That Netflix Video everyone is sweating over

Looking at my feeds yesterday and this morning I'm seeing the reaction from the video:

. Additionally there's the Approved Camera List Netflix put up. Actually there's lot's to consider in their Partner Help Center.

It got me thinking about the post Tim Buttner put up asking Blackmagic to consider making the 6K Pro available with PL Mount as well as SDI. I suspect the SDI issue would have been the decider as to whether or not the 6K Pro would have made the list.

I really like shooting the 6K Pro. The built-in ND's work really well for how I shoot. Like Tim, I also would love to see better connections without always worrying if the HDMI connection will break off the circuit board. Having an interchangeable mount would be really cool. I would use both lens mounts depending the project.

Although Tim did not mention this, additional dynamic range would be a huge plus. I am finally getting ready for either a G2 or its next iteration to add this but I like the smaller form factor of the 6K Pro. The box form is not an issue for me.

Don't get me wrong, I really like shooing the 6K Pro a bunch. I'm thinking more about the shot and less about the gear which is the right direction for me to go. But having a more robust camera would be really good. I know it would add to the cost. And it would likely be added to the prized Netflix Approved list of camera's too. Woo Hoo!!!

BMD has a 6k pro with PL and SDI, it’s called the URSA Broadcast G2.

Neither feature is keeping the 6K off the Netflix list.

That video has drawn predictable responses: all this talk about image quality, but meanwhile Netflix has qualified consumer-level mirrorless cameras but rejects most Alexa models.

Why this list gets so much attention, particularly for persons not under contract with Netflix, is a wonder. It probably did drive the development of the Alexa 35, but I wonder if Netflix would want to take credit for that, given all the Alexas which didn't qualify.

What? Is this some sort of crypto test or something? When deciphered does it say earth is going to be hit by an asteroid or something? I'm surely out of touch. I don't watch netflix let alone sweat over it.

There is nothing scientific about their camera list. The 12k should be on it. The Broadcast G2 should be on it.

Something to do with dedicated time code input. UM12k took it off.
A big retailer in the UK is selling the UM G2 £300 more than the UM12k.

I not understand why people are too worry about this list. If you work on a original Netflix you work good covered, is not a problem to choose one or another cameras, if not they accept everything’s did, in Italian catalog they had also a miniDv movie be cause it receive award from some festivals.
Netflix approved list is another way to give work to pixelpeeper people to play with cameras and tool. For their requirements most of filmstrip movies not meets the basic requirements of dinamic range ….


Inviato dal mio iPhone utilizzando Tapatalk

There is nothing scientific about their camera list. The 12k should be on it. The Broadcast G2 should be on it.

Both of these are the most interesting cameras that I haven’t used. You could add the BMPCC6K2 as well since I don’t need the internal IRND option. It’s all about the best image quality.

Interesting discoveries while colouring the music festival shot on the original UM4.6K in lossless CinemaDNG Gen 3 Colour Science with the old 4.8 firmware. I can understand if anyone felt it’s a mistake for me to use that out-dated and tragically flawed camera sensor. Years ago I can remember having to deal with the magenta clouds and killer fixed pattern noise sometimes for example.

Well the magenta issue literally evaporated into thin air after a couple of years and in spite of dealing with many varieties of scene lighting with theatrical LED lights, I’ve never had a hint of fixed pattern noise on recent shoots.

But rather than bury the antique camera I’m here to sing the praises of the Colour science. I had assumed it’s a no-brainer that I should update all my Gen 3 clips to Gen 4 in the Camera RAW tab of DaVinci Resolve. And that might be true in many situations, but theatrical lighting can be difficult. Make no assumptions.

I found many CDNG clips did benefit from applying Gen 4; particularly with the medium close beauty shots under the intense spotlights. Gen 4 just provided better sculpting of the images adding depth when Gen 3 looked flat. Very good. But there were also several shots where Gen 4 actually caused severe artifacts and ruined a shot that Gen 3 handled well. Perhaps Gen 5 would have done better but that’s not a CameraRAW option for media shot with Gen 3. And several shots that were medium or wide where my smaller faces were lightened and just looked better staying with Gen 3. Best to colour each clip on its own merits.

So some pleasant surprises from the old camera that remains my workhorse. The producer was pleased and has complimented the quality of my camera gear. As you would understand, no kudos to the operator!

Something to do with dedicated time code input. UM12k took it off.

