The Blackmagic Cinema Camera vs Canon EOS 5D Mark III

by planetmatt25 Comments


Blackmagic Cinema Camera vs Canon EOS 5D Mark III notes

So much thanks to Marco Solorio and OneRiver Media for his detailed post comparing the cheaper Blackmagic Cinema Camera (BMCC home) ($3000) to the Canon EOS 5D Mark III ($3500).

Please note that they had a lot of Vimeo trouble uploading the video and finally had to compress it down to 18mbps from their 40mbps lowest quality standard.

Also remember this is h.264 4:2:0 compressed color space on Vimeo, so even when you download the original video source, there is slight banding and artifacts present in the video that are not present in the original BMCC footage.

The quality of the Blackmagic camera is simply so good, like RED or other RAW cameras, conforming to web services is incredibly hard to retain the full detail and color range.

Aside from that, enjoy the video and let me know what you think. Try and download it, Vimeo limits downloads to 100 per day.

Comparing the Cinema Camera & 5D Mk III

Description

After several DAYS of trying to upload this video to Vimeo, we’ve ultimately had to reduce the bit-rate compression down to 18mbps (Vimeo recommends 5mbps, ha!), which is down from our minimum quality level of 40mbps. This means the SOURCE file you can download will also inhibit some amount of compression blocking and smearing, even in the Cinema Camera footage, which doesn’t originally exist in our ProRes master file. We’ve tried EVERYTHING, multiple types of uploads, different encoding methods, you name it. This is as best as it will get unless someone can host our 40mbps H.264 file (about 3GB) on their server that the world can download from.

Although the downloadable source file is a little better than the streaming version, it still doesn’t compare to the original ProRes source file which imposes no banding, compression artifacts, or chrominance sub-sampling (down from 4:4:4 to H.264‘s 4:2:0 space). Please keep this in mind when viewing.

Unfortunately Vimeo only allows 100 downloads per day, so check back to download the 2GB file if the queue is filled. And remember to always watch in FULL 1080 HD or you will have added scaling and moiré issues on some of the tests than what is already been added by Vimeo and our horrid bit-rate restrictions.

Background: This video compares the Blackmagic Design Cinema Camera and the Canon 5D Mark III in several tests. This includes dynamic range, sharpness, pushing levels, banding, artifacts, rolling shutter, chromakeying, wide/telephoto lengths, DOF (depth of field), low light, macro blocking, contrast, and more.

Thanks for watching. Hope this is as informative for you as it was for me making it.

Marco Solorio

What do you think?

Did the Blackmagic Cinema Camera crush the Canon EOS 5D Mark III?

What did you think of the Blackmagic Cinema Camera footage? Does either camera look more filmic or more like video?

Sound off below!

(cover photo credit: snap from the video)

Comments

  1. Pingback: Canon EOS 5D Mark III vs Blackmagic Design Cinema Camera | Paolo Cervesato

  2. jd

    If only the bmcc would find a global shutter solution. Its hard to think doing serious action and camera tracking shoots with such high wobbling. But id adopt this as personal camera any day

    1. Author
      planetMatt

      The RED does not have global shutter – yet. At this point in time, dynamic range is decreased and noise is increased with global shutter on CMOS sensors. I think our best bet is to increase read out speed and develop the tech further. It will come to CMOS sometime.

      To take a step back for a sec, you ever notice the severe whip pans people do to test rolling shutter? I mean, if you had a feature with those types of pan movements, I think you’d make someone sick.

      As long as you are properly stabilize to eliminate vibration and use your cinematic skills to further the story, you’ll be fine.

      1. Federico

        Simon and Matt, thanks for taking your time to share your knowledge about these subjects, however different your opinions may be.

  3. Simon

    Looks like video.

    I’m a D800 fan but I think you were unfair to the 5DMk3. A little sharpening goes a long way with the Mk3 and why were the colours and contrast so washed out in the Mk3 footage. It can do a lot better than is shown here.

    With the Flaat_11p picture profile for the 5DMk3 you can get 11 stops of DR. 12 stops for the D800 which is equally as sharp as the BMCC.

