Yi 4K red and orange the same.

Your expectation of the camera's ability to perform by itself is rather ... uninformed. What a camera body is technically capable of is rather different a beast than what any one user gets in various settings unless you understand 1) precisely what each setting does in every situation with that specific body and 2) how every step of your process (from exposure through any post) inter-relates.

And this doesn't matter if it's a relatively cheap 1,000 pound DSLR to a 20,000 pound RED. The user's knowledge of how to utilize the camera's capabilities in the end matters far more than the cost of the camera.

I've only had 40 years of professional photography as a career, before adding video 5 years ago. In the portrait stills crowd, I'm considered "somewhat" of a tech nerd. I built profiles for analyzing color negatives for printing back when we had our full color lab, for our own cameras as well as other studios. Each film needed a "standard" exposure profile, plus -2, -1, +1, +2 exposure profiles. A very intense detailed process to determine exactly the "standard" exposure rating for a film (which was never exactly what the manufacturer listed, for reasons of great depth), and then to make the entire set of negatives for every printing device or enlarger ... they all needed different sets.

I learned how to use a meter in the camera room to precisely set complex lighting and get every part of it nailed, which ... is not using a meter the way it was normally taught. I couldn't get the precision of the base exposure the way it was supposed to work, but from testing an altered method ... I learned to always got the base exposure nailed. Which was easy to test for in printing.

After we switched from medium-format film to digital in 2003, I had to test every setting on every camera to see what it did precisely, and select which settings worked best for our needs, what ISO ranges we could use, and how to stretch at need.

Including, I've naturally tested the actual output of our two bodies that do video for me in the same way. Both very good cameras, a bit "up" shall we say from yours, and yet, they have very different results for basic color/hue range of a scene, the way they handle skin tones, roll off highs & shadows, everything. One a Panasonic, one a Nikon.

According to your assumption, they should produce nearly identical images. They don't. (They do both produce very good images ... just very different color and some difference in tonality.)

And knowing how to blend their results in post takes ... work. Training. Testing. Practice. Shoot & test. From those results, shoot again & test. Again. Again.

Chris said he found it fascinating that a 3d LUT made from a DSLR color match works similar to that for an Alexxa. I have no surprise with that. Camera makers of those actual spendy beasts go to amazing lengths to create perfectly profiled LUTs to recreate a scene technically correct ... therefore, neutral. (Read: technically correct but boring.) Based on detailed test charts & such. But they only work perfectly IF you start with the camera and the same lens in the same studio with the same lights and scene set exactly the same.

Which is why Van Hurkman and others teach applying the normalization LUT to a clip such that you can go to controls processed before that LUT, and doing basic exposure, tonality, and WB corrections while watching your  scopes & image to fit the image through that LUT to the program monitor as the best "neutralized" image possible. First step of grading ... and there will be several more passes.

Many of the colorists I know will take a manufacturer's LUT, apply it to a clip while watching their scopes, and see what it does. And then go about building one for that camera as their base LUT for X camera, that to them is a more ... pleasing ... look. And then create different variants of it for different situations, probably using a very different shadow curves shape for night shots, for instance, both in general tonality and color response.

And never use the manufacturer's LUT again.

When confronted with a new camera shooting some form of encoding that needs normalization, the first thing they do is quickly run through the normalization LUTs they built and stored for the other cameras they've worked with. Several will probably work well, but ... for this scene, one might be better. Then they just tweak it. Why rebuild London Bridge?

Also ... they don't tend to apply LUTs simply to every clip of media on a line, except for the ones in some Log or other specialized encoding that need a normalization process. Why? They're far faster at doing a quick trim for tonality & color balance than they can possibly match by spending time trying several LUTs. And they hit a few clips of one cam, the settings seem mostly similar, copy/paste to the rest of that cam in that scene. Done.

There's no perfection guaranteed by gear. The cameras are just a tool. And cannot be calibrated in the factory to yield "perfect" color/exposure in every scene, as there isn't any perfect exposure standard. Perfect for what? The skin, the sky beyond that's so much brighter, the shadows of that building? They are all different. What do you need most to have a particular look? Then ... what else do you need to have detail & color information of a certain sort? Build your on-set color & exposure around those, that gets you to post with something you can than work to where you want it to be.

Neil