Huh?

The 12K has a dedicated TC input.

JB

Something to do with dedicated time code input. UM12k took it off.

Huh?

The 12K has a dedicated TC input.

JB

Actually I thought the Lanc port was the time code, that was missing in the UM12k. Anyway is this still timecode related the reason for it to be rejected?

Why care about Netflix? I stopped my account with them when they raised their rates last winter. Got Disney now and some free Documentary services. I've shot on film, prism sensor, CMOS Bayer and like the RGBW of the 12K best.

Why care about Netflix?

The main reason is, they're due to spend $18 billion on "content" in 2022, which will be of some small interest to filmmakers.

But you also have to ask, why is Netflix creating public camera qualifying lists which govern private contracts? And why "qualify" $<5000 cameras which no Netflix signatory is going to use as A-cam on a theatrical production anyway? Or detail requirements which could be set aside in a private contract?

Here's guessing the list is a publicity stunt which is why, despite being a joke in in the industry, it doesn't embarrass anyone at Netflix. It's doing its job. And would-be Netflix filmmakers ("I own that camera!") will want to be Netflix subscribers, in the same way the main audience for mostly unwatchable American indie films beloved by Sundance has always been would-be American indie filmmakers.

Why care about Netflix?

The main reason is, they're due to spend $18 billion on "content" in 2022, which will be of some small interest to filmmakers.

Most of their content spend doesn't go to "filmmakers"

JB

Well, the movie business..... But the slate is still enormous:

https://en.wikipedia.org/wiki/List_of_N ... since_2022)#Feature_films

and that doesn't include episodic or foreign. Most of us it's trash of course, but even so....

I used to hate the whole "Netflix Approved" BS, but I've recently come to appreciate certain aspects of what they're trying to do. One advantage of their list is that they provide a handy best practices guide to all their cameras which give some good data points when working on set with various cameras you may or may not be used to.

You can use these guides as a reference for various questions that come up such as "How often do you APR on a Venice?" "How often do you black shade on a Panavision DXL2?" "What Braw compression rates should we use?" "What about on board proxies?" etc. etc. And I've been able to settle many discussions by being able to say "Netflix would recommend doing 'X'."

Over the past years, the Netflix Partners site has become a very good resource for stuff like this.

Well, the movie business..... But the slate is still enormous:

https://en.wikipedia.org/wiki/List_of_N ... since_2022)#Feature_films

and that doesn't include episodic or foreign. Most of us it's trash of course, but even so....

It's a diversion, but let's talk about it.

The headline gigantic number Netflix spend on content seems to be impressive, but it's misleading.

Many Netflix originals have a huge number of "producers". Most of them are only involved with the show peripherally, maybe agents of actors involved, development people that might have had very little to do with the actual production other than having had a hand at the start.

And yet they take take a massive chunk of the budget in fees. It's something like 30-40% of those headline budgets you see.
https://www.imdb.com/title/tt8787802/fu ... _=tt_cl_sm
So a show like this with an Oscar winner has 17 producers. Only a couple of them do the *actual* job of a producer. Most of them just take a margin for somehow being involved in the early days of setting up a show.

Or this massive budget show with 25 producers...

https://www.imdb.com/title/tt2261227/fu ... _=tt_cl_sm

I know crew that worked on this show. At the time it was one of the most highest budget shows. BUT, not a lot of it actually ended up on the screen.

That's because Netflix have created an artificial bubble. They are paying exorbitant fees for talent and IP and still trying to make the show for cheap.

I have friends that have made indie films in Australia. Even though the film is pristine, hasn't had a theatrical run, they will typically offer 35-50K for WORLDWIDE rights.

So unless you have make a film for less than that, you don't get any money and you give up the money you could make by selling individually in territories to have a coveted "Netflix" release.

And now it's starting to dawn on the investment community that the crazy things Neflix have done to the market aren't sustainable. That's why they've had a massive dump in their share price.

They've borrowed insane amounts of money, to pay OVER INFLATED prices on things just to try and buy market share, mostly throwing money at everything and hoping something "good" lands. Stranger Things being one of the few bright sparks for them. Meanwhile, other companies that have been around longer and have more established relationships with actual filmmakers (HBO et al) are continuing to have a higher strike rate for content that people want to actually watch.