    And as Bloom mentioned the battery lasts only a little over an hour and a half. Yes you can choose external power….

    No weather sealing compared to a DSLR that you can stand in the pouring rain with and the only thing you need to worry about is keeping your glass clean.

    No HDMI out….

    And that sensor is way too small.

    Far too many compromises.

    I love Blackmagic products but this is a storm in a teacup.

    1. Author
      planetMatt

      I’m sorry Simon, but you are severely mistaken. Let me try and wrangle some of your thoughts/questions/statements.

      1. 5dmkIII footage was done with a proper flat profile – the Canon did better here than anywhere I have seen it. It is severely soft, sharpening in post does *slightly* better than 5dmkII but is not perfect. You get halos quick.

      Mush cannot be sharpened – it turns to sharp mush. There is soft, then there is smeared detail that simply cannot be resolved. Some of the foreground larger elements ok, but the background brush? NOt a chance to save in post.

      You forget neither the 5dmkII or mkIII or D800 are 1080P resolution cameras. They do not even resolve 720P.

      Look at the charts.

      As far as color, 8 bit compressed 4:2:0 vs 12 bit RAW. Which do you think is gonna win? lol

      2. D800 does not have 12 stops in video mode. D800 is similar to the 5dmkII in video – the 5dmkIII is better maybe a bit softer but does not alias and moire like the D800.

      To state the “D800 is equally as sharp as the BMCC” is quite offensive actually. You are saying sub 720P resolution is EQUALLY the same as 2,500 lines of resolution (greater than 1080P).

      With your logic, why did we even go to HD then since 720×480 is the same as 1080P right?

      WRONG! If they do look similar, and I’m being serious, maybe you should get some glasses. When our eyes lose their sharpness, it is indeed harder to tell these things.

      3. I would not want to put my 5dmkIII in the rain all day. The BMCC got wet, kept working according to John Brawley.

      4. No HDMI out – better than HDMI – HD-SDI out! Standards here!

      5. Too small of a sensor? Did you even watch the video? He got the same size bokeh as the 5dmkIII (it looked way better actually) and EVEN GOT A WIDER SHOT than the 5dmkIII with the Sigma 8-16 and Canon 8-15 Fisheye.

      Remember, we have over 100 years of film shot on a sensor smaller than the BMCC – 16mm film!

      1. Simon

        Its hard to know where to start since everything you said is wrong, or whether I should even bother since your such a noob.

        I’ve been a camera operator for 20 years and compositing and grading for the past 12 years as well.

        I’ve worked with too many cameras and the D800 is the sharpest and most detailed I’ve come across. Line skipping doesn’t mean its not resolving 1080p. It is. It just means theres a little moire and it is just a little and so rare its never been an issue. Every line of the D800’s 1080p is unique information and to say its not even 720p shows what a noob you are.

        With the Flaat picture profiles the D800 can reach 12 stops. 12.7 with the Flaat_13 profile but thats too noisy for me. Flaat_11 is beautiful and actually seems to reduce colour moire patterns on a test I did with black mesh.

        As for the 5DMk3 I’ve seen much sharper footage shot with it here on Planet 5D including Mitch’s camera test with the D4 it might have been and the 5DMk3 looked great.

        You wrote —— “With your logic, why did we even go to HD then since 720×480 is the same as 1080P right?” You think 720p means 720×480, maybe thats because you a web designer, but 720p is 1280×720. The 5DMk3 and D800 are 1080p. There is no sampling doing on.

        However the BMCC is 2500 (actually 2432 x 1366) pixels wide meaning it then has to be downscaled to 1920. Even if you capture to prores its not going to be fun to edit in. When I said the D800 is just as sharp as the BMCC you can assume I meant in 1080p since the BMCC has to be downsampled to 1080p to play anywhere. Up or down sampling is never ideal. Makes me wonder if in that video did he downsample the BMCC footage to 1080p then edit or up-res the 5DMk3 footage to 2500 pixels wide then output a centre crop which might account for it not looking as nice as I’ve seen.

