BMCC6K Facts and Findings

Low Light Test
No noise reduction or grain added.

Quick battery test with included battery.

- 57 minute record time (6k DCI Q5)
- 26 minutes of operation after battery indicator went red

EDIT: Wasabi 570 battery only went about 43 minutes total record time with 23 minutes being in the red battery indicator

Thank you for the test. What was the screen brightness?

Screen brightness at 50%, recording to an internal 1TB CF Express card.

Is there any sense in a multi-camera comparison in which each camera has a different sensor and inherent differences in the image processing, that the camera footage is only minimally adjusted for exposure illumination? Yes, it does show how they compare in terms of apples-to-apples.

But realistically when you are baking an apple pie with four different varieties of apples, no cook will keep all the ingredients the same. You don’t do that, the results are meaningless. You assemble each pie with different varieties slightly differently to bring out the best in each variety. Seriously think about it, it’s not a silly analogy. Each apple hybrid is painstakingly bred for certain characteristics that must be respected.

It’s exactly the same with these four cameras. You want your camera tests to be adjusted to their own strengths, you want to test in various situations such as low light, normal light, various time of day perhaps, and so on. But in each camera in each scene, you want to go with a particular camera’s strengths as you would of course if you actually owned each camera.

This is a little long winded to make one point: why are the cameras all not shooting with their best recording codecs for example? I’m not shooting with CinemaDNG lossless and BRAW Q0 because I love eating up my precious recording space; the next full day, my post media workflow accommodates that I’m shooting with the two best codecs I have on my two cameras to give me the best shot at my best deliverable. It’s a lot of storage but it’s only for a few hours, not the mission impossible of three or four cameras each with six hours of footage every day of a month where you’re forced to use a more efficient codec.

So please reconsider the purpose of your tests. Every apple needs to be respected. At the end of the day, it may be difficult to tell which apple was used in which pie, because they’re all delicious. Great test result. Doesn’t mean you had to use an Alexa 35, it means, you learned to make a tasty pie with every apple you could buy.

This is a little long winded to make one point: why are the cameras all not shooting with their best recording codecs for example?

Because those margins are inconsequential (and frequently invisible), outside the head of the shooter. The same is true of other distinctions some here claim to see, not least of all because the captured image is not the final one. The difficulty here is, you can't make "scientific" comparisons of image "quality" if you grade each image, but comparing them without grading is meaningless, because it has no real world value -- unless your only production is camera tests. DR and latitude *are* measurable, but the differences are again too small among these cameras to be decisive. Alexa 35 is dandy, but you don't need one to make a movie.

Or consider 16mm or 35mm: in any ten or fifteen year period, countless productions used the same 1 or 2 stocks. And yet everyone's movies looked different, at levels far more obvious than divergences between sensors among the better cameras available today, because production values overwhelm all else. This is even truer where those values are rudimentary. I.e., the last thing people without money should worry about is marginal differences between sensors (or codecs).

We are likley at the stage now (or have been for a while) where image quality is no longer an important factor when choosing a camera within a particular price bracket.

It’s exactly the same with these four cameras. You want your camera tests to be adjusted to their own strengths, you want to test in various situations such as low light, normal light, various time of day perhaps, and so on. But in each camera in each scene, you want to go with a particular camera’s strengths as you would of course if you actually owned each camera.

It is an interesting point. I have often wondered about this same thing. It would seem like the most realistic and helpful way to do it. I suppose that would require the testers to know the strengths or best settings for every camera. More work? Maybe this is why we don't see this more often. Just wondering aloud really.

It’s so impossible to organise tests.

First the access to the gear. The location. The subjects. The lenses. The lighting. The time. It can take days to organise.

Most importantly. The methodology.

Then everyone tells you how you didn’t do it right.

I regularly test for myself at the beginning of shows. It’s easy to spend 100k, that’s what a day costs. How do you pay for that if you’re just being altruistic ?

I’ve stopped sharing mine because every expert out there wants to tell you why you didn’t do it right or that you’re cooking the books somehow.

Entitlement makes it hard to be generous any more.

