helped me get started-- and frankly I haven’t looked back. My Blender experience has been great. I typically use the beta products and they are remarkably robust. I’m currently working on a AAA model with over 2m tris with over twenty 4K PBR texture maps and it’s very smooth.

The damn thing hardly ever-- and I mean HARDLY EVER crashes. As much as I thought I hated working with non-supported OS software-- Blender has changed my mind. Heck, they literally have a new build EVERY DAY. In fact I have many more problems with Substance Painter than I have with Blender.

I know Blender isn’t for everyone–heck I thought for sure it wasn’t for me. But, I would challenge any of you to give it a try. The non-destructive modifier workflow is just amazing. Yes, just like every other 3D app out there, it’s not 100% intuitive (tried C4D and had just as much trouble figuring it out-- it’s just what you’re used to and/or have taken the time to learn).

My 2 cents…


I have the feeling that they have to change much more under the hood than they thought. There are a lot of things that don’t work as expected in R21. I tried to use the UV-gizmo different times and always got weird problems. Sometimes it didn’t show up, sometimes it did. Picking UV polygons with the transformtool selected in the viewport didn’t work anymore and stuff like that.
All things that are confirmed as “bugs”.
But the whole openGL viewport seems buggy to me. Gets black sometimes. Selecting in the HUD is sometimes possible, sometimes not.
The fact that rendering in the viewport makes everything totally blurry first (when you happen to use the interface scaling in combination with a highres-monitor in Windows 10) is another thing.
Sometimes I wonder, if these things will be solved ever.


I’m sure they’ll fix those bugs.

Maxon and Red Giant finished their merger today, Check the lineup.

Stu Maschwitz is now Chief Creative Officer, quite a role for a guy who has spent years working in VFX, doing FXPHD classes on DSLR shooting, doing blogs on different aspects of After Effects live action compositing, and writing a book on indie action movie filmmaking. I’d now be really surprised if Cinema 4D remains ‘just’ focused on mograph stuff in the coming years.


The management structure has grown like Adobe bloatware, there’s more heads than Blender has full time developers! Hasn’t Maxon heard the idiom, too many cooks spoil the broth?

I’m sure the customer base will be most happy knowing their Subscriptions pay for these managers before a single line of code is written.


Hard not to disagree. I don’t care who is nominally in charge, I care about the release schedule and future technologies. More senior management does not, necessarily, a better product make. I hate to keep bringing up the ‘B’ word, but it’s really fascinating (and quite exciting) seeing Blender’s new features being discussed and developed ahead of time. (Besides, I’ve moved on from Adobe After Effects, so at the moment Red Giant is of limited interest to me.)

I hope good things are coming, but I’ll believe it when I see it.


Infograph. I knew you’d find something to whine about. You surely won’t have a kind word to offer about R22 and R23 when they come out either.