        No HDMI means no way to attach a decent monitor without first having to add another box etc.

        By the way HDSDI doesn’t mean better quality. HDMI is lossless and 10.2 gigabits per second. HD-SDI is 2.97 gigabits per second. Standards here! as you said.

        As for wide angle you do know that the same lenses mount on both cameras right? Another example of the bias nature of that “review”. So how wide do you think an 8-16mm lens would be on a full frame body? It would be 8-16mm. On the BMCC it would be 16-32mm since its 2x crop. Although I thought I heard it was 2.3 times crop which is even worse but that can’t be right.

        You can’t put a 70-200 2.8 on and stand back every time you want nice DOF. And thats why the BMCC looks like video. Personally I think the mid tones are mushy because they’ve pushed their curve to far to capture highlights and shadows. Yet to see nice skin tones with it.

        P.s. If you want to record RAW on this you’ll get about 30 min per SSD according to Bloom but thats ok since you’ll only need to buy 3 SSD’s since the battery lasts an hour and a half. Thats assuming you turn it off when ever your not recording which makes prep and setup interesting. Or have a bunch of external batteries.

        Putting CINEMA CAMERA on its name was marketing genius tho.

        1. Author
          planetMatt

          A few things to clarify: name calling without justification shows your immaturity.

          On to the reality: prove in the charts where the D800 resolves 1080p. There is a reason we have charts. Do charts mean everything? No. Is there a difference between perceived resolution and actual scientifically proven resolution? Absolutely.

          That said, I have four words for you: Talk to Barry Green. Read this older article, but the D800 and D600 are in the same league as the Canon 5dmkII resolution, aliasing and moire speaking.
          www.dvxuser.com/articles/article.php/20

          On to 720 x 480 – you do know I was talking standard DV definition right? As in saying, your logic is like comparing standard def to HD and saying a little sharpening in post does the trick.

          Now I think, since you did not realize 720 x 480 is standard DV widescreen that you Simon are the ‘noob.’

          Also, HDMI is not lossless – it is 8 bit color, some claim 10 bit but that is not accurate. You can get 10 bit through HD-SDI. You are getting 4:2:0 out of the GH3 right now in HDMI. Do you call that lossless? Crazy talk.

          Another noob point you state is about the lens choice. The BMCC can be wider than a FF because Canon’s 8 mm fisheye does not cover edge to edge on the FF (and has severe barrel distortion, but covers the 2.3 crop beautifully on the BMCC.

          Also, if you know about pin cushion and various barrel distortions, you would understand that a fisheye on the BMCC does not equal a 2.3 crop, it flattens out and goes to less than 1.5 crop.

          Looks like video? Wait, I thought video was a term for low res, bad motion, artifacts, blown highlights – oh yes, something DSLR’s do daily.

          I’ve pixel peeped far more than you, guessed each camera on the Zacuto shootout, I know filmic and video. I would like specifics where the BMCC looks like video.

          Battery – well you know, actual cinema cameras like the RED and Arri Alexa don’t even come with internal batteries – you get bricks and various power sources. I wish I had the option on my DSLR to power externally with as much ease as the BMCC.

          1. Simon

            Keep talking, I don’t need to say anything but as for names you said I was blind and needed glasses so :P

            I would suggest you put some serious time in behind the lens before armchair experts like yourself start making claims.

            HDMI is whatever you want to put through it. If a GH3 puts through 4:2:0 its because of the GH3. If a D800 puts through 4:2:2 its because if the D800. But the medium itself is not to blame. If the BMCC was HDMI it could output the same as its HD-SDI. I’m guessing it was done to be more compatible with Blackmagic’s other hardware.

            Thanks for clearing up the whole standard def and hi-def thing. I had no idea they were different…….

            For wide angle, if my 14-24mm f2.8 has almost no distortion then I’m sure an 8-16 set at 14 will still be wider than the BMCC as for distortion thats up to the quality of the lens. Notice that lens is also f2.8.