Whoops..forgot about Squid Game. They love to claim as their own Netflix Original, but they bought it after the fact too. It apparently only cost 21 million to make. $10 million an episode is a typical higher series budget these days...

https://www.looper.com/651415/how-much- ... m-netflix/

But the show was so successful it's now worth more than a billion to them
https://variety.com/2021/digital/news/s ... 235091156/

I guess that pays for a few other flops...

There will no doubt be further pain and correction. This "Netflix approved" moniker is mostly just a branding exercise, like the 4K requirement to give them a point of difference in the market, to create the prestige they so desperately crave as a brand.

The above linked video gives the appearance of telling you something while actually telling you nothing about how they evaluate cameras.

JB

I have friends that have made indie films in Australia. Even though the film is pristine, hasn't had a theatrical run, they will typically offer 35-50K for WORLDWIDE rights

But it's always been like that. When the Hollywood majors still thought there was money and prestige in running boutique art film divisions, they either produced movies with stars on the cheap (it's art, everybody takes a pay cut!) or offered unknown filmmakers tiny advances, or more often nothing at all, for finished films. Or they'd tell the filmmakers to put the movie in a theater at their own expense, and pay for their own publicity, and then maybe we'll talk. I remember one savvy guy declined a small advance on the condition the studio put that money into publicity. They agreed. But never spent it. The thing died after one weekend.

And even getting "product" for nothing, these companies couldn't make money. I remember an American indie which got rejected by dozens of festivals, large, small, obscure. The director/producer eventually put it into a theater at his own expense, and got the best reviews ever seen for a low-budget non-Hollywood American. He still didn't make his money back.

Meanwhile, Apple TV paid $15 million for a wretched, insufferable, positively emetic Sundance movie, written, directed and starring a kid with one prior indie feature. First letter of the title is "C". You really can't win.... Too many people to either despise or envy. Or despise *and* envy.

But it's always been like that.

Of course you're correct.

The point remains.

Netflix aren't any different and are spending even more money that mostly isn't going on the screen.

JB

In time however, the big screen business will not be the commodities as home viewing of new releases becomes the norm via Netflix, Amazon, and their likes. I am expecting the Sony Production of the world will turn more towards this model than traditional Hollywood filmmaking as the market shifts. Big screen productions will be the exception and to a few. Producers and productions will always find the lease cost investments that yield high returns. You will also see a-listers moving to signup with this new cottage industry that is for now.

BTW, I am not sweating over this Netflix video. That is just propaganda to make people think they have the right stuff. Doesn’t bother me an aota.

There is nothing scientific about their camera list. The 12k should be on it. The Broadcast G2 should be on it.

Both of these are the most interesting cameras that I haven’t used. You could add the BMPCC6K2 as well since I don’t need the internal IRND option. It’s all about the best image quality.

Interesting discoveries while colouring the music festival shot on the original UM4.6K in lossless CinemaDNG Gen 3 Colour Science with the old 4.8 firmware. I can understand if anyone felt it’s a mistake for me to use that out-dated and tragically flawed camera sensor. Years ago I can remember having to deal with the magenta clouds and killer fixed pattern noise sometimes for example.

Well the magenta issue literally evaporated into thin air after a couple of years and in spite of dealing with many varieties of scene lighting with theatrical LED lights, I’ve never had a hint of fixed pattern noise on recent shoots.

But rather than bury the antique camera I’m here to sing the praises of the Colour science. I had assumed it’s a no-brainer that I should update all my Gen 3 clips to Gen 4 in the Camera RAW tab of DaVinci Resolve. And that might be true in many situations, but theatrical lighting can be difficult. Make no assumptions.

I found many CDNG clips did benefit from applying Gen 4; particularly with the medium close beauty shots under the intense spotlights. Gen 4 just provided better sculpting of the images adding depth when Gen 3 looked flat. Very good. But there were also several shots where Gen 4 actually caused severe artifacts and ruined a shot that Gen 3 handled well. Perhaps Gen 5 would have done better but that’s not a CameraRAW option for media shot with Gen 3. And several shots that were medium or wide where my smaller faces were lightened and just looked better staying with Gen 3. Best to colour each clip on its own merits.

So some pleasant surprises from the old camera that remains my workhorse. The producer was pleased and has complimented the quality of my camera gear. As you would understand, no kudos to the operator!

