Why choose 2020 Output Device Transform if you actually have no 2020 device?

FWIW, there's Gamma and other stuff:

https://en.wikipedia.org/wiki/Transfer_functions_in_imaging

https://en.wikipedia.org/wiki/Perceptual_Quantizer

The only way I can output HDR10 is to use the rec 2020 1000 nits display transform as per the manual.

I don't think I can monitor it in rec 709 and change the color space in the render settings to rec 2020 1000 nits and be accurate at all. I have a TV and my ninja V set up so I can see the brightness differences but as for color, I already graded it for rec 709 for the SDR project.

The V-Log L in the GH5 and my AG-CX350 uses the rec 709 color space and not the full V-Gamut space since Panasonic only gives the full V-Log to more expensive cameras that can capture 14+ stops. My CX350 only captures about 12.5 stops so they give it the cheaper V-Log L.

Yes, precisely. I think that is what some folks missed when considering the original question, which has since morphed into a broader discussion.

It's very frustrating but I sort of understand why they do it.

It's really the best option for delivery, for the time being. Kind of a stopover until the broadcast standards catch up and sell lots of new hardware, at least in the US.

One question I do have, @Musicvid regarding this, is since that is true, why does my color get flat like it's really rec 2020 when it's just rec 709 in the the rec 2020 container? Is it stretching it out past rec 709 limits to cause this? As soon as I activate the rec 2020 1000 nits transform it turns flat and unsaturated on rec 709 screens while it shows fine on rec 2020 screens.

Yes, I think that's close. It's the old "5 gallons of water in a 10 gallon bucket" analogy I've been using for years. Properly conformed on an HDR display, the 709 source will look flat alongside wide gamut source; it doesn't grow to fill the space, but it can be mapped to a wider, but less dense plat. Properly conformed on a 709 display, they would look roughly the same, but the wide gamut probably warm in comparison, owing to the unequal vector distance shift from the white point.

I think we're both learning a lot from the process of this discussion, @Teagan, and it requires more than a two-dimensional thinking model to conceptualize, my own understanding being far less than complete. I'm probably too old to get it completely.

Why choose 2020 Output Device Transform if you actually have no 2020 device?

FWIW, there's Gamma and other stuff:

https://en.wikipedia.org/wiki/Transfer_functions_in_imaging

https://en.wikipedia.org/wiki/Perceptual_Quantizer

The only way I can output HDR10 is to use the rec 2020 1000 nits display transform as per the manual.

I don't think I can monitor it in rec 709 and change the color space in the render settings to rec 2020 1000 nits and be accurate at all. I have a TV and my ninja V set up so I can see the brightness differences but as for color, I already graded it for rec 709 for the SDR project.

Of course you can't, you'll have to regrade and remaster - using HDR10 capable monitor. :)

The only way I can output HDR10 is to use the rec 2020 1000 nits display transform as per the manual.

I don't think I can monitor it in rec 709 and change the color space in the render settings to rec 2020 1000 nits and be accurate at all. I have a TV and my ninja V set up so I can see the brightness differences but as for color, I already graded it for rec 709 for the SDR project.

Of course you can't, you'll have to regrade and remaster - using HDR10 capable monitor. :)

Would my Ninja V with 1000 nits HDR display work, even though it's capped at 100% rec 709?

I tried using it and the only benefit I get is the brightness part of the grading, the colors are washed out, but when I render the project it looks perfect on the same screen, with perfect rec 709 colors.

The only way I can output HDR10 is to use the rec 2020 1000 nits display transform as per the manual.

I don't think I can monitor it in rec 709 and change the color space in the render settings to rec 2020 1000 nits and be accurate at all. I have a TV and my ninja V set up so I can see the brightness differences but as for color, I already graded it for rec 709 for the SDR project.

Of course you can't, you'll have to regrade and remaster - using HDR10 capable monitor. :)

Would my Ninja V with 1000 nits HDR display work, even though it's capped at 100% rec 709?

I tried using it and the only benefit I get is the brightness part of the grading, the colors are washed out, but when I render the project it looks perfect on the same screen, with perfect rec 709 colors.

 

I don't know, don't have a Ninja.

What confuses me is why you're getting the "correct colors" once you've rendered your project to a file, but not when grading. :)

Would my Ninja V with 1000 nits HDR display work, even though it's capped at 100% rec 709?

I tried using it and the only benefit I get is the brightness part of the grading, the colors are washed out, but when I render the project it looks perfect on the same screen, with perfect rec 709 colors.

HDR is Rec2020. You cannot have perfect color in HDR with Rec709. If you cannot preview Rec2020, you'll need to get a proper HDR monitor. I'd recommend OLED for color reproduction capability.

+1 to all of the above.

My camera shoots HLG but not HDR10. I'll have to give it a shot.

https://www.eoshd.com/news/sony-a7s-iii-10bit-image-quality-vs-same-camera-in-8bit-with-surprising-results/

https://www.eoshd.com/news/sony-a7s-iii-10bit-image-quality-vs-same-camera-in-8bit-with-surprising-results/

Yes it's interesting but there's no way I'm shooting in 8 bit and delivering in 10 bit, of which HDR10 is 10 bit - especially when I am tweaking exposure in large increments in post sometimes. There's no way that 8 bit gives me that much to work with when I'm doing that.

I don't have the A7S III / FX3 yet, so that's another thing. I'm saving for the FX3 after deciding to ditch the S5.

https://www.eoshd.com/news/sony-a7s-iii-10bit-image-quality-vs-same-camera-in-8bit-with-surprising-results/

Andrew really needed to shoot a greater variety of scenes with smooth gradations in parts and higher contrast overall. I understand it's overcast there but these examples aren't adequate to draw conclusions from. It's not relevant to HDR either as high dynamic range combined with inadequate bit depth to describe the values within that range is a recipe for poor image quality.

I look forward to working in HDR spaces, but until adoption of HDR screens is more widespread, and I can afford an accurate grading-level HDR monitor, I can't do much to test it now.

Good luck Teagan and let us know what you come up with!

@Teagan And heres some more ... http://www.xdcam-user.com/hdr/

VLogL doesn't use rec709. It uses a version of vgamut made to work with 12 stops. This is why VlogL to rec709 luts exist. I use vlogL alongside my full VLOG cameras in HDR projects all the time, the same luts work for the most part (I do a few minor tweaks), as do the same view transforms in ACES.

VLogL doesn't use rec709. It uses a version of vgamut made to work with 12 stops. This is why VlogL to rec709 luts exist. I use vlogL alongside my full VLOG cameras in HDR projects all the time, the same luts work for the most part (I do a few minor tweaks), as do the same view transforms in ACES.

I've asked panasonic reps about this, if the AG-CX350 has full v-gamut in V-Log L and the answer is no, the GH5 and AG-CX350 use rec 709 and not v-gamut for the color space. The gamma may be v-log, but the color space is not v-gamut.

Just to update, I have found that this is not true, both cameras DO use v-gamut in v-log L. I believe the confusion was coming from the color space it looks like it's in when unpacked/flat on a chromaticity scope on davinci resolve. After unpacking it with a view transform to rec 2020 it fully expands. With an SDR camera I have some footage in the same project, that maxes at rec 709 on the same scope, while the ag-cx350 fully expands to rec 2020.