          2. Author
            planetMatt

            Well technically I didn’t say blind – maybe your eyes are a bit blurry ;) About time behind the lens: I may only be 29 and have done pro-bono work, personal work and corporate video/film (though I have been on three features), but I actually started linear editing in 5th grade while homeschooled of all things. I was able to be one of the first students in CA to work on a Media100 system and taught myself 3d modeling with Strata as a fifth grader.

            Since then, I’ve been building computers, doing web/graphic design and cinematography. As far as the eye goes – graphic design has taught me far more about aesthetic and composition than any prestigious film school regarding cinematography.

            Sure there are specifics to learn and terminology that I need to learn, but my point is, you have no idea about my history or time behind the lens.

            Let’s stay with what we do know and discuss these cameras without blurring reality.

            The BMCC facts:
            1. 2.5k resolved resolution is far greater than 680 ish lines of the majority of DSLR’s (minus the GH2/3).

            2. Sharpening in post, even on the 5dmkIII does not fix mushy detail because aspects of images were not rendered.

            3. RAW 12 bit images have far greater color detail, gradient rendition and post grad-ability (keying, day for nights, fixing color balance) than compressed 4:2:2 or 4:2:0 whether recorded off HDMI or not.

            4. No DSLR on the market at this time can come close to matching the dynamic range of the BMCC.

            5. The same depth of field and bokeh can be achieved on the BMCC as a FF by simply stepping back 6ft with a 50mm.

            6. The BMCC can actually obtain greater field of view than a FF camera (talking EF lenses here) because of the Canon 8-15 fisheye and Sigma 8-16 lens.

            7. BMCC has numerous options, rather cheap, for long lasting external power that is industry standard – like the RED and Arri Alexa.

  4. weixing

    Hi,
    What’s the point of this review?? Yes, Blackmagic Design Cinema Camera is better than Canon EOS 5D Mark III for video production. I’m not surprise… that’s why 5DIII is a DSLR and Blackmagic Design Cinema Camera is a Cinema Camera.

    I think a lot of people had forgotten that 5D Mark III is a DSLR which mean it is design to took still images. Video in DSLR are suppose to be a nice addition features, not a must have feature. If video professional want to do video production, do yourself a favor and get a camera design for that purpose.

    So, please do not use DSLR for your professional video production, so that DSLR manufacturer can concentrate in making better DSLR that took still great images that most of us paid for.

    Have a nice day.

  5. frank suero

    We are sold as well. We are working with a local supermarket to do product shots for their weekly specials and first tiem in my life i had a customer that knew about video and photography and he just ask us, why is the footage a bit soft? My reaction was just simple, stop, think, and themtold the customer, ther is always a post work to do on the files.

    My main concern is Lenses, now we need to get at least a 16 to become a 35, actually we shot only with Contax lenses from Zeiss, and we do have the 35 1.4, 50 1.4 and the 85 1.5 and thingking about the 135 f2 but now after we get the BMCC we will go to the Zeiss 15mm even though it is really expensive, butm hey it works for all cameras as well.

    Thank you so much for the great review, i love my canons, but man this BMCC is way superior in terns of video to the DSLR’s.

  6. Tom

    I think it is always comparing apples ad pears. Camera’s are completely different. BMC has some dawbacks, anmd I agrree with Bloom amd Laforet in their remarks.
    But I have to say outside of the comparison the picture quality of the BMC is outstanding what I have seen to this point.
    But keep in mind the camera doesn not make movies, it is the director/cameraman/editor and may be more that make the magic of storytelling. I am afraid that buying this camera does not make bad cameraman better, but it is a chance for talented cameraman to have at least a nice picture quality. Bring enough SSD’s would be my advice and buy a more power full computer. The consequence might be that it will be not that cheap after all.

    1. Author
      planetMatt

      Buying any camera does not make a bad cameraman better or a good movie great – especially if the script/acting/editing is poor – BUT – having a camera that has a cinematic look, is sharp, clean, holds highlights and doesn’t blow out things is great.

      Look, us Indie folk cannot afford to create a huge set and have tons of lighting and electricians. Being able to use available light better comes in so handy.

