BMPCC 6K Pro vs Sony A7S III?

Why does shooting for HDR 'absolutely demand' an HDR monitor?

Because you can't judge an HDR image on an SDR display.

That would obviously be the case during post production but why during image capture? As long as you can expose correctly via the normal exposure controls built into cameras like zebras, false colour etc I can't see the 'absolute demand' for this even if it's nice to have. In addition viewing an HDR monitor in a working i.e brightly lit environment defeats it's HDR capability although the increased nits of these displays certainly makes them usable in bright environments but that's not the same thing.

With a monitor, you're not only looking at things like exposure and shadow and highlight clipping using tools like zebras, waveform monitor and false color - you're also judging framing, color, focus (and if light stands and other stuff are in the frame! hehe). Issues that may not be of concern in SDR and whose effect is impossible to see on an SDR monitor can be detected early on with an HDR monitor. I use the Ninja V to judge things like the ratio between background highlights and the light falling on my model's face. That ratio is different in HDR and SDR. So I'm always checking both my scopes (usually toggling between false color and waveform monitor) and confirming visually using the Ninja V. Glare from jewelry and other bright light sources can ruin the image and are best avoided by using a proper HDR monitor. I shot many of my earliest HDR videos using just the internal codec and zebras on the camera's LCD, so it's certainly possible to shoot HDR without a monitor; but false color, waveforms and being able to see exactly how the image will look help reduce any anxiety! When I use an external monitor, I've got confidence, knowing that what I'm seeing on the display is how it will appear in the NLE. And of course, shooting RAW gives me more flexibility and richer colors than shooting with the internal codec.

I never shoot in brightly lit environments, so monitoring with the Ninja V has never been a problem.

So it's not an 'absolute demand' - thanks for clarifying.

So it's not an 'absolute demand' - thanks for clarifying.

You can ride a motorbike without a helmet, too. A celebrated DP in our midst has shot without an onboard HDR monitor as well but it's certainly preferable to have one. For HDR, you should always shoot RGB 4:4:4 or RAW and avoid Y’CbCr 4:2:2 codecs, making an external recorder indispensable for most mirrorless cameras today - and more bits means more leeway in post. The Ninja V also has waveform monitor, vectorscope, false color and other useful features often lacking on mirrorless cameras and is even helpful for accurately monitoring log footage not intended for HDR.

magine if you will that due to present circumstances we're no longer able to shoot Hollywood blockbusters but instead, we're compelled to shoot mundane documentaries, travel and nature films, product photography, YouTube videos, music videos, corporate training videos, weddings, real estate, fashion and beauty, food or fitness videos. Cringe, right?

I'm shooting in these applications and I still find there is a compelling case for an URSA 12K or 4.6K if you are making a living from it. It's still a relatively small box camera where you need maybe 2 batteries, 2 cfast cards and a lens to get going.

Where I use a hybrid in my workflow, it's now the Pocket 6K Pro with EVF. I've found that finally to be the replacement for corporate, etc where I was still clinging to wanting a Canon C100 for internal ND, EVF, sound recording. I'm getting IS from the Canon lenses in those situations. Now I feel like I have the ergonomics with the 6K Pro without adding a monitor and have the benefit of the internal BRAW and Color Science Gen 5 with the swappable "Video" and "Film" with lots of room to push the images.

I don't feel locked to the eco system, but that along with Resolve it's consistently met my needs without headache or trying to figure out what piece of metal goes with what. Even if it's not substantially cheaper than another camera system, I still think it's a solid choice for the applications you mentioned.

I actually still used my HD Micro Cinema Cameras recording internal ProRes this past winter quite a few times mounted to helmets for snowboarding apparel. Everything else still feels too big without going back to GoPro .

For HDR, you should always shoot RGB 4:4:4 or RAW and avoid Y’CbCr 4:2:2 codecs,

I'd like to see some evidence that HDR from an internal codec (as long as it has the bit depth, gamut and DR to meet REC2020 output) is to be avoided. BRAW is not RGB 4:4:4 or 'technically' RAW so does this make the grade or is only ProRes RAW acceptable? In every situation either SD or HDR you will get more flexibility in grading with better codecs (internal or external) but as numerous tests have proved it depends on what you are shooting and how you grade as to if you actually need the extra info or the hassle of an external recorder.

There’s nothing to stop other brands of camera manufacturers from using BRAW. I suspect it’s only a matter of time before one of them pay the not insubstantial asic chips dev costs to do so. BRAW is FPGA based, but as far as I know, no one else is other than Arri.

There’s massive restrictions to anyone using ProRes RAW internally even if someone wanted to.

One is open, and CAN be used internally. The other isnt open and can’t be used internally.

I know the lead developers of BRAW. They have had many asking about implementing BRAW internally. BMD have a long history of using open systems and making tools more accessible, not less or licensed. I think it’s only a matter of time.