JB

I hear you John. Don’t think there’s such a thing as an overall good test. Which is why, I think, people can always find something to complain about. Need a well-specified goal that the readers of the test accept.

Way back when I worked at Microsoft they had something called SMART goals for Specific, Measurable, Achievable, Realistic, Time-bound. There’s even a Wikipedia on it: https://en.m.wikipedia.org/wiki/SMART_criteria

Most tests don’t start out with something like this so… oh well!

I was thinking though… would be cool if a group with specific interests discussed what one or more useful tests might be - given their common area of interest. Then members of the group who might want to run tests would have a reference for what the group thinks would be useful. Better than doing tests and then having the group just throw insults after the fact.

Guessing that’s how most of the benchmarking apps got started.

My own testing methods are just what I want to test FOR.

It usually comes from wanting to try something specific for a show with a creative goal or exploration in mind.

Testing for me to is also about how the setup works, will it work in a production environment. How is it to use in the hand with assistants. With workflow.

JB

Cinematographer Sittipong Kongtong from Bangkok, Thailand:

Alexa 35, Alexa mini LF, Red V Raptor XL 8K and the new comer, Blackmagic Cinema Camera 6K FF.
No any color adjustments, I only slightly adjust global exposure to match all 4 cameras. Nothing else."

Are you guys talking about this test? I think it's a great comparison in a high contrast daylight scene with a person in frame and ACES pipeline, tells me a lot (mainly that they are all great cameras with Alexa 35 being exceptional). I prefer these to chart stuff anytime.

Battery testing continued…

Blackmagic has always been quite conservative on their voltage readings with v-mount batteries, but the BMCC6k is VERY conservative.

I’m about 2 hours into a test with an IndiPro Micro 98Wh v-mount and the camera reads 13.2V while the actual battery reads 14.58V at the terminals on my multimeter.

Fully charged, I expected this battery to last about 3 hours, but I think it's safer at about 2.5 hours after testing.

Lastly, after a couple NP-F to 12V adapter plate tests, I can't recommend this solution for the BMCC6k, it seems the power requirements outpace what can reliably be supplied by these plates (Tilta plate and a no-name plate). Stick to stock NPF570 battery or v-mount.

Nontechnical comparison by videographer Anthony Rodriguez (for him, BRAW from the 6K FF is more pleasing than from the mirrorless one (vía external recorder), and is clearly better than S5iix internal H265 recording)

As it happens with many camera tests these days, the key moment of the review, the comparison scenes, are just too few, and they last just a couple of seconds, so you don't really have the chance to experience too much. Most of the time is spent in talking about his preferences and experiences (though his analysis is still informative).

On the live stream announcement, Blackmagic said (I´m paraphrasing here) the inside of the camera has completely re-designed electronics interior, and at 26min 20secs say:

"And it even has room to grow in the future, as we´ve got space in the camera... for colour correctors etc"


What exactly are we expecting here with this? What other things can be added??

I mean the Pocket 4K is basically a new camera from when it was introduced five years ago with the inclusion of Braw, new monitoring tools, and the gyro + gyro stabilisation. Let's hope the Cinema Camera 6k sees as much incremental improvement, but as always: buy for what the camera can do NOW, not what it may do in the future.

Second that. Some things we were hoping for (but were not promised) to be added to the UMP never came.

...as always: buy for what the camera can do NOW, not what it may do in the future.

This is very true!

Just got to test out some ND filters with the CC6k.

I bought two different kinds:
Nisi True Color VND 1-9 stop kit
Firecrest Ultra IRNDs 0.6, 1.2, 2.1 (superslim)

I'm happy to report that I saw no noticeable color shift in any of the filters. I only tested the Nisi to 7 stops, but there was maybe just a tiny bit loss in saturation. Honestly, really happy with both sets, whether I'm running a matte box or just need a VND on the go.

Some more findings/comparisons:

Part of the lure (for me) of the Pocket 6k Pro was that I could show up to a shoot with a lightweight zoom and really have all I needed to film a simple interview and capture some B-roll.
I finally got an electronic EF to L mount adapter (Viltrox) and I borrowed a friend's Tamron 24-70mm f2.8 G2.
It looked really nice on the CC6k! IS and aperture worked fine.