      What the BMCC has shown us is that DYNAMIC RANGE not ISO sensitivity is the key for properly using available light.

  7. Pingback: Thoughts About The Black Magic Camera – dvfreelancer.com

  8. Simon

    Hi PlanetMatt. You had me at Media100. I can remember the last time I heard that name and I completely agree about design being a great way to learn framing. Designers make great cameramen (or women) and cameramen make great directors.

    I dont believe in sharpening in post but I’ve seen sharper footage from a 5DMk3 and believe the 5D footage in that review was skewed. The poster image alone for the video depicts the 5D as being unable to capture even a blue sky without blowing out. Also why am I the only one here sticking up for the 5d on a site called Planet5d, especially when I’m a Nikon fan.

    2.5k I’m not fussed about. When your going to output in 1080p anyhow I prefer the fast workflow over the option of being able to crop in a bit.

    Dynamic range seems to be this camera’s only feature that is not a negative one with 12 stops. But with a decent colour profile like Flaat, the 5D is 11 stops (without loosing data in the midrange as the BMCC looks to me), the D800 12 stops with Flaat and the D4 13 stops out of the box without Flaat, so I’d love to see what a D4 can do with some non-standard profiles. There is also 2.7x cropped mode which has an awesome and sharp image but way too cropped. Would be great for sport and doco’s tho as would the BMCC if it had less rolling shutter.

    I watched the making of for the D600 launch film made by Florian Schulz & Salomon Schulz. Love the film but at the start of the making of video check out the EXTREMELY fast whip pans they are doing in the car, almost no shutter lag!!

    P.s. The D800 and D4 have become the first DSLR cameras to pass the very elaborate European Broadcasting Union (EBU) test, commonly referred to as the ‘BBC test’. It was that test that showed the D4 to have 13 stops. It also mentioned the D4 and D800 exhibited the lowest amount of rolling shutter he (Alan Roberts) has ever seen on a CMOS equipped camera. Apparently the results are online which I want to find.

    Planet5D surely news like that deserves its own post? :)

    With a fullframe camera a shot like someone changing gears in a car, camera close to the shifter in focus looking up towards the driver who’s very out of focus from the f2.8 14mm lens. You can’t get this with a BMCC. Theres no 6 feet to walk back to for shallow depth of field. And do you really want all your shots visually compressed every time you wand shallow DOF?

    I’ll take 4:2:2 uncompressed over the storage and workflow issues of RAW any day. Keep in mind that while RAW has its benefits, theres still only so far you ever want to push an image RAW or not before it looks odd to the eye. RAW doesn’t mean magically seeing in the dark. Storm in a teacup.

    1. Author
      planetMatt

      Ya Media100 with 10GB of disk space was the WOW factor, lol :) I couldn’t believe real time fades and basic effects haha – on SD footage of course.

      1. To me, the 5dmkIII footage looked pretty legit here – they had the flat profile with superb glass and exposed properly, I have a MkII and thought the MKIII was holding that sky better than mine.

      2. For softness, I was one of the first people to complain when I saw test footage leak out of Japan comparing the MKII and MKIII with the windmills. To my eye, the lettering on the MKIII was softer than the old Canon MKII and now people have shown this to be true.

      OneRiver Media is a legit media house – why would they soften, skew or hamper the 5dmkIII especially when they are such a publicly known company? Maybe if I had done this test you could challenge my ethics but I hold firm to the test being legit. Did you download the file?

      3. For 2.5k just realize it is not just good for re-cropping when outputting to 1080p but for creating a beautifully crisp, properly debayered and downsampled 1080p image. Remember, we are talking about CMOS and not CCD chips here, so debayering comes in.

      Technically, no 1080p CMOS sensor is 1080p once you factor in debayering. This is why RED offers 5k for crisp 4K footage and the C300 captures at 4k internally for amazing 1080p 3 channel color (once debayered and downsampled).

      So again, 2.5K is that much better for outputting to 1080p.

      4. Regarding dynamic range: the BMCC is rated 13 stops. I have not seen, heard or obtained charts from any DSLR coming near 12 stops in video mode. in RAW still mode, sure. But I am all for seeing the charts.

