Hi Steve,

Thanks for using Google and the image. I had to experiment a little to reproduce it.

The color engine in After Effects has no idea what it get sometimes, you need to set up the Color Space for each project, and based on what Ae version you have. When that space is defined, Ae will give you the colors you expect.
.

My suspicion is you missed a step.
https://projectfiles.maxon.net/Cineversity_Forum_Support/2025_CLIPS_DRS/20250820_Wrong_Colorspace_assigned.mp4
Pixels are code for a specific color space, including Gamma.
The ACES specification states that only ACES 2065 (AP-0) needs a colorspace flag set. The ACEScg, which has a smaller space (AP-1), does not have such a suggestion for a flag. ACEScg is supposed to be used internally, never for exchange, hence the idea that the artist knows what they are doing.
So, for years my request was not fulfilled by Adobe to make sure that it recognizes ACEScg or ACES 2065, which leads, obviously, even today to problems.

My best guess is that the image you opened first was left without specifying it as an ACEScg file, and the pasted rendering was converted.

Here is a little video, and I hope it illustrates this problem.

Many people (?) have Photoshop running on sRGB, based on sRGB or perhaps P3 screen, the limitations might not show up easily. Another guess is, that calibrated screens are not the standard.

You might find in the video that I left my screen for this in Low Dynamic, to record it accordingly. We live in an era where more and more HDR is requested, and in the past 15 years, I have advocated here working in linear as it was clear in 2012 with the ACES release that we would get there. Low Dynamic with sRGB will not hold the water when shown close to HDR, like most of the advertisements "breaks" have that Low/ High sandwich contrast, surely if not now then soon.

I mention that, as not diving into Color Management is not an option any longer. I'm surely biased, as an Alumni of the "Colour Academy in Hollywood for professional Colorists", and countless other Color Gradings trainings for around 18 years by now. However, I remember how tough it was to get any information on "where Gamma is stored" while asking in my network in the '90s for help. It took forever to clear that point, yes that was years before YouTube, and web was dial in. It has become more complex these days, but there is a lot of good material available to remedy that.

I hope that allows to reconsider, and dive in. You will feel how much better things go when this theme turns into muscle memory.

Enjoy

Hi kariomart,

What pretty much every Gerator does is generate all new content newly, and so does the Connect Generator. With that all things get pulled into a new object here. Without the Correction Deformer, that setup wouldn't work as I wanted. Without the complete setup, most questions are more general, and harder to answer, as they need to work "in general", meaning including all possible variations, with a more or less complex interaction of several things that is difficult to estimate.

The Correction Deformer can work with just the one Field moving along the "cube-zipper", but to ensure that the influence will not reach too deep, the other Field protects the other "half" of the plane. It is optional.

That you found a solution is great.

My best wishes for your project.

Hi kariomat,

Please have a look here:
CV4_2024_drs_24_MGcd_01.c4d

There is now a second Spline as Rail, which is pretty much a Target function.
I had to replace the Spline, as it seemed like something was off with them.

You could also use a Matrix Object to Deform, and then use the Matrix as Mode> Object

All the best

Hi kariomart.

N-Gons are great for modeling; after that, they should go.

With the Remesh Generator, we have a great and fast way to produce lovely meshes for a wider variety of objects. Is that always working? It depends. Trial and error from time to time, sure.

Anyway, please have a look here.
CV4_2024_drs_24_RSdp_11.c4d

Compare the results when rendering the scene. The scale will change if you move the time slider; frame 30 is perhaps a good starting point. I think the object on the right will clearly show the minimum quality. It's the same material and the exact same RS Object Tag. Just the underlying polygon setup differs.

All the best

Hi kariomart,

No, it should not be needed at all. It should work. Please have a look below.

I have filled out a report and hope it will improve.

I had to stop last night at midnight, as we had our first meeting at 7 am (L.A. time). Now, I try to catch up after a busy, long day.

I explored your file a little bit more. So, here is my suggestion: it might render longer, but it needs a higher Sample radius for this file.

Try "eight" as the start value and measure in the Picture Viewer (CMD+Right mouse down) with the Alpha in the view.

Screen Shot 2023-12-12 at 7.16.06 PM.png

That should work; if not, "ten" seems stable. I have not tested the whole file; I have only a laptop here as the fastest machine.

Sorry for all the waiting time.

My best wishes for your project

Hi kariomart.

Typically, any Dynamic system, to my knowledge, is based on the idea that Dynamics take over, except you have forces enabled, then it mixes the data.

While not having a file here, I would suggest that you eventually cache the dynamic solution you have, then use a Current State to Object.

With a non-dynamic copy of your object, use the Pose Morph and drag the "Current State to Object" results inside. Relative should work fine.

Please note that the Pose Morph can't replicate the complexity of the movement from start to finish; it will be a straight line between the "same" points.

CV4_2024_drs_23_CLtc_01.c4d

All the best

Hi kariomart,

Thank you very much for the kind feedback. You're very welcome!

Going by your scene description, Action, Static, Action, I explored the setup from a different "angle.
To showcase that, I created two similar scenes. The one on the right is different. I explored the Scene Time here. Instead of sucking all the energy out of the simulation, I just dropped the time for a moment. Then, when the wind starts to blow, it is set back to 1.0

Please have a look here.
CV4_2034_drs_23_SIts_01.c4d

Screen Shot 2023-11-14 at 7.47.44 PM.jpg

In the F-Curve Manager, you can fine-tune the settling of the spheres.

Let me know what you think about it, as I have never used it this way in the relative "young" Simualtion-Scene options!

Please cache the scenes to see it working without the delays. Around frame 100-150, the one on the left shows movement, the one on the right is still.

