Blackmagic Pocket Cinema Camera 4K Hands-On Review

October 30th, 2018 icon / message-square 43

In this first-look hands-on review with the long-awaited Blackmagic Pocket Cinema Camera 4K, we took a close look at all aspects of the camera: Apart from the quality of the footage itself, we judged ergonomics, handling, design as well as low-light and dynamic range performance. Watch above and read below to learn more!

The Unicorn Is Here

Ever since its announcement at NAB in April this year, the Blackmagic Pocket Cinema Camera 4k has become the most-anticipated camera release of the year. We shot an exclusive interview with Blackmagic Design founder Grant Petty in Las Vegas, where he answered all of our questions about the newly announced device. Blackmagic promised to start shipping the camera in September – and for once, they somewhat managed to keep that promise. Very small numbers of the BMPCC 4K started being shipped out at the end of September, however they are still nowhere close to meeting demand. It seems like production has been somewhat slow, but at least by having had a press loaner for about two weeks we can confirm that it works, and it works really well.

Size, Monitor, Handling

What’s most striking about the new Pocket 4K is the fact that you need very big pockets. Just like Blackmagic’s URSA Mini was never “mini”, the Pocket 4K doesn’t fit in a pocket. Compared to its predecessor, the very popular Blackmagic Pocket Camera camera, the BMPCC 4K is very large. However, that is thanks to a gorgeous 5 inch multitouch display at the back of the camera. It’s 1080p and brilliant in every way, particularly for a camera that costs so little. It’s super easy to judge focus and the controls are very responsive. What’s missing is a viewfinder, which means loupe adapters like Zacuto’s Z-Finder might have a renaissance. However, as you are required to use the screen with your hands, such a loupe might prove impractical because you will have to take it off constantly in order to be able to control the menus and change most settings. So that only leaves you with adding an external viewfinder to this camera when working out in sunshine (where you really can’t read the screen well).

While the screen is really nice and very responsive to any input, it’s unfortunate that Blackmagic decided against a floppy screen or even just a tilt screen – especially because there is no viewfinder as an alternative whatsoever. Very often you find yourself bent over backwards with this camera when you are shooting with it alone without assistants.

The BMPCC 4K feels good in the hand when you have large hands like me, but I know that some other people will have a problem with the hefty grip of the camera.

There are two microphones left and right of the lens mount which should in theory bring nice stereo sound, however you find yourself touching these microphones all the time when holding the camera, because they are not ideally placed. However if you are aware of it, it’s of course much easier to avoid doing when filming.

In terms of menu design, Blackmagic Design is still my personal champion … the menu is so straightforward and easy-to-use that you really never need to touch a manual or anything like that. When you see camera menus like this, you start wondering why all the other (mostly Japanese) camera manufacturers never managed to make straightforward menus where you can actually find stuff more easily.

Battery Life

The BMPCC 4K uses Canon LP-E6 type batteries. The same standard as the 5D Mark II and III, and many many other popular Canon cameras. This is a nice move since it means that you can continue using your old batteries.

However, the power management of the camera seems to be quite poor. At random times below 20% charge, the camera actually simply shuts off without warning. Also, a battery also typically only lasts 35-45 minutes, which isn’t long at all.

We recommend using an external power solution like for example the Core SWX Powerbase EDGE with a dummy battery, which can last you many hours, if not a full day! It makes your footprint with the camera bit bigger, but also more powerful and reliable.

Core SWX Powerbase EDGE Battery for Blackmagic Design Pocket Cinema Camera 4k

Sensor and Sensitivity

The BMPCC 4K comes with a MFT sensor that is supposed to be quite light sensitive, with a dual ISO of 400 and 3200. In my video test I could verify that these two ISO values were indeed the two with the least amount of noise in the picture. If you are aiming for high quality images, I would not shoot beyond ISO 6400, however I think it’s totally acceptable to do so for documentary or other reality content. The noise of the sensor is quite pleasant and should be fairly easy to get rid of using de-noise in DaVince Resolve (which, by the way, comes for free with the camera in its Studio version – the normal version is free anyhow).

This camera is miles better than the original Pocket camera when it comes to low light, and it’s a lot better than many others in the market. During the shoot of the intro of this review, we were losing light rapidly as all of this was shot after sunset in the woods. With a 25mm MFT f/1.2 lens from Olympus, I was able to get exposure by increasing the ISO progressively, with the lens always wide open. As you can see above, even the ISO 16,000 shots are tolerable, albeit noisy.

