Attila Bakos

It's not a good idea to judge banding by youtube footage.
If I apply more contrast to the youtube frame to show banding, this is what I get:

Now if I go to vimeo and download the original ungraded version, and apply about the same contrast, this is what I get:

 
And the original ungraded version uploaded to vimeo is still 8bit! In 10bit you will have even less banding.

It's not a good idea to judge banding by youtube footage.
If I apply more contrast to the youtube frame to show banding, this is what I get:

Now if I go to vimeo and download the original ungraded version, and apply about the same contrast, this is what I get:

 
And the original ungraded version uploaded to vimeo is still 8bit! In 10bit you will have even less banding.

It's not a good idea to judge banding by youtube footage.
If I apply more contrast to the youtube frame to show banding, this is what I get:

Now if I go to vimeo and download the original ungraded version, and apply about the same contrast, this is what I get:

 
And the original ungraded version uploaded to vimeo is still 8bit! In 10bit you will have even less banding.

It's not a good idea to judge banding by youtube footage.
If I apply more contrast to the youtube frame to show banding, this is what I get:

Now if I go to vimeo and download the original ungraded version, and apply about the same contrast, this is what I get:

 
And the original ungraded version uploaded to vimeo is still 8bit! In 10bit you will have even less banding.

While it's true that H.265 can have the same quality as H.264 with approx. 50% of the bitrate, it's only at low bitrates, like streaming. As far as I remember at such high bitrates like 400Mbps the difference would be almost non existent, but I can't show you any test at the moment that proves this, maybe someone else can.

To me it seems like 4K60 10bit 4:2:0 internal, and 4:2:2 external.

There has been a lot of updates of the app, so give it a try, and let me know what you think.
To save the generated LUT, press the "cloud download" icon.
Android app can be downloaded here: https://play.google.com/store/apps/details?id=com.kinoseed.matchcolor
Here's how things look like now:

Thanks a lot! I don't have a Fuji body at the moment, but I have some savings for the X-T3. If it turns out to be good then I'll make new LUTs. Until then visit my experimental project: http://colorizer.net/index.php?op=lut_contrast
With this tool you can remove the contrast curve of a 3D LUT. With the method I use the curve of the Eterna LUT can't be removed without distortion of the RGB transformation, but you won't necessarily see it in real-life footage. Maybe try 50% strength or less.

I recently finished this project. If you like the results, download the LUTs for free on colorizer.net.
 

and here is a complete work (some art event deep inside russian territory):
 

Confirmed.

Resolve makes it very easy to convert the color space from Rec.2020 to Rec.709, using the built-in color space transform plugin.
In case you are using another NLE, I generated a LUT using 3D LUT Creator that does the same conversion: Rec.2020_to_Rec.709.cube
I couldn't test it, but it should be fine.

I believe it's user error, and a common one unfortunately, even among people who actually teach log grading. He shot many clips in S-Log2/S-Gamut, and forgot to convert color space during color correction.
This a screenshot from his video:

Here it is corrected with the color space transform plugin in Resolve, I only added a bit of contrast:

Now if I skip color space transformation and only apply contrast and saturation, see what happens:

And finally this is the uploaders grade:

 
So it's pretty clear what's happening here, it's not the camera's fault.

I recently finished this project. If you like the results, download the LUTs for free on colorizer.net.
 

Make sure you check the article as well, it has tons of comparison images and information.

Resolve makes it very easy to convert the color space from Rec.2020 to Rec.709, using the built-in color space transform plugin.
In case you are using another NLE, I generated a LUT using 3D LUT Creator that does the same conversion: Rec.2020_to_Rec.709.cube
I couldn't test it, but it should be fine.

In case you change your mind, she's on pornhub. Or so I've heard...

I believe it's user error, and a common one unfortunately, even among people who actually teach log grading. He shot many clips in S-Log2/S-Gamut, and forgot to convert color space during color correction.
This a screenshot from his video:

Here it is corrected with the color space transform plugin in Resolve, I only added a bit of contrast:

Now if I skip color space transformation and only apply contrast and saturation, see what happens:

And finally this is the uploaders grade:

 
So it's pretty clear what's happening here, it's not the camera's fault.

I believe it's user error, and a common one unfortunately, even among people who actually teach log grading. He shot many clips in S-Log2/S-Gamut, and forgot to convert color space during color correction.
This a screenshot from his video:

Here it is corrected with the color space transform plugin in Resolve, I only added a bit of contrast:

Now if I skip color space transformation and only apply contrast and saturation, see what happens:

And finally this is the uploaders grade:

 
So it's pretty clear what's happening here, it's not the camera's fault.

I believe it's user error, and a common one unfortunately, even among people who actually teach log grading. He shot many clips in S-Log2/S-Gamut, and forgot to convert color space during color correction.
This a screenshot from his video:

Here it is corrected with the color space transform plugin in Resolve, I only added a bit of contrast:

Now if I skip color space transformation and only apply contrast and saturation, see what happens:

And finally this is the uploaders grade:

 
So it's pretty clear what's happening here, it's not the camera's fault.

I believe it's user error, and a common one unfortunately, even among people who actually teach log grading. He shot many clips in S-Log2/S-Gamut, and forgot to convert color space during color correction.
This a screenshot from his video:

Here it is corrected with the color space transform plugin in Resolve, I only added a bit of contrast:

Now if I skip color space transformation and only apply contrast and saturation, see what happens:

And finally this is the uploaders grade:

 
So it's pretty clear what's happening here, it's not the camera's fault.

The camera eats the batteries one by one, the one in the body comes first, then it switches to the ones in the grip. When this switch happens, recording stops. The camera won't turn off, you just have to press record again. It's easy to forget, and because the screen is on, it's easy to remain under the impression that your recording is on. Then comes the surprise, and blood pressure rises to 200 in like 1 second.

The camera eats the batteries one by one, the one in the body comes first, then it switches to the ones in the grip. When this switch happens, recording stops. The camera won't turn off, you just have to press record again. It's easy to forget, and because the screen is on, it's easy to remain under the impression that your recording is on. Then comes the surprise, and blood pressure rises to 200 in like 1 second.

The problem with the grip is that the body can't switch batteries while recording. This is rarely mentioned even though it can be a real pain in the ass. Imagine you have 5 mins left in the battery inside the body and two fully charged batteries in the grip. You will still be limited to 5 mins of continuous recording, as the recording stops when the currently used battery dies. So you have to keep an eye on that, and in this case remove the battery if you plan a shot longer than 5 mins.

The problem with the grip is that the body can't switch batteries while recording. This is rarely mentioned even though it can be a real pain in the ass. Imagine you have 5 mins left in the battery inside the body and two fully charged batteries in the grip. You will still be limited to 5 mins of continuous recording, as the recording stops when the currently used battery dies. So you have to keep an eye on that, and in this case remove the battery if you plan a shot longer than 5 mins.