Cheers

All the best

Hi kariomart,

Thanks for the file. Which shows up here as a 2023 file.
The Connector can't be alone; it must be a Simulation object.

Anyway, I would suggest caching Cloth and using the Surface Deformer here.
https://stcineversityprod02.blob.core.windows.net/$web/Cineversity_Forum_Support/2023_PROJECTS_DRS/20231114_CV4_2024_drs_23_CLtsb_01.c4d.zip

Often, a Constraint is suggested for Buttons, but I have not found a stable setup with Simulation with it. I.e. I assume there is something, but I need to explore it more.

Let me know if that works.

Here is an example done in 2024
https://stcineversityprod02.blob.core.windows.net/$web/Cineversity_Forum_Support/2023_Clips_DRS/20231115_2024_Button_01.mp4

All the best

If this works for now as a workaround, Greg, then let me say thank you for being so flexible.

I hope this is fixed soon and you get what you expect.

My best wishes for your project.

You're very welcome,

Speed set the Voxel size for the time of exploration as large as possible. Keep in mind that between 10 cm and 0.1 cm, the difference in amount of voxels is around 100^3 or 1,000,000. This is super simplified, but keep the values for the size large (Low amount!) and go only smaller (Higher!) when needed.

Perhaps download Cinebench and google for results to see how your setup works comparatively. An activity Monitor or Task Manager allows for some exploration if the system reaches a limit.
Switch off all apps and functions that eat into your performance. Google is typically one app that asks for a lot of performance. Enabling redshift only when needed is a current circulating tip.

Please remember that tutorials can be recorded and sped up, then the commentary is placed on top. The tutorial should be slower, as the screen recording is active at the same time.

I do not want to go deeper here, as this forum is not about hardware. But perhaps I have to suggest at one point to have an Ask The Trainer show about performance optimizations of hardware stuff.

However, as a friend told me three decades ago when he got his first NeXT system: Whenever something gets faster, we put more options in our software and have it a little bit slowed down in return. I feel that is not so much different these days. Of course, the primary trend is that everything gets faster (Silicon and code), and expectations might run hot at one point.

All the best

Excellent, thank you, miss-bear, for the reply!

There are often many ways to get somewhere; finding the one you like is the key.

My best wishes for your project.

Cheers

Hi miss-bear,

The point I made above is that you need to have the same settings. Taking overshoots out was not meant as a rule.

In the initial example, the Key settings were different for the second Joint than the first. That might work sometimes, but more often, it doesn't.

The first Joint does something that the second joint need to do while countering the movement of the first, as it is a child of the first. Any slight derivation from the settings will lead to different results.

When I select your first example, Keyframe 75 from Joint and Joint.1, the fields show Mixed or [ - ] in many areas. There should only be one Mixed, and that is the Key value. These two values should be like "positive two" and the other "negative one", in ratio to each other.

All the best

Screen Shot 2023-04-15 at 1.25.39 PM.jpg

Hi miss-bear,

Thanks for the kind feedback.

CAD files often have quite a bag of problems if visualization is needed. They are all created with different targets. Based on their core modeling technique, most of the higher-end CAD systems run mathematically based. From that data, polygon-based models are retrieved. Here the settings are the key to getting either bad or better models.

What I am trying to say, there is a wide variety of CAD systems and an even wider variety of polygon results. This makes it so hard to say anything in general.

My typical advice is to get in contact with the person who creates the CAD data in the first place and try to make him/her understand what you need. As mentioned, what is done in CAD doesn't mean targeting visualization in the first place. So, don't fix what could be created from the start in a better way. For a CAD operator, that is often the same work either way. I'm sure everyone likes to get a thank you instead of perhaps a cool OK. So, I stick with the idea that the workflow has room for improvement when I get a question like yours.

I was an office principal/manager/lead designer in the early '90s, and my first project was to connect my teams of architects with our CAD department. That was quite new back then. It took a while; I trained the CAD people a little bit in Architecture, and the architects went with me to Nemetschek and enjoyed some CAD training. Building bridges. After that, the CAD room was overbooked, and everyone was happy. Since then, I have seen many times that things have options to get better. Try it.
Yes, my roots with Nemetschek (Allplan, ArchiCAD, and Vectorworks) are deep, and I love them. It was fun to see Maxon joining this great company. In short, I believe there is always something possible.

There are methods to improve, but again, I can't say anything in general. I need to see the model. Sometimes it is as easy as using the Untriangulate option or combining tools. This is kind of experience based. Take a copy and mess around with it. Of course, ask, but please with an example.

I'm happy to look into it. After a while, you will see what is possible. As everything has quality, it needs some time to develop.

Cheers

P.S.: Please have a look here, where Jonas goes in depth into Noise and distortion.

https://youtu.be/m-1LvPKhGjc
https://youtu.be/c5uhSkEiPRc
The long version with more details:
https://youtu.be/Lf7JaGaeDP4

Hi miss-bear,

You haven't specified what render engine you like to use.

Here is an example (left_lower) for the Standard Render, while using an RS Ramp in Redshift requires some UV work (right lower).

I have two options in the top row using Vertex-based setups, manually weighted or via Field, both Redshift 3D.

I hope one will work for you.

CV4_2023_drs_23_REcg_01.c4d

Cheers

Thanks for the feedback, miss-bear.

Great! Thanks for the feedback.

I hesitated with the UV suggestion, as that is not just a general do-this-do-that suggestion. In other words, if you have a UV solution, like UV Packaging, those can't be just rotated (well, sometimes, sometimes not). Hence why I always ask for a project file.

All the best