Codecs, Resolutions and Frame Rates

What sets Blackmagic cameras apart is the fact that they record RAW and various flavours of ProRes, as opposed to highly compressed proprietary codecs like most other manufacturers. That means 12-bit RAW recorded internally in a camera, and that is a big deal – especially at this price point.

The Pocket 4K cannot record in Blackmagic RAW yet (we have reported about it in an interview with Blackmagic here), but it records RAW in DNG sequences (which is of course data intense!) However, the RAW images end up being 12bit, while ProRes HQ is only 10-bit (which is still much better than what 90% of professional cameras in the market can do).

What’s impressive is the capability to record up to 60fps in full quality in 4K. Only when you want to go even higher with your frame rates, you have to switch to 1080p and activate a dedicated “sensor crop” button. Under that sensor crop, you are then able to record 120 frames per second.

Dynamic Range Tests and Questions …

Our technical expert Gunther Machu ran a couple of cinema5D Lab Tests with the BMPCC 4K before it had to go back to the manufacturer. Before you read on, in case you haven’t read it yet, please look at our recent article explaining the methodology of our lab tests – here’s the link.

The result: 11.6 stops of dynamic range for a signal to noise ratio of 2 (12.7 stops for a signal to noise ratio of 1) at ISO 400 (10.5 and 11.8 stops for ISO3200), in ProRes 4K DCI 25p.

Interestingly, when we tested the same DCI 4K 25p RAW (1:1) we found that the dynamic range reading was lower (by about a half stop). Only when we selected the “highlight recovery” option in DaVinci Resolve and then exported frame grabs into IMATEST we got similar results to ProRes.

This is a bit strange, because we tested also the old Pocket Cinema Camera and it had the same dynamic range reading in ProRes and RAW – and when we enabled the “highlight recovery” option in Resolve, we actually got about 0.6 stops higher DR in RAW that with ProRes.

How come that the DR of the new Pocket 4K is lower in RAW, and only matched ProRes when “highlight recovery” is selected in Resolve? Is ProRes doing this HL recovery automatically?

We reached out to Blackmagic Design, and received a reply very promptly:

Reply from Andy Buckland, Blackmagic Design (after a call with Gunther Machu):

The Pocket Cinema Camera 4k uses our new demosaic algorithm developed with Blackmagic RAW, before encoding to ProRes. This new algorithm helps to reduce noise and improve the dynamic range in ProRes capture. With Cinema DNG RAW, the demosaic is still performed in DaVinci Resolve. That enables full control and manipulation over debayer settings, noise reduction, and highlight recovery to increase the dynamic range measurement beyond that of ProRes.
 
By applying minimal temporal noise reduction to the Cinema DNG RAW files in DaVinci Resolve – the lower steps on the waveform can be cleaned up nicely, bringing the available dynamic range of RAW in line with those seen in your ProRes tests. Then by enabling highlight recovery, the available dynamic range in Cinema DNG RAW increases beyond that of ProRes.
 
ProRes users benefit from the new debayer algorithm in terms of dynamic range and reduced noise at capture, resulting in a faster turnaround.
RAW users benefit from the full control and manipulation of the image, to decide which parameters they would like to tweak and finesse.

Conclusion

There are many more things to say about the Blackmagic Pocket Cinema Camera 4k, but please do watch the video if you haven’t seen it yet, as many things are covered there in more detail. Blackmagic really did their homework with this one and listened to as many customers as possible when designing the successor to their original Pocket Camera. The image out of it is beautiful and “cinematic”, the dynamic range is better than the one of similar cameras in the market right now, and it’s just very enjoyable shooting with it. It’s not a perfect camera of course, but it has a lot to offer. I can’t wait to took a much closer look at the camera soon.
The music used in the video above is courtesy of Music Vine. Get 25% off your next music license with code C5D25 (valid for one use per customer).
What do you think about the BMPCC 4K? Did you pre-order one, and are you still waiting for it? Please let us know in the comments below. 

43
Leave a Reply

guest
1
1
- 3
Filter:
all
Sort by:
latest
Darek
Darek
Guest
November 10th, 2020

How long blackmagic pocket 6K can work continuously without a break?

lrjjoy
Guest
March 3rd, 2019

I think this the best camera

MrJACK
MrJACK
Guest
December 19th, 2018

Does it crop in HD 60fps ? or only in 120 fps?

Alf
Alf
Guest
December 18th, 2018

What lenses would you recommend for music video shooting? I know that’s super vague, it’s a gift for someone. But I was looking at MFT mount 25mm lenses, any $700 or less recommendations? Thanks –

 Tony McGuire
Member
December 17th, 2018

Are some of you people commenting like Panasonic whores? I don’t get it, these people have put in HOURS of work and you come in and trash their work.