THEN I thought, might as well try the APS-C Canon 17-55mm f2.8 IS in 4k DCI crop...
And the field of view was almost exactly the same as shooting 6k on the Pocket 6k Pro. which brings my sensor readout down to 15.13ms, which is not great, but coupled with the IS in the lens, it's fine for most of the stuff I'm doing right now.

I also weighed the two cameras and they came out to near-exact weight:
CC6k + Viltrox + 17-55mm = 4.6lbs
P6kPro + 17-55mm = 4.7lbs

So the more I play around with the CC6k, the more I like it. The more I justify using it as my 'grab & go' camera.
The Nisi True Color VND looks just fine and doesn't take long to swap to.

And when I want to use my FF cinema glass and 'kit it out', I can run a cage, PL adapter, external monitor, etc.

I´m still interested in the BM6K FF despite a few things mentioned already, as I think it´ll be a great full frame camera for what I need.

But as people are comparing cameras, broadly speaking, what is the next step up in terms of a full frame camera? The FX9 or Kinefinity mavo edge 6K or 8K?

a comparison between the Pocket 6k Pro and the BMCC6k.

a comparison between the Pocket 6k Pro and the BMCC6k.

I downloaded his footage. He prefers color and contrast from the FF 6K over the 6K Pro but I only had to slightly adjust color temperature and tint to get them indistinguishable.

I'm a bit disappointed with the support for actual L mount lenses. So far I'm two for two for being disappointed.

On the one hand there is this issue of EMI induced flickering banding that basically makes the Lumix 16-35 unusable and apparently also affects the Lumix 24-105. As opposed to say the CMOS smearing induced banding that gets all the attention I'm not confident this will be fixed via firmware/postprocessing.

The other lens I have is the Lumix S 50/1.8. That one doesn't seem to have the interference issue but the focus by wire is strangely implemented. It has acceleration by default and also a significant dead zone for low speeds. When rotating the ring slowly the focus doesn't change at all. This doesn't appear to be inherent in the lens. The encoder clearly picks up on slow rotations on my Lumix S5. The S5 also allows configuration of the focus ring behavior to "linear". So I guess this is firmware fixable.

That's a problem on my GH5 and GH5S also, the Pana-Leica and Lumix MFT lenses focus behavior is non-linear in manual.

That's a problem on my GH5 and GH5S also, the Pana-Leica and Lumix MFT lenses focus behavior is non-linear in manual.

That's a different mount though. On the S5 using the exact same lens this can be configured and it doesn't have the odd dead zone (dead speed?) thing that it exhibits on the BMCC.

I´m still interested in the BM6K FF despite a few things mentioned already, as I think it´ll be a great full frame camera for what I need.

But as people are comparing cameras, broadly speaking, what is the next step up in terms of a full frame camera? The FX9 or Kinefinity mavo edge 6K or 8K?

The fx9 will probably be the first to receive an upgrade, as it’s the oldest of the fx series (2019). The fx9 is actually 6K downsampled to 4K. Unless you need s35 mode, which the fx9 has, the fx6 is as good or better in every way and you don’t need an extra unit to do raw.

Unless you need s35 mode, which the fx9 has, the fx6 is as good or better in every way and you don’t need an extra unit to do raw.

Are you certain Keif? The B&H website seems to be saying the fx6 has raw output like fx9 via SDI or HDMI.

Unless you need s35 mode, which the fx9 has, the fx6 is as good or better in every way and you don’t need an extra unit to do raw.

Are you certain Keif? The B&H website seems to be saying the fx6 has raw output like fx9 via SDI or HDMI.

I believe you need to purchase the $2500 extension unit to get raw from the fx9, whereas the fx6 just needs a recorder like Atomos directly into the HDMI port

I know Panasonic cooks their books with the S5 II and its noise reduction but it definitely doesn't have this banding problem (I own one), which is odd

Back to the BMCC6k…

Anyone seeing odd status light behavior when plugged into the AC adapter and charging the OEM battery?