This is now off topic, but Rick have you tried transforming Pocket 4K GEN 4/5 back to URSA 4.6K GEN 3? Your post gave me the idea to try it on the Pocket 4K/6K and the URSA 12K. I was advised not to move the 4.6K GEN 3 forward to GEN 5 but so far no one has advised against going backwards other than it should be common sense.

It does alter some of the colors that I've felt are muted towards the outside of the gamut in GEN 5, greens and violets and has a pleasing effect on the URSA 12K for product imagery.

Ryan, I don’t know how to take my Gen 5 ProRes footage from the BMPCC4K back to Gen 4 or Gen 3 which would match the CinemaDNG footage from the UM4.6K camera. I’m willing to try. Would this be done from the Media Page or in the CST as the Camera RAW tab doesn’t apply to ProRes.

Some ProRes shots looked better on Gen 4 and some looked better on Gen 3 (especially if they had excessive use of those super-saturated LEDs).

Ryan, I don’t know how to take my Gen 5 ProRes footage from the BMPCC4K back to Gen 4 or Gen 3 which would match the CinemaDNG footage from the UM4.6K camera. I’m willing to try. Would this be done from the Media Page or in the CST as the Camera RAW tab doesn’t apply to ProRes.

Some ProRes shots looked better on Gen 4 and some looked better on Gen 3 (especially if they had excessive use of those super-saturated LEDs).

The CST Tool for ProRes from GEN 4 or 5 to 3, and if you have BRAW you can still use the CST too.

I had abandoned the URSA 4.6K with GEN 3 for a while because I wasn't using GEN 4, but recently needed it for an extra angle and was struggling to wrangle it to match. Now going backwards from GEN 5 to GEN 3 it's at least an extra option to explore.

Screenshot 2022-07-13 180112.png (10.66 KiB) Viewed 4593 times

Ryan, I don’t know how to take my Gen 5 ProRes footage from the BMPCC4K back to Gen 4 or Gen 3 which would match the CinemaDNG footage from the UM4.6K camera. I’m willing to try. Would this be done from the Media Page or in the CST as the Camera RAW tab doesn’t apply to ProRes.

Some ProRes shots looked better on Gen 4 and some looked better on Gen 3 (especially if they had excessive use of those super-saturated LEDs).

The CST Tool for ProRes from GEN 4 or 5 to 3, and if you have BRAW you can still use the CST too.

I had abandoned the URSA 4.6K with GEN 3 for a while because I wasn't using GEN 4, but recently needed it for an extra angle and was struggling to wrangle it to match. Now going backwards from GEN 5 to GEN 3 it's at least an extra option to explore.

Screenshot 2022-07-13 180112.png

This is good to know. Thanks.

[CaptainHook]

FYI Gen 3 is camera and white balance dependant, so different transforms are needed depending on the white balance selected on camera (or in the RAW tab) and the space itself unique to each camera. The CST tool assumes 6000K (a Resolve team decision) so the further away you are from that kelvin setting the less accurate your conversion into Gen 3 will be. Whether or not you're happy with it is another thing, but it would not be the same as native Gen 3 support in camera or the RAW tab. Gen 3 gamuts were also only ever developed for the 4K sensor line (Production Camera 4K/URSA Mini 4K/URSA) and 4.6K sensor line/cameras. So there are no Gen 3 gamuts/spaces for any of the pocket cameras, 12K, etc. In other words, YMMV.

Thanks, CaptainHook. I shot the BMPCC4K ProRes and the UM4.6K at 3200 K. That’s a long way from 6000 K.

Thanks, Ryan. I can give the CST a try going back to Gen 3, but I suspect it will be problematic given there is no true BMPCC4K Gen 3 Colour Science option.

At least I’m not mixing CinemaDNG Gen 3/4 (three evening concerts) and ProRes Gen5 (one matinee) in the same video. So the lack of perfectly matching the two codecs won’t be a show-stopper).

Did you try color management? It's meant to bring a variety of sources into "agreement", a common reference point for grading, which is what you're trying to do(?) Set the input color space appropriate for each Prores clip in the media pool, and see how the different cameras play out on the timeline.

Thanks, John. I used ACEScct 1.3, but not too late to use Resolve Colour Management.

Hey Rick, Color Management certainly will work whether with DaVinci or ACES. But, maybe all things don't have to look the same. For example, maybe the grading might differ with differing performance and when all together can create a contribution which adds positively to each performance. Creative grading could be used to mask any technical difficulties.