The GH5 is not the camera of G-d, it’s a tool, and it’s very limited just as many DSLRs are, I have a Sony F3 that KILLS the A7S2 image and the GH5 and it’s only 1080p, I could scale it up very easily with a smart scaler and it would be very usable and more pleasant.

I have a RED MX that still holds up and an A6300 that when lit properly can work with an ARRI for short smart cuts.

Please STOP with the brand whoring. They’re cameras made by corporations who in the end, don’t care about “Tom Roper” or “Winthrop Holstrom,” and are trying to make as much money as possible, with the exception of RED that lately has taken a severe turn to that side.

Aside from that, thanks for the review, I can interpret my own graphs and I really thought I’d see 12.5ish to 13.5ish on this camera but as usual, the manufacturer overestimated it.

Rod Compton
Guest
December 12th, 2018

All I want to know is can I plug an electronic eyepiece viewfinder to shoot with completely manual focus, manual aperture, 100mm macro, 200mm macro, 300mm and 500 mm long lenses for wildlife – with suitable adapters of course http://www.londonbutterflysurvey.org.uk

Dave
Dave
Guest
November 4th, 2018

Not an video expert, just a hobbyist. Here’s where I wonder about your results. Your entire test depends on the software to calculate the signal to noise ratio, but that itself depends on what the camera is doing or not doing to denoise the image internally. So if camera A has superior denoising processing but camera B has no denoising, your results will not reflect the true sensor’s dynamic range. So this is why I don’t think your results are 100% useful or true. Also if one does use a denoiser in post, you can recover some dynamic range which the software rejects.

Member
November 3rd, 2018

Thank you for the article C5D! Have the URSA Mini 4K and 4.6K been reevaluated since the Dynamic Range Test reboot? I’m curious to see how the then 8.5 and ~12 stops of DR compare to the 11.6 of the BMPCC 4K.

John B.Cummings
Guest
November 2nd, 2018

Yikes. Didn’t see anything in that initial review to make me want this camera. But I did notice Gunther even has chip charts on his glasses. That’s dedication!

Leo Hynes
Leo Hynes
Guest
October 31st, 2018

I’d just like to say what a great job you’ve done here Nino and Gunther. You’re DR tests are invaluable to me and many and I hope you continue the great work, despite all the people getting quite sensitive over dynamic range!

andi
andi
Guest
October 31st, 2018

@Gunther Machu are you ever tested the Dynamic Range based on Post Production @ Color Grading ? I think this area are interesting

Member
October 31st, 2018
Reply to  andi

Hi Andi, so far not – while I find it very interesting too, there are so many more factors coming into play, that it will be very difficult to control them all – hence to be fair in a comparison with other cameras as well.

 Mercy Warhol
Mercy Warhol
Member
October 30th, 2018

Pre-ordered with Hot Rod on the 9th of April, still don’t have my camera. People who ordered with Adorama in August are getting theirs. Lesson learned.

Scott Armstrong
Scott Armstrong
Guest
November 2nd, 2018
Reply to  Mercy Warhol

I ordered mine on June 30th from B&H. I got an email two days ago saying that the camera I ordered is shipped and should arrive on Monday. I was late in my Pre-Order. When I ordered the camera it was already listed as a best seller before it even shipped. So I knew I would have to wait until my turn to get it. Rumors seemed to indicate that they would really be shipping in early November. I’m excited to see what the camera can do. I purchased it because of the great reviews, features and the price.
I have an old 14x B4 video camera lens and a FotoDiox adapter I plan on using with the camera. I’m excited to see how it all works with this combination.

mike
mike
Guest
October 30th, 2018

the BMPCC4k’s shot is overexposed in the DR test. you can see glow around the light square.

Max Mustermann
Guest
October 30th, 2018
Reply to  mike

Scroll up and read the answer Gunther Machu gave for “German Polozov”.

Arnold Finkelstein
Guest
October 30th, 2018

Sorry, you’re using the word literally the wrong way. In fact, you’re using it in exactly the opposite way it’s supposed to be used. You can’t say the camera literally ate batteries because you mean it did so figuratively. You can’t use the word that way – it’s very, very wrong.

Ballard Jones
Guest
October 30th, 2018

The article mentions the “price point.” Has Blackmagic released it’s price point and demand curve for this camera or are you using “price point” incorrectly? Do you mean price? If so, here’s some great news: you can just say price. It takes less time and you’ll be correct.

