Blackmagic casually announces 12K URSA Mini Pro Camera

[Super8]

29 minutes ago, TheRenaissanceMan said:

It's RAW. You don't lose any information pulling color casts out of shadows. That's the entire point of the format.

Alexa skews green out of the box. Canon has issues with blue-ish grass, inaccurate reds, and blue fringing around bright highlights (depending which matrix you use). RED has had heavily documented color issues since they've been around, including massive IR problems with all their pre-Dragon sensors.

"Perfect color out of the box" is a myth perpetuated by pretentious gearheads; every camera (and many a lens) has quirks that must be managed. Try using the heaviest internal ND settings on a C200 and not seeing IR pollution!

I understand that you were unable to screw an IR filter on some BM cameras in the past, which bit you in post. That's unfortunate; however, the issue is easily managed with skilled workers who do their homework. 

That's not the point is it?

Quote

Built in Precision ND Filters

URSA Mini Pro features high quality ND filters that let you quickly reduce the amount of light entering the camera. Designed to match the colorimetry and color science of the camera, the 2, 4 and 6 stop filters provide you with additional latitude even under harsh lighting. The IR filters have been designed to filter both optical wavelengths and IR wavelengths evenly to eliminate IR contamination of the images from which many ND filters can suffer. URSA Mini Pro ND filters are true optical filters with a precision mechanism that quickly moves them into place when the ND filter dial is turned. Filter settings can even be displayed as either an ND number, stop reduction or fraction on the LCD!

The above info is from BM website landing page about the ND filters and what they claim to do.

The issue is not can you correct it or not.  But if you go down that route and ignore that BM says their ND filters eliminate IR contamination and don't, you are left with adjusting for IR contamination which can cause other problems depending how severe.

 

Quote

That's unfortunate; however, the issue is easily managed with skilled workers who do their homework. 

I would say Black Magic is lacking in engineers and quality control.  This includes whoever shot and approved that 12K footage that they display.   Looks like they had the same issues that I had, right?

 

42 minutes ago, TheRenaissanceMan said:

It's RAW. You don't lose any information pulling color casts out of shadows. That's the entire point of the format.

Alexa skews green out of the box. Canon has issues with blue-ish grass, inaccurate reds, and blue fringing around bright highlights (depending which matrix you use). RED has had heavily documented color issues since they've been around, including massive IR problems with all their pre-Dragon sensors.

"Perfect color out of the box" is a myth perpetuated by pretentious gearheads; every camera (and many a lens) has quirks that must be managed. Try using the heaviest internal ND settings on a C200 and not seeing IR pollution!

I understand that you were unable to screw an IR filter on some BM cameras in the past, which bit you in post. That's unfortunate; however, the issue is easily managed with skilled workers who do their homework. 

Except BM claims to have a 100% solutions for IR contamination.

This is clearly my issue with false advertising and lack of trust.

[Savannah Miller]

There are 2 types of ND filters you can use to control IR pollution. IRND (full spectrum) and hot mirror filters which have IR cut built into them. Blackmagic does NOT use ND filters with IR cut built into them but rather full spectrum ND filters. What that means is you do get IR pollution when using them since they're not perfect and that's really unnavoidable.

[kye]

7 hours ago, Savannah Miller said:

Something interesting is on linkedin I see quite a few "computer vision" engineers working for blackmagic design in colorado city, colorado. Not sure what they do but it could be related to autofocus. Unless blackmagic has bigger goals

Computer vision is about the camera understanding what it is seeing.

From wikipedia:

"Computer vision is an interdisciplinary scientific field that deals with how computers can gain high-level understanding from digital images or videos. From the perspective of engineering, it seeks to understand and automate tasks that the human visual system can do."

It could be about AF, in the sense of choosing what to focus on.

In a sense, the logic of PDAF is something like this:

1. Look at all the focus pixels on the screen and go towards the middle / closer objects
2. One things are broadly in focus, employ face / animal recognition to identify faces
3. Identify faces and rank which ones are the most important, and where to focus to (eg, could be one face or many)
4. Use PDAF to focus on that face / focus the best on all target faces

The problem with modern PDAF focus systems is that they go wrong because they're focusing on the wrong thing, not having problems with focusing itself.