BRAW is already available to an external recorder on some major brands of hybrid. Nikon’s. Sigma. Also Panasonic and Canon pro cameras.

HDR on set monitoring isn’t really necessary or possible. For starters many HDR monitors aren’t really HDR monitors. More like high NIT monitors. Great for viewing in bright ambient lighting. Not really designed for critical monitoring.

I shot a whole season of The Great without any HDR monitoring because it can’t be done accurately on set. Certainly not from a low cost on board. The only way you could do it is to use higher end (20k) HDR monitors in the DIT tent. Even then it’s still problematic to get consistent results.

Do I monitor waveforms and vectors using HDR tools ? Yes.

Do I monitor a HDR image using HDR monitors ? Not really technically possible or useful. Of course you can put a monitor that has a HDR badge on a camera. That doesn’t mean you’re getting anything useful from that monitor for critical colour / contrast assessment. I’m also including BMDs HDR monitor recorder here.

JB

BMPCC 6k Pro the so call HDR screen for me is definately more to do with higher nit the ability to see in bright sunlight.
Atomos's AtomHDR I can set it to see into the shadow and not over expose the highlight. It may have little use on what is actually recorded but gives a pretty picture.
BMD Video Assist HDR 12G (though mine still have the magenta cast) gives a more solid colour and less crushed shadow and looks more real.

BMPCC 6k Pro the so call HDR screen for me is definately more to do with higher nit the ability to see in bright sunlight.
Atomos's AtomHDR I can set it to see into the shadow and not over expose the highlight. It may have little use on what is actually recorded but gives a pretty picture.
BMD Video Assist HDR 12G (though mine still have the magenta cast) gives a more solid colour and less crushed shadow and looks more real.

The HDR screen on the 6K and the HDR Video assist are the same unit.

JB

BMPCC 6k Pro the so call HDR screen for me is definately more to do with higher nit the ability to see in bright sunlight.
Atomos's AtomHDR I can set it to see into the shadow and not over expose the highlight. It may have little use on what is actually recorded but gives a pretty picture.
BMD Video Assist HDR 12G (though mine still have the magenta cast) gives a more solid colour and less crushed shadow and looks more real.

The HDR screen on the 6K and the HDR Video assist are the same unit.

JB

That is a surprise. To me they look different.
According to the spec
BMPCC 6k Pro screen rated at 1500 nits
VA 12G HDR rated at 2500 nits

Because of energy consumption, maybe?

VA 12G HDR has options to change contrast and saturation.
BMPCC 6k Pro along with the two other pocket cameras since firmware 7.3 has option to calibrate screen temperature and tint. If 6k Pro screen is the same as the VA 12 HDR then I would hope BMD allow screen calibration in future firmware as mine is suffering from magenta cast and others have reported the same.

While it's true that HDR monitoring on set is costly; that not every monitor labeled HDR is suitable for high-end production work; and that HDR projects continue to be shot without an HDR monitor - what is not true is that monitoring on set is unnecessary and impossible.

"For season two, we knew we were going to finish in Dolby Vision HDR from the beginning. With that in mind, and taking what we had learned the first season, I did extensive testing prior to our preproduction period to explore how it might affect my lighting and exposure. I worked with the DI colorist, Eric Weidt. He and I developed a color pipeline and on-set monitoring system that allowed us to monitor in HDR using an ACES workflow in Dolby PQ gamma and Rec.2020 color. For us, the setup worked great and I’m really proud of what we put together. We settled on the Canon range of HDR field monitors with the DP-2420 as our primary monitor. If one is going to finish in HDR, I believe it’s absolutely essential to monitor in HDR on the set, as HDR directly affects lighting ratios and exposure. [italics added] During the DI of the first season I found we were stretching our highlights out to fill the added space in the HDR gamma space, which sometimes led to odd-looking practically lit night interiors. That is because, in most cases, I was not making full use of the sensor’s contrast, because my lighting choices were informed by an SDR monitor and waveform. Conversely, when monitoring in HDR I found I was less concerned with highlight detail, because I could see exactly where my exposures fell relative to the gamma of the final finish, and therefore was able to use more of the sensor’s dynamic range. It meant much less fill and more “exposure to the right” while still feeling confident I could work in the tow of the exposure range for most of the scene’s action." - DP Erik Messerschmidt on Mindhunter

Messerschmidt brings up an important point about lighting and how crucial it is to monitor in HDR. During Season 1, he says, "...my lighting choices were informed by an SDR monitor and waveform". What ends up happening is that 99% of HDR films on Netflix and elsewhere, while technically HDR, are in reality no more than SDR movies with a few brighter specular highlights thrown in here and there.