Jay Wolff
Jay Wolff
Guest
November 2nd, 2018
Reply to  Ballard Jones

Ballard, you can also say “its” when you mean “belonging to it” instead of “it’s” (contraction of “it is”) and you’ll be correct.

Tim Erickson
Tim Erickson
Guest
June 22nd, 2019
Reply to  Ballard Jones

Holy pedantry, batman!

Eric St-Martin
Guest
October 30th, 2018

Good job Nino.

Tom Roper
Tom Roper
Guest
October 30th, 2018

GH5s has 1.1 stop more DR in HLG recording mode than the tested V-Log/L according to EBU tech 3335 (https://tech.ebu.ch/docs/tech/tech3335_s29.pdf) which also uses Imatest, doesn’t clip the red channel prematurely as P4K, doesn’t have trick like “highlight recovery.” And why the first two usable stops on GH5s and A7SII so much less bright, a full stop shy of clipping? Gamma curve matters to DR and HLG should have been used as this was known, V-Log/L which doesn’t write above code value 768 is the wrong choice. Similar observations for XT3, A7SII, Appreciate the effort Nino, but still a ways to go for consistency, and this was pointed out to Gunther last week before this review because it looked then like a conclusion could be forthcoming not really supported by impartial testing. In other words, efforts made to extract a better number from P4K through “highlight recovery” and not seeking advice and assistance from Sony or Panasonic as was made of Black Magic Design. It’s in small details like these that questions of impartiality arise when the intent may have been to do the opposite, to address the questions that were spawned by the EBU testing on GH5s. In light of this, the EBU tech 3335 method comes out looking more objective and impartial.

Max Mustermann
Guest
October 30th, 2018
Reply to  Tom Roper

I don´t think it is important if you measure 11 or 12 stops or whatever. The good thing about the Cinema5D Setup is that they can compare cameras. You can see the differecne between camera A and Camera B. When Cam A is 11 and Cam B is 12. I know the difference is 1 stop. And that is what is important to know.

Tom Roper
Tom Roper
Guest
October 30th, 2018
Reply to  Max Mustermann

The difference is 1 stop from V-Log/L but that gamma doesn’t yield the most DR from the GH5s. That’s the problem with the comparison. CineLikeD and HLG profiles both yielded more DR for EBU than V-Log/L, the latter by more than a full stop.

Member
October 30th, 2018
Reply to  Tom Roper

Hi Tom, the GH5s VLOG result is what it is – the result when the VLOG Gamma is used. It may or may not be different when e.g. the HLG is used – if we find time we will retest the GH5s with HLG. Regarding the EBU tech 3335 method, of you read our article about our methodology of testing carefully and the compare to the EBU tech result for the GH5s, I am quite stunned that we are getting very very similar results. In the EBU tech article it is stated that the max (absolute) DR for the GH5s is 14.7 stops – our results in IMATEST show a maximum of discernible patches of 14.7 stops (irrespective of the fact that the image is usable or not).

Winthrop Holstrom
Winthrop Holstrom
Guest
October 31st, 2018
Reply to  Gunther Machu

GH5S native ISO for V-Log is 800 ISO, not 400 ISO. The wrong ISO was used to test the dynamic range of the GH5S. Panasonic makes it quite clear that for maximizing dynamic range, the native ISO for the GH5S should be set to 800 ISO.

Winthrop Holstrom
Winthrop Holstrom
Guest
October 31st, 2018

Just to clarify, for V-Log on GH5S for highest dynamic range use 800 ISO. For GH5S in V-Log for lowest noise use 5000 ISO. Those are the numbers provided by Panasonic. The native ISO values are different when not using V-Log, whereas it would be 400 ISO and 2500 ISO respectively. So please retest GH5S using correct 800 ISO for V-Log.
Thanks again.

Winthrop Holstrom
Winthrop Holstrom
Guest
November 1st, 2018

Nothing Gunter?
Hmm… no reply.

Winthrop Holstrom
Winthrop Holstrom
Guest
November 1st, 2018

Sorry,. Gunther.

 Colin Elves
Member
November 3rd, 2018

Setting the ISO to 400 rather than 800 just pulls the entire curve down. You can see this has happened from the wedge chart above as instead of clipping at around 80 IRE it’s more like 70 IRE.

In all likelihood they’ve just underestimated the DR by about a stop. So the GH5S probably has the same, if not higher, DR than the pocket 4K.

I’m sure Cinema 5D will edit the article above I acknowledge this mistake, as it would be completely unethical to fail to do so and I’m sure they have more integrity than that.