Jeffrey, thanks; Resolve Colour Management might be interesting since the ACES implementation doesn’t quite cover all the bases. Still I’ve seen some differences that don’t look bad, just different. So those could be my ‘creative’ choices as you suggest.

So interesting that the only Blackmagic Camera on the Netflix approved list is the URSA Mini Pro 4.6K G2. The URSA 12K and the P6K pro didn’t even make the cut.
https://www.bhphotovideo.com/explora/vi ... ma-cameras

Although this may not be of any worries for the folks here, the marketing appeal is huge and producers and entering filmmakers are going to read this and will be mindful for only looking at the options posted here as what’s the safe investment. So that 4.6K G2 is more valuable than the 12K in the eyes and mind of a producer.

The Netflix camera qualification only has any relevance insofar as they are a very important employer or customer of content - and that is not at all insignificant. They have a lot of weight, in this regard, though with their falling subscribers and their persistence with a high subscriber fee, that is not inevitable, forever. Their approved list has no technical nor scientific validity. Only Alan Roberts, who has since retired, conducted proper camera tests, for the BBC and EBU, that I know of. And since he's retired camera manufacturers have exhaled a collective sigh of relief, believe me.

Although this may not be of any worries for the folks here, the marketing appeal is huge and producers and entering filmmakers are going to read this and will be mindful for only looking at the options posted here as what’s the safe investment.

The list probably does affect consumer-level camera sales among aspirational filmmakers, and some technically uninformed producers may insist on a listed camera for non-Netflix productions, but that would be one-off rental house business anyway, where they may well be directed by staff to more credible choices (i.e., Alexa).

It's too obvious to insist on, but in the absence of actual published specifications (agree or disagree with them), the list is meaningless. And the whole question of image quality is all but meaningless anyway. You either have money for production value or you don't, and dozens of cameras are capable of satisfying theatrical requirements -- even forgetting that most movies, if they're seen at all, will be viewed in highly compressed streaming form (and then recompressed by pirates to HD and lower, for the bulk of the worldwide audience). Not exactly a pristine workflow.

Alan Roberts was a great source of objective camera evaluation, although I'm not sure he kept up the times..... The idea that the image is made in post, not right out of the camera, sometimes seemed at odds with his later reviews.

Alan Roberts was a great source of objective camera evaluation, although I'm not sure he kept up the times..... The idea that the image is made in post, not right out of the camera, sometimes seemed at odds with his later reviews.

He very much 'kept up with the times', since science, is not a fashion. He told me that camera manufacturers and broadcasting organisations, simply stopped paying him.

But he was evaluating only broadcast performance. Aesthetics, usually of concern to cinema, don't enter in to it. If he evaluated film, he have to describe the color as inaccurate, wholly unsuited to broadcast. Does that mean you shouldn't shoot film?

There's a lot to be said for objective measurements, but (OTOH) he reviewed the GH5 (favorably) and gave it 14.6(!) stops in HLG and 13.9 for Cine-D, which is way of whack with what other tests report and experience demonstrates.

Despite this positive review, most people here probably wouldn't prefer a GH5 to a commensurately priced BMD camera. And other approaches to testing might lead to other conclusions.

There's no one methodology which will satisfy everyone or answer all questions asked.

But he was evaluating only broadcast performance. Aesthetics, usually of concern to cinema, don't enter in to it. If he evaluated film, he have to describe the color as inaccurate, wholly unsuited to broadcast. Does that mean you shouldn't shoot film?

No he wasn't, he evaluated the objective, scientific merits of camera technology, with proven technical tests. Cinematographic or broadcast are not necessarily different or mutually exclusive when evaluating problems in camera technology. The 'Film look' he espoused at the time was problematic for me, since that was ultimately subjective but then, then, no one was doing that.

Okay, I'll put it this way: his choice of criteria was selective and tailored to broadcast, and the fact that a measurement is objective doesn't make it more significant than a characteristic which wasn't measured or can't be measured.

This is seen most obviously in resolution. The fact that the number is easily quantified doesn't make it worth measuring and doesn't make the camera with the higher number "better" than another.

Alexa color is admired because of its aesthetics (or suggestions of film), not because it's truer to reality than other cameras. And we know that viewers find departures from reality (and analog artifacts) pleasurable. Beyond a certain performance threshold, which many cameras satisfy today, measurable technical distinctions on this order become meaningless for dramatic cinema. I'm not persuaded these kinds of tests mean much today.