This is puzzling. You've said nothing here to indicate that you're shooting material for broadcast or cinema exhibition; you'd lack not only these on-set resources, but the post-production expertise and equipment to support broadcast-level HDR delivery. For no-budget production, this controlled HDR pipeline is plainly fantasy.

Similarly, for actual crewed low-budget production, reliable HDR monitoring doesn't appear to be either practical or affordable at this point. With higher budgets, John Brawley presumably does have these resources, but he's already indicated he finds HDR monitoring unworkable and, evidently, inessential.

If Netflix HDR productions are not up to your standards, which perceived shortcomings you attribute without evidence to SDR monitoring (the techs on set really have no idea what they're doing?), you might want to consider that another HDR medium, 35mm film, was shot for 100 years with no monitoring at all, and said deficiency didn't deter DPs from exploiting the full potential of the medium.

Granted, it's self-evidently true HDR monitoring may be "preferable" [EDIT: Jon Pais has since removed this assertion, as well as made other alterations to the comment immediately above, after this post appeared. This post reflects his original comment.], so you probably won't get much of an argument there. I hope that's not too disappointing.

Fantasy is correct.

I tried to get HDR precision monitoring on The Great but when we did it (early 2019) the only actual good HDR monitors that were able to be used cost many tens of thousands of dollars per monitor. It wasn’t practical and they just weren’t available in the UK even if we could spend the money.

They’re coming down to be field affordable but you’re still looking at 10k per monitor. That’s 30k in monitors that get trashed within a few months of use being wheeled around. Most DITs I work with won’t spend that kind of money on monitors until they get paid properly and even big shows don’t want to spend that kind of money.

And still it’s a problem to try and set up what is essentially a grading suite on location. Ambient light makes this nearly impossible, especially with HDR because it affects the blacks so much.

A lot of people with HDR focus on the highlight dynamic range. Most of the time I found what was more interesting was to use the extra shadow range.

I just asked Eric by the way. He used Sony 17 HDR monitors on the movie he just wrapped. A 120 million dollar budget.

They’re not even 4k. You have to buy a HDR license on top of the 11495 you spend here.

https://www.bhphotovideo.com/c/product/ ... nitor.html

True critical HDR monitoring on set is still very difficult. No one but the most resourced can claim to do it well.

A HDR badge on a monitor doesn’t mean you’re monitoring on HDR.

JB

what's more, an arri alexa or red on the SDI output doesn't even send out
a reasonable true color signal, not to mention HDR ; -)

and as john mentioned, it is impossible to create an environment on set
that allows HDR to be judged properly.

.

He and I developed a color pipeline and on-set monitoring system that allowed us to monitor in HDR using an ACES workflow in Dolby PQ gamma and Rec.2020 color. For us, the setup worked great and I’m really proud of what we put together.

They had to invent the workflow that would make on-set HDR possible for Fincher who would accept nothing less than uncompressed 16bit, 18 stop, 12K.

My guess is that they created a PQ LUT that maps white to something more production friendly like 300-500 nits.

The difference is that the monitoring gamma won't be the same as the deliverable gamma but they can both be HDR. It's about time, log is worse than 709.

Good Luck

If the DP of an HDR production genuinely wants to shape the look of the image from the start, viewing in SDR on set is woefully insufficient. Which may account for the fact that so many HDR productions fail to thoroughly exploit the dynamic range HDR is truly capable of - and that viewers have come to expect. At the other end of the spectrum, that is, those not working on high-end productions, it is entirely possible for anyone with a Pocket 4K, a7s III or any other camera that shoots RAW, to create stunningly beautiful HDR PQ videos.

The forum member who claims that RED and ARRI cameras do not output good color over SDI is living in an alternate reality.

It’s also possible to create stunning HDR images without HDR monitoring on set.

It’s been happening for years.

JB

It’s also possible to create stunning HDR images without HDR monitoring on set.

It’s been happening for years.

JB

We are in agreement that the shadow range is by far the most interesting.

Krzysztof Sieniawski, who shot period drama Born for the Saber on a budget of around $120,000 using a GH5 and recorded to ProRes with Atomos Ninjas, sold his film to the History Channel. The production’s decision to record in ProRes using a Log image had another advantage. After completing the SDR version of the film, an HDR version of the film’s trailer was also commissioned. Therefore, by shooting in log, the filmmakers now have the option to also grade for HDR. Thus they have a backup if they need to deliver an HDR version in the future. While it would have been preferable to shoot in RAW, there is no question whatsoever that an Atomos recorder can be used for HDR monitoring on low-budget productions.

You’re trolling now. It was better when you weren’t making redundant statements.