Micah
Micah
Guest
July 16th, 2020
Reply to  Tom Roper

Just never add any contrast to any image so you can make sure you have the absolute most DR. In fact just de-contrast it, boom another “3 stops” of DR. Get outta a here, its ridiculous to see people crying because they don’t have the best camera. Go make a video, if you own any camera today you already own something great and capable of producing great images that werent even fathomable 7 years ago.. Try making something good with an old camcorder or a gl2 and a G3 power pc. Both the GH5 and BMPCC 4k cameras are above your capabilities.

Max Mustermann
Guest
October 30th, 2018

Wondering why the DR Image is blurry on one image. And why the strange white balance? And at Sony A7S the “left squares” are still gray and not blown out. Also Gh5s. I am not saying that the test isn´t accurate. I think the numbers are fine. But the image is a bit confusing.

Max Mustermann
Guest
October 30th, 2018
Reply to  Max Mustermann

Another thing. Would it make sense to have “real life test footage” as well. And under and overexposure the image and try to get all the details back in post. I´ve seen more and more professional YouTubers are doing the plus one, minus one, plus two stops, minus two stops test and I think it is really good to measure DR, too.

Member
October 30th, 2018
Reply to  Max Mustermann

Hi Max, please look at my reply to German above.

Max Mustermann
Guest
October 30th, 2018
Reply to  Gunther Machu

Thank you!

 Henry Eckstein
Henry Eckstein
Member
October 31st, 2018
Reply to  Max Mustermann

The cause of the blurry image on the LEFT side can ALSO be due to a high SLEW rate (an ultra fast rise or drop in signal difference) which means the image area being photographed has TOO MUCH CONTRAST, having a very sharp difference between black (or dark) and white (or bright) pixels which the camera cannot resolve fast enough.

This results in adjacent pixels of a very high contrast ratio being excessively ‘smoothed’ and blurred by the digital signal processor on the camera. The LESS blurriness between areas of black and white during the left-to-right scan of the camera’s pixel readout means the BETTER the DSP processor which then means you get BETTER quality image!

On terms of overall quality, the Black Magic camera is ABSOLUTELY FINE for documentary and low-budget cinematography! You just can’t get that 12-bit image quality ANYWHERE ELSE at it’s price point!

To make up for any current noise and dynamic range issues, all you have to do is spend 10 Euros ($15 US) at your local Tesco, Asda or Walmart and buy those cheap LED strip lights which have a bank of 20 to 40 LED’s on them.

Put some Colour Correction Gels on the LED lights such as CTO-Temperature Orange (3200 degrees Kelvin – Tungsten lighting), CTB-colour Temperature Blue (5400+ Kelvin-Sunlight) or CTG-colour temperature Green (4200K – Fluoro Lights) to match your prevailing light and you will get MUCH BETTER IMAGE QUALITY because you will actually be lighting up your scene much better! You get less video noise and much better colours on faces and in-scene objects!

That’s ALL you have to do! Spend 10 Euros on cheap LED strips and some colour correction gels to get MUCH BETTER VIDEO out of the Black Magic camera! We are going to order a few ourselves as cheap B and C cameras and and as an emergency ‘Backup’. For low-light shooting, REMEMBER to LIGHT THE NIGHT by adding more cheap LED strips to your camera gear bag!

.

Member
October 30th, 2018

Why you cut first two usable stops at sony a7sii in this test? I downloaded your screens and check it with waveworm – it’s not cliped like bmpcc4k and gh5s. wtf?

Member
October 30th, 2018
Reply to  German Polozov

Hi German, those stepchart images are just for visual reference – of course they are from the camera and the stepchart, but every camera is clipping highlights at a different Code value and also the Noise floor code value is different. That’s why I always push or drag the highlights and also push / drag the shadows up or down such that the images in a comparison look comparable. So don’t Download them and run them through waveforms / etc .. they are heavily manipulated for visual purposes.

Member
October 31st, 2018
Reply to  Gunther Machu

LOL!

Clayton Burkhart
Member
October 30th, 2018

Very nice review guys. So pleased to see how well you are all doing these days. Lots of energy, passion and involvement in your craft.

One question, the dynamic range test mentioned in the video was done with Prores outputs. Is there any difference with the distribution of values in RAW ? I have occasionally found slightly more DR in raw on other cameras, however the distribution of the values changes as well. Frequently there is slightly more detail in the highlight rolloff, but also more noise in the shadows. Can anyone test this on the pocket 4k ?

Jonathan Bailey
Guest
October 30th, 2018

If you can get hold of one….

Joe
Guest
October 3rd, 2019

Our company actually has in stock! The pocket camera has done really well!

1
1
- 3
Filter:
all
Sort by:
latest

Take part in the CineD community experience