I’m using the battery grip and it’s been charging all three OEM batteries all night - this morning, the status light on top is toggling between amber, red, and off. I don’t think I’ve ever seen my Pocket 4k cameras do that. Anyone else see this?

EDIT: A quick Google search showed others having issues with various BMD cameras and odd amber/yellow status light behavior…but no definitive answers from either anecdotal experience nor BMD support.

EDIT 2: I talked on the phone with BMD support. They are having me test each individual battery, which I have already done. Can anyone confirm that their BMCC6k shows an amber/yellow status light when plugged into AC without the battery installed?

I´m still interested in the BM6K FF despite a few things mentioned already, as I think it´ll be a great full frame camera for what I need.

But as people are comparing cameras, broadly speaking, what is the next step up in terms of a full frame camera? The FX9 or Kinefinity mavo edge 6K or 8K?

The fx9 will probably be the first to receive an upgrade, as it’s the oldest of the fx series (2019).

"The next Sony camera is neither the A7sIV nor the A1II…it’s a video focused model (FX?)"

https://www.sonyalpharumors.com/confirm ... -model-fx/

The 6kFF sensor is noticeably warmer and more saturated colours than the 6kS35. Some prefer warmer looks, I know I do most of the time.
Pushing the temperature to get a warmer and more saturated look is not ideal because you are basically increasing the colour channels and driving up colour noise that why colour filters and jell or lighting is more preferable. Similarly if you want a cooler image if you best avoid pushing the temperature or you get a lots of blue noise.

Basically, pushing the look to cooler colors will always be worse than filtering the light to get the desired look. Like stabilisation, it's always best to combat a problem at the origin.

Electronic cameras are least sensitive in the blue part of the visible spectrum, so that channel is always the noisiest. Since our eyes don't resolve blue as well as green and red, you can often get quite far by separating the channels and applying more NR to the blue channel. You may get rid of most of it without sacrificing visible detail.

The 6kFF sensor is noticeably warmer and more saturated colours than the 6kS35.

I don't think it's more saturated. I get the same scene shot with both cams exactely the same by adjustung temp and tint. Iirc that's making the s35 6K warmer and less magenta. Making it warmer results in an Image that appears more saturated, especially when skin is involved. Adjusting the s35 to match doesn't push noise and the s35 appears to be cleaner anyways.

this morning, the status light on top is toggling between amber, red, and off. I don’t think I’ve ever seen my Pocket 4k cameras do that. Anyone else see this?

This is the behavior I also observed after putting two third party (Patona) batteries into the battery grip. So I chalked it up to them being somewhat used up "foreign" batteries for now. I ordered two BM batteries but they weren't immediately available here. I might report back when those arrive.

The 6kFF sensor is noticeably warmer and more saturated colours than the 6kS35.

I don't think it's more saturated. I get the same scene shot with both cams exactely the same by adjustung temp and tint. Iirc that's making the s35 6K warmer and less magenta. Making it warmer results in an Image that appears more saturated, especially when skin is involved. Adjusting the s35 to match doesn't push noise and the s35 appears to be cleaner anyways.

Adjusting to warmer temperature you are pushing the warm colour channels higher, red, orange and yellow and with it the noise of those colours. It maybe your eyes a filtering the different and you don't notice.

Ideally keep temperature native at 5600k and ajdust with colour lightings or filters.

this morning, the status light on top is toggling between amber, red, and off. I don’t think I’ve ever seen my Pocket 4k cameras do that. Anyone else see this?

This is the behavior I also observed after putting two third party (Patona) batteries into the battery grip. So I chalked it up to them being somewhat used up "foreign" batteries for now. I ordered two BM batteries but they weren't immediately available here. I might report back when those arrive.

My BMCC6k does this even with all three batteries being OEM Blackmagic NP-F570 batteries. But I’m exchanging one as I believe it has an issue. We’ll see…

Adjusting to warmer temperature you are pushing the warm colour channels higher, red, orange and yellow and with it the noise of those colours. It maybe your eyes a filtering the different and you don't notice.

Ideally keep temperature native at 5600k and ajdust with colour lightings or filters.