I suspect that Panasonic with its DFD system it hoping to 'out-smart' the PDAF systems by being able to intelligently work out the scene using some kind of AI.  For example, if you see a blurry human-shaped blob that is sitting on top of the rest of the image, it's pretty easy to conclude that it's a person (by the shape), relatively how far away from the focal plane it is (how out of focus it is and where the focal plane is now), and that it's closer than the focal point rather than behind (if other things are more in focus but this isn't), and that it's the closest object (because it's over the top of everything else).  If it's also the biggest and completely in frame then it would be reasonable to assume that might be the focus of the shot too.

I suspect that this is how human vision works, in a way.  These things are pretty straight forwards.

PDAF is only good for telling if a given pixel is out-of-focus to the front or back, and by how much.  At the point when you have an AI engine that can look at an image and understand it, blurry things and all, then there is no advantage to having PDAF.  Our eyes don't have PDAF, we have killer AI, and out eyes are pretty good at focusing on objects and basically never get things wrong. 

I suspect that that'll be what Panasonic is doing with it's computer vision people.  If they get it right and get it to fulfil the promise of the technology, then it will out-perform other systems because it will understand a scene even when everything is out of focus.  Of course, they're a long way from that being market ready at the moment.

[Andrew Reid]

9 hours ago, androidlad said:

BMD patent document on the URSA 12K sensor design:

https://patentimages.storage.googleapis.com/88/08/39/428dbbc9e5dfca/US20190306472A1.pdf

 

Actual CFA pattern:

 

In the document, it specifically discusses colour-aware pixel-binning to increase sensitivity and double framerate, at the expense of lower resolutions. So the "oversampling" mentioned in marketing is without doubt BS, at least for full FOV framerates higher than 60fps.

Because if it oversamples from full 12K, then the oversampled modes cannot exceed 60fps which is the full readout max fps.

Maybe there's a feature similar to R5's "4K HQ" mode.

BS maybe.

"In-sensor scaling" = pixel binning or oversampling...That is the question. Because binning, skipping, implies you are not sampling all the data, but with averaging and combining data from all photosites, the story is very different.

I am curious how the 12K RAW stuff does not infringe on RED's patents BTW.

Changing the CFA is enough?

I doubt that.

16 hours ago, Laurier said:

Yet you still hear people complain about the Alexa green cast, the Red yellow wash and the pinkish skin from canon colours.

Put a Zeiss lens you get a blue cast, put a canon or a Cooke you get yellow, but a sigma you loose some red/yellow ect...

Shoot on film, you get several colour cast depending which lab is going to develop .

If the image is striking , no one is complaining about the sky being the wrong shade of blue.

Once you cut through the marketing BS and learn how to use a grading suite, none of that matter.

Well put.

[Logan]

On 7/19/2020 at 8:12 PM, hijodeibn said:

my gosh...who threw peanuts to the monkeys?

lol, this clown is going through my post history and downvoting everything. What a cesspool.

[Andrew Reid]

@Super8 final warning. Behave or it's over. Full IP address level blocking and ban.

Topic after topic ending in You vs 10 others.

[Super8]

16 hours ago, Laurier said:

If the image is striking , no one is complaining about the sky being the wrong shade of blue.

The 12K image did have alot appeal so we'll see.

Quote

 

[Andrew Reid]

[Savannah Miller]

What patents does RED have that blackmagic would infringe upon? BRAW uses in camera partial demosaicing so the raw format they record gets around RED;s patent. Unless there are other patents I am unaware of

[steven abrams]

15 hours ago, androidlad said:

BMD patent document on the URSA 12K sensor design:

https://patentimages.storage.googleapis.com/88/08/39/428dbbc9e5dfca/US20190306472A1.pdf

 

Actual CFA pattern:

 

In the document, it specifically discusses colour-aware pixel-binning to increase sensitivity and double framerate, at the expense of lower resolutions. So the "oversampling" mentioned in marketing is without doubt BS, at least for full FOV framerates higher than 60fps.

Because if it oversamples from full 12K, then the oversampled modes cannot exceed 60fps which is the full readout max fps.

Maybe there's a feature similar to R5's "4K HQ" mode.

The cfa you show is figure 5, but the patent says the cfa 'invention' is figure 6 (a different one). Others have been posting figure 6 as well, I think you're wrong here about the cfa.

[androidlad]

8 hours ago, Andrew Reid said:

BS maybe.

"In-sensor scaling" = pixel binning or oversampling...That is the question. Because binning, skipping, implies you are not sampling all the data, but with averaging and combining data from all photosites, the story is very different.

