Quote from: stoker2014 on December 29, 2023, 12:07

Quote from: Zero Talent on December 29, 2023, 12:01

Quote from: stoker2014 on December 29, 2023, 11:58

Quote from: Zero Talent on December 29, 2023, 10:45

This is probably because ProRes 4:2:2 is a native 10-bit codec.
https://www.loc.gov/preservation/digital/formats/fdd/fdd000389.shtml

"While ProRes is a 10-bit native codec, it can be used with either 8- or 10-bit sources and 8-bit sources"

Modern cameras can also shoot 10 bits in 4:2:0.

Sure. So when anyone sees a ProRes clip, they can safely bet that it's a 10-bit clip, even if they don't have Davinci Resolve or other tools to check it.

It's a lie! I can recode any video from any crap phone into prores 422.

The de facto bit depth of the video will be 8. Whatever Davinci Resolve shows. 

Many people, after reading this topic, will begin to render any of their videos from their phones in prores and upload them to stock agencies. 

I agree, but it's not a lie.

It is cheating.

The fact of the matter is that Davinci Resolve will show 10 bits, because the ProRes files are 10 bits, even if you cheated and invented 2 bits out of thin air, by falsely encoding an 8-bit file with a 10-bit codec.

The same goes for any other codecs. You can encode an 8-bit clip and sell it as a 10-bit H.265. MediaInfo will show 10 bits, Davinci will show 10 bits, but your client will have no way to know upfront, that someone is selling fake 10 bits clips.


So it doesn't have anything to do with ProRes, but with your integrity.

The only way to discover this cheat is when the buyer is trying to edit the clip, only to realize it has ugly sudden transitions and banding specific to 8-bit clips.

And, btw, my S22 Ultra can make native 10-bit clips in Pro mode.

Conclusions: there are no programs that show the video bit depth. Very few people have your DaVinci Resolve Studio software.

Zero Talent, I remain of the opinion that people buy your video not because it has 10 bits. But only because it is encoded in prores. Or perhaps just a good story.
 

Quote from: Zero Talent on December 29, 2023, 13:47

Yes, I learned something about MediaInfo, indeed. No shame in admitting it: it doesn't show the bit-depth for ProRes clips.

And yes, as I said, ProRes can also work for 8-bit clips. After upscaling them to 10 bits. All ProRes outputs will always be 10 bits minimum, even if the native file was 8 bits.

Below is the info for an old 8-bit H264 clip. Next to it, is the ProRes version obtained by asking Adobe Media Encoder to encode it with ProRes. I even removed the check forcing the encoding at maximum depth.
As expected, the output is a ProRes 10-bit clip.

I am sure that by using this cheat you may fool some buyers interested in quality 10-bit clips, but it is also likely that it may bite you back. The choice is yours.

+100

I think that buyers buy prores only because they are comfortable working with this codec. Also, the buyer can theoretically return any purchased clip if he doesn�t like it. I've heard of such returns.
The author of the video will never receive any claims or threats from the buyer. All these issues are regulated by stock agencies, and on the websites of these agencies there are no requirements for the authors you write about here.
I also believe that this is a defect of the prores codec, which represents 8-bit video as 10-bit. Moreover, I'm not sure that DaVinci Resolve even detects the video bit depth in the prores codec. Most likely, this program decided that if it is prores, then it means there are 10 bits. Thus, most buyers will never know what depth of bits they purchased. Many video editing programs will also not show how many bits are in the video in the prores codec.
 

You are writing about some kind of scaling of 8 bits to 10 bits. I'm not sure that this is even possible and that Media Encoder does this. You need to ask Adobe about this. And even if you are right, then so much the better, 8 bit videos will become 10 bit. 

You are already well known for denying the evidence, even when it is screaming in your face.

Unlike yourself, high end users and media players already know that ProRes 422 output is always 10-bit, by defintion. Period. End of line. No switch to 8-bit. 10-bit only.

Why do you enable the option to render with maximum bit depth? Your prores codec already renders everything at 10 bits, as you claim.
 

Quote from: stoker2014 on December 30, 2023, 06:22

Quote from: Zero Talent on December 30, 2023, 06:08

Why don't you try it yourself?

I tried it. The files have different sizes. I rendered 10 bit video.
The question is why? 

I'm not sure I understand what you did. What were your input and output settings?

Even inside the 10-bit ProRes family (LT, vs normal, vs HQ) the output files have different sizes depending on how much information the lossy compression is discarding.

Quote from: stoker2014 on December 30, 2023, 06:48

Quote from: Zero Talent on December 30, 2023, 06:37

Quote from: stoker2014 on December 30, 2023, 06:22

Quote from: Zero Talent on December 30, 2023, 06:08

Why don't you try it yourself?

I tried it. The files have different sizes. I rendered 10 bit video.
The question is why? 

I'm not sure I understand what you did. What were your input and output settings?

Even inside the 10-bit ProRes family (LT, vs normal, vs HQ) the output files have different sizes depending on how much information the lossy compression is discarding.

10-bit ProRes LT.
The first rendering is with maximum bit depth.
The second rendering is without maximum bit depth.

I took one source file and made 2 different renderings from it.

And both final files are 10-bit ProRes, right?
Except for the test above, I never experimented with that setting switched off.

I found this explanation, which means that the setting has a different purpose. Maximum depth means 32 bit for internal calculations instead of 8 (or 10?):
8-bit means the result of every calculation on the image can only have 256 possible levels per channel (RGB or YUV). That�s not a lot of accuracy. With multiple effects on a clip, this means we�ll get rounding errors for each calculation, with the danger of introducing banding and blocking.

In 32-bit, we�re doing all calculations with results that can have more than 4 billion different levels per channel, so rounding errors with multiple effects are totally eliminated. We can also store levels way beyond 100% white and 0% black.

So, it looks like that setting is mainly used for internal transcoding calculations. Maximum depth means more accurate calculations, not necessarily the bit-depth of the final file, which is fixed in our case, by the 10-bit ProRes standard.

Keep it checked. It may take longer to render, but the original information will be better preserved.

This is why it's better to shoot timelapses in RAW, process and export them as 16-bit TIFF