I really think you are playing 4D chess where you don't need to. You will not notice a difference when adjusting the Pocket 6K to match the BMCC 6k under same conditions unless you pixel peep at 800%...maybe. If anything, the s35 will be cleaner.

You can download samples from both cams from the same scene here and try yourself.

If I remember correctly the Pocket 6K is a bit colder and bit more magenta, that's the only difference.

I really think you are playing 4D chess

If only.... It would be fascinating to know what otherwise immortal movie will be ruined, destroyed, forever blighted, by
small post-production adjustments. Great story, great director, great cast, great production values -- but murdered by the tint knob, thanks to added blue channel noise nobody actually notices.

Meanwhile, some of the same voices endlessly extol the immortal 2013 BMPCC -- with that noisy Fairchild sensor. Go figure. Or better still, don't.

Ideally keep temperature native at 5600k and ajdust with colour lightings or filters.

What leads you to believe that the “native” color temperature is 5600?

Good Luck

I wouldn’t be so sure about that, but it’s definitely not Tungsten for silicon ‚eyes‘.

Shouldn't the color temperature just be metadata for a raw codec anyway? It shouldn't affect the actual raw data. Or is BRAW different in this regard to other raw formats?

WB is metadata in the context that selecting 2000 or 10000 is irrelevant. The actual lighting you choose will have some effect. FWIW the P4K and P6K are allegedly closer to 3200 though well south of the Plankian locus.

viewtopic.php?f=2&t=114403#p630899

Good Luck

[CaptainHook]

As I mentioned in that thread if your goal is to try reduce noise than a wb setting that would result in equal gains for the RGB channels (what some might think is "native") would be less preferable to a light source that would end up lowering the R/B gains (a light that is more in the blue/magenta area). Assuming you wb for something 'neutral'.

I'm also assuming an ideal black body radiator light source with gels or similar. Not necessarily suggesting setting an LED as blue/magenta as possible. As always if this is something one finds really concerning or just interesting, test a bunch and see what works best for you.

For many cases it might be far down the list of things to worry about.

For many cases it might be far down the list of things to worry about.

That sums it up.

I really think you are playing 4D chess where you don't need to. You will not notice a difference when adjusting the Pocket 6K to match the BMCC 6k under same conditions unless you pixel peep at 800%...maybe. If anything, the s35 will be cleaner.

You can download samples from both cams from the same scene here and try yourself.

If I remember correctly the Pocket 6K is a bit colder and bit more magenta, that's the only difference.

Its in the wording you use. If you need to "adjust" you are pushing something from its original native source. Adjusting to "correct" you are basically uncorrecting from native source.

Ideally keep temperature native at 5600k and ajdust with colour lightings or filters.

What leads you to believe that the “native” color temperature is 5600?

Good Luck

What made you think its not?

Research.

Ideally keep temperature native at 5600k and ajdust with colour lightings or filters.

What leads you to believe that the “native” color temperature is 5600?

Good Luck

What made you think its not?

Research.

Good Luck

I am finding more noise in my footage than I thought I would at 1250 ISO. If it's brightly lit, yeah, it's fantastic.
In my quick, run-n-gun shoots, where I'm using available light, I'm trying to figure out what I need to help compensate. (I WAS using an s35 lens in crop mode, which may have exacerbated the noise, for sure)

But even as I'm analyzing some shots from when I was shooting 6k DCI in decent light, there's definitely more noise right off the bat. I feel like 1250 should be a lot cleaner.

Yeah there's a ton of noise in the second gain, to the point where it makes peaking pretty useless. Definitely not a no light camera.

I feel like dynamic range is a lot better in the second gain stage though (when compared to the og 6k), so that's really nice.

I am finding more noise in my footage than I thought I would at 1250 ISO. If it's brightly lit, yeah, it's fantastic.
In my quick, run-n-gun shoots, where I'm using available light, I'm trying to figure out what I need to help compensate. (I WAS using an s35 lens in crop mode, which may have exacerbated the noise, for sure)

But even as I'm analyzing some shots from when I was shooting 6k DCI in decent light, there's definitely more noise right off the bat. I feel like 1250 should be a lot cleaner.

What camera are you coming from? Other BMD cameras?