JB

If the DP of an HDR production genuinely wants to shape the look of the image from the start, viewing in SDR on set is woefully insufficient. Which may account for the fact that so many HDR productions fail to thoroughly exploit the dynamic range HDR is truly capable of

This is why you as an HDR ‘enthusiast consumer’ can’t grasp what goes on in the real world of the content creator. A DP uses above all else their eyes and experience to shape the look of the image and not any monitoring. You yourself claim our eyes can see 20stops so what HDR monitor is as good as that anyway? Monitoring just confirms choices and (I repeat from my earlier post) basic camera exposure tools are perfectly capable of ensuring the full DR of the sensor is exploited and no clipping or shadow blocking will occur in the deliverables to post. This is all that’s required on set and the final HDR result that you seem to be so disappointed is a decision in post production and not a failure to monitor in HDR on set. I would go further and say that the main reason for the present limitation on HDR is not at capture but at production as the cost of a proper HDR monitor (not an HDR capable OLED consumer TV pressed into service as a monitor) is prohibitively expensive for most productions. Even monitors for accurate REC709 are only just now ‘affordable’, for DCI-P3 they are still very expensive and for REC2020 - out of reach for most.

Our eyes can perceive a range of about 20 stops, but not at the same time.
Going from dark to light will take a second or two, but going from full sunshine into a dark room the biochemical process can take between 10 and 20 minutes.

Our eyes can perceive a range of about 20 stops, but not at the same time.
Going from dark to light will take a second or two, but going from full sunshine into a dark room the biochemical process can take between 10 and 20 minutes.

He is gaslighting. Never did I write anywhere that our eyes can see twenty stops. Like all his other drivel, in and of itself, that statement is completely meaningless.

Our eyes can perceive a range of about 20 stops, but not at the same time.
Going from dark to light will take a second or two, but going from full sunshine into a dark room the biochemical process can take between 10 and 20 minutes.

He is gaslighting. Never did I write anywhere that our eyes can see twenty stops. Like all his other drivel, in and of itself, that statement is completely meaningless.

Yes you did in another thread when I questioned this in relation to HDR and the static DR capability of the human eye. I will paraphrase but it started with the usual personal attack such as ‘nonsense’ and claimed most experts say 20stops is what the human eye is capable of. You now seem to have deleted it so who is doing the gaslighting…….
I see now you are keeping up with the personal insults that have got you banned from numerous other forums.

Our eyes can perceive a range of about 20 stops, but not at the same time.
Going from dark to light will take a second or two, but going from full sunshine into a dark room the biochemical process can take between 10 and 20 minutes.

He is gaslighting. Never did I write anywhere that our eyes can see twenty stops. Like all his other drivel, in and of itself, that statement is completely meaningless.

Yes you did in another thread when I questioned this in relation to HDR and the static DR capability of the human eye. I will paraphrase but it started with the usual personal attack such as ‘nonsense’ and claimed most experts say 20stops is what the human eye is capable of. You now seem to have deleted it so who is doing the gaslighting…….
I see now you are keeping up with the personal insults that have got you banned from numerous other forums.

Cut it out, buddy. Never did I write that our eyes can see 20 stops, nor was it necessary to delete such a patently ridiculous falsehood. I was clearly talking about the DR of HDR television sets, not the human visual system. My writing style is fairly simple and direct - it would be difficult for anyone to misunderstand what I wrote.

.....
The forum member who claims that RED and ARRI cameras do not output good color over SDI is living in an alternate reality.

then take a closer look at the signal and compare it with the recorded material ; -)

after 25 years of editing, now i work as video operator and DIT the last 14 years.

.

I'm living in that reality too. The signal for viewing on those cameras is a just an approximation of real thing and every serious DP who is using a Red or an Arri knows that.

One pretty solid source on the DR of the human eye (scroll down): https://tinyurl.com/ypeu2f2k

And now I'm off this thread!

You yourself claim our eyes can see 20stops so what HDR monitor is as good as that anyway?

The Sony BVM-X300, to name just one! I thought we already had this conversation...

Even now you are editing and rewriting posts. I stand by my memory and as I was not the only one reading and contributing to that thread why would I make something up. An archived version before you deleted will be stored so I’m happy if you want to ask the moderator for for this as proof as it’s you who are accusing me of gaslighting but it’s you who have ‘form’ (as John P noted ) of retracting and editing previous posts.

.....
The forum member who claims that RED and ARRI cameras do not output good color over SDI is living in an alternate reality.

then take a closer look at the signal and compare it with the recorded material ; -)

after 25 years of editing, now i work as video operator and DIT the last 14 years.

.

Correct. RED have gotten better but their first SDI outputs were total garbage compared to what was recorded. Check out the banding in the corners on this photo of a red monitor.

https://www.flickr.com/gp/johnbrawley/15fMZ8

It's why RED for a long time were never used in broadcast situations. The output wasn't good enough, and it's another reason why many users have traditionally eschewed external recorders, except for maybe proxy recording.

JB