I am curious how the 12K RAW stuff does not infringe on RED's patents BTW.

Changing the CFA is enough?

I doubt that.

Well put.

[cameraeye]

14 hours ago, Logan said:

lol, this clown is going through my post history and downvoting everything. What a cesspool.

Yeah someone did that to me too. It’s pathetic. I didn’t know 3 year olds were allowed to create accounts on here.

[majoraxis]

8 hours ago, androidlad said:

In the Camera RAW settings in Resolve, you can do the subsampling from 12k to 8k, 4k or even 2k. Is Blackmagic doing Oversampling in the Resolve and in camera it is a different type of math? We may see a quality difference when comparing the two methods (in Camera and in Resolve).

Subsampling to 2k looks to be straight averaging so that should look about the same in camera or in Resolve.

Because it is not an even ratio of RGB pixels when going from 12k to 4k, the subsampling conversion should benefit from doing advanced math to get the most out of the subsampling process as you are turning a 36 pixel 6 x 6 group of pixels in to 4 pixels. There is a lot of data to work with and maybe simple math, if that’s all that is available in camera, won’t be a big difference when going from 12K to 4k in camera vs in Resolve, if the math is different in camera vs in Resolve.

For the 12k to 8k subsampling coversion it is a 2.25 ratio so you are going from 36 to 16 pixels. I would expect to see the most difference between capturing at 12k and subsampling to 8k in Resolve vs subsampling to 8k in camera and creating a 8k BRAW file, if the math is different in camera vs in Resolve.

If the subsampling math is simpler in camera, I expect the 12k to 8k in Resolve subsampling to look better. It is possible that they may look very similar now and eventually look better in Resolve when higher math options are available. Testing and time will tell.

 

[BenEricson]

On 7/20/2020 at 9:17 AM, TheRenaissanceMan said:

It's RAW. You don't lose any information pulling color casts out of shadows. That's the entire point of the format.

Alexa skews green out of the box. Canon has issues with blue-ish grass, inaccurate reds, and blue fringing around bright highlights (depending which matrix you use). RED has had heavily documented color issues since they've been around, including massive IR problems with all their pre-Dragon sensors.

"Perfect color out of the box" is a myth perpetuated by pretentious gearheads; every camera (and many a lens) has quirks that must be managed. Try using the heaviest internal ND settings on a C200 and not seeing IR pollution!

I understand that you were unable to screw an IR filter on some BM cameras in the past, which bit you in post. That's unfortunate; however, the issue is easily managed with skilled workers who do their homework. 

That's not how IR pollution works though. Raw might help you slightly, but you'll spend hours applying masks to the effected areas. 

I'm not sure BM ever even officially announced the IR pollution issues with the bmpcc or bmmcc, even though the camera is basically unusable without one.

For what it's worth, I use the C300 Mk2s constantly and have never seen a trace of IR pollution. The baked in color profiles on a Canon C series camera are pretty damn close to "perfect color out of the box." 

[Savannah Miller]

52 minutes ago, BenEricson said:

That's not how IR pollution works though. Raw might help you slightly, but you'll spend hours applying masks to the effected areas. 

I'm not sure BM ever even officially announced the IR pollution issues with the bmpcc or bmmcc, even though the camera is basically unusable without one.

For what it's worth, I use the C300 Mk2s constantly and have never seen a trace of IR pollution. The baked in color profiles on a Canon C series camera are pretty damn close to "perfect color out of the box." 

That's because there's two ways to handle IR Pollution. Hard IR cut or gentle. Gentle gives you an more accurate color at the expense of needing to use IRND filters.  ARRI uses gentle IR cut and so did RED up until their newer cameras. Blackmagic used gentle IR cut but it's a little too weak so you sometimes may see IR cut without even using ND filters.  Blackmagic is either lazy or for whatever reason they haven't bothered to source a better IR filter. Because blackmagic uses a softer IR cut you end up having to screw on your own which defeats the purpose of using smooth IR cut in the first place. A lot of blackmagic's early camera choices were inferior copies of what ARRI was doing and a few of them have stuck throughout all their cameras.

[TheRenaissanceMan]

1 hour ago, BenEricson said:

That's not how IR pollution works though. Raw might help you slightly, but you'll spend hours applying masks to the effected areas. 

I'm not sure BM ever even officially announced the IR pollution issues with the bmpcc or bmmcc, even though the camera is basically unusable without one.

For what it's worth, I use the C300 Mk2s constantly and have never seen a trace of IR pollution. The baked in color profiles on a Canon C series camera are pretty damn close to "perfect color out of the box." 

I wasn't saying you should fix IR in post as a rule--just that this specific instance didn't seem to lose much. 

My experience with Canon has been mostly good, especially in regards to pleasing color.  However, the extended ND settings on the C200 invariably needed an IR cut to decontaminate the shadows. On the 2 commercial projects I've shot with newer Canons, I've run into issues with cycs lit by blue LEDs turning purple. Turns out I'm not the only one--there's a whole topic about it on DVX User (the answer from Canon: "the camera is operating within spec" lmao). And while rather pleasing at first glance, the color isn't what you'd call accurate. It took a ton of secondary work in post to get the company colors dead on without fucking up everything else. 

All I'm really driving at is that every camera (and brand) has its issues to mitigate, be it form factor, color accuracy, IR, codec, sensitivity, etc. In the scheme of things, considering what the 12K offers, remembering to use some IR cut is pretty damn manageable.

 

At least for my kind of work. Ymmv.

[Savannah Miller]

4 hours ago, TheRenaissanceMan said:

I wasn't saying you should fix IR in post as a rule--just that this specific instance didn't seem to lose much. 

My experience with Canon has been mostly good, especially in regards to pleasing color.  However, the extended ND settings on the C200 invariably needed an IR cut to decontaminate the shadows. On the 2 commercial projects I've shot with newer Canons, I've run into issues with cycs lit by blue LEDs turning purple. Turns out I'm not the only one--there's a whole topic about it on DVX User (the answer from Canon: "the camera is operating within spec" lmao). And while rather pleasing at first glance, the color isn't what you'd call accurate. It took a ton of secondary work in post to get the company colors dead on without fucking up everything else. 

All I'm really driving at is that every camera (and brand) has its issues to mitigate, be it form factor, color accuracy, IR, codec, sensitivity, etc. In the scheme of things, considering what the 12K offers, remembering to use some IR cut is pretty damn manageable.

 

At least for my kind of work. Ymmv.

I think what would be even better than all of this is if someone could design a replacement IR cut for the blackmagic camera. John Brawley uses rawlite OLPFs in his cameras but he mainly benefits from the IR cut it provides. I would LOVE to see a replacement IR cut for blackmagic which has slightly better IR cut without being harsh. Just enough IR cut that you don't see any when using the internal ND or when shooting under tungsten

[ade towell]

John Brawley has said they are still tweaking the IR filter and it will be a different one on the final camera release

From.dvxuser.com

IR filtering is still being tweaked. My comments wouldn’t mean much because I had a prototype camera that had a different IR filter than what will ship so that’s a bit difficult to draw conclusions from.

I used all internal ND’s on my shoot. There was also a calibration difference with my camera that wasn’t picked up till recently. Most of the RAW files are shot at the default WB of 5600K +10, which is the “standard” BMD daylight default.

BUT

When you start to go into he RAW tab you’ll find it’s really about -500K to -1000K (Less warm) and -5 to -10 tint (less magenta)

I didn’t change these on my initial clips, but I have on these ones...

https://vimeo.com/440479199

Which also by the way show the 8K and 4K modes 🙂

[majoraxis]

The footage is impressive! The more I see, the more I am sold on how special the image is.

[Savannah Miller]

2 hours ago, ade towell said:

John Brawley has said they are still tweaking the IR filter and it will be a different one on the final camera release

From.dvxuser.com

IR filtering is still being tweaked. My comments wouldn’t mean much because I had a prototype camera that had a different IR filter than what will ship so that’s a bit difficult to draw conclusions from.

I used all internal ND’s on my shoot. There was also a calibration difference with my camera that wasn’t picked up till recently. Most of the RAW files are shot at the default WB of 5600K +10, which is the “standard” BMD daylight default.

BUT

When you start to go into he RAW tab you’ll find it’s really about -500K to -1000K (Less warm) and -5 to -10 tint (less magenta)

I didn’t change these on my initial clips, but I have on these ones...

https://vimeo.com/440479199

Which also by the way show the 8K and 4K modes 🙂

That is the best news ever. I believe blackmagic has used the same IR filter in every camera they've ever put out so this time they might have strong enough IR filtration for once.