      From this test, we certainly could see the MKIII was not hitting 12 stops. Again, in RAW still mode, sure.

      5. I’m eager to see the D600 rolling shutter – that would be great if does better than the MKIII or BMCC. The RED does better than both. Still, I think the MKIII is slightly better than the MKII – could be mistaken though.

      6. Yes, while we have shown one can get super wide on the BMCC with the Sigma 8 or Canon 8 Fisheye, the speed is another issue.

      BUT – hold the phone – BMCC is now offering a MTF that will fit all types of beautiful mounts including PL mount lenses and .95 voigtländer lenses.

      1. Simon

        Here are the links to back up the what I mentioned earlier about the D4 having a dynamic range of 13. Also I found the EBU tested the D800 to be 11.9 (call it 12) stops of dynamic range, and that was using the standard colour profile (not flat).

        mynikonlife.com.au/pro/pro-news/nikon-d4-d800—first-dslr-cameras-fit-for-broadcast

        tech.ebu.ch/camtest/camtestreports04

        tech.ebu.ch/camtest/CamTestReports01

        I’m not mentioning this to push the Nikon’s but rather to show there are other options with excellent dynamic range that don’t have as many workflow issues but each to their own.

        Also reading through some of the other reports the C300 faired very well with 12 stops and the Red one with 10.76 (but that was a very early model).

        1. Author
          planetMatt

          Thanks for the links – I still could not find the actual stops listed in ebu regarding dynamic range in video mode. 12 stops seems likely – I know the D800 has at least a half a stop more than the 5dmkIII.

          Regarding early versions of RED: even with less stops, you still had RED RAW and 12 or 14 bit color which is a night and day difference in post retaining highlights and doing recovery. Take 8 stops compressed codec and 4:2:0 color space vs 8 stops RAW and it will be different.

          Also, remember, dynamic range stops do not tell the entire story. Starting from middle grey, do the camera’s favor highlight or shadow retention? How many stops do they move in each direction?

          The beauty of film, the Arri Alexa, RED and the BMCC is that on highlights they have a nice kneed roll-off instead of blowing out. You could have two 10 stop cameras in dynamic range. One reads further into the shadows, the other into the highlights. And then they both could roll off differently.

          I think after our discussion, having a shootout with a D800 and BMCC would be good. MY only issue is that the shootout between the 5dmkIII and D800 showed:

          1. 5dmkIII cleaner overall in high ISO’s
          2. D800 slightly sharper
          3. 5dmkIII less aliasing and moire
          4. D800 .5 stops better dynamic range (in the shadows)

          The BMCC blew away the 5dmkIII in these tests very clearly where as the 5dmkIII and D800 were a lot closer to each other in their respective tests.

          NoFilmSchool has a 5 part test between the two:
          nofilmschool.com/2012/05/markii-vs-markiii-vs-d800-candlelight-revisited/

          1. Simon

            All the EBU dynamic range measurements are in video mode only. If they were from the stills mode they would have been much higher. And remember those tests were using STANDARD as the colour profile. Love to see how much higher it could go with Flaat.

            That RED measurement was from RED RAW. Your dynamic range is your dynamic range, raw doesnt mean seeing beyond that. But as I mentioned it was an early model.

            This is a quote from the D800 test in relation to the knee roll-off.

            “From the gamma-curve plot, the over-exposure range is about 3 times,
            1.58 stops, and the gamma curve appears not to break off into a linear portion near black, it more closely resemble a film-type logarithmic curve. Therefore, the only limit on low exposure is the noise level.”

            For 5DMk3 vs D800 I know what you’ve written there is the general consensus but I think its misunderstood by most.

            Example. Hi ISO’s on the D800 are treated differently in camera preferring to give you the data as is. That looks bad in reviews because the 5DMk3 is keeping blacks black and doing a lot of noise reduction in camera where as the D800 is leaving that to you which is the way I prefer it. When you grade the shadows and blacks to the same level as the 5D almost all the noise is GONE and a little noise reduction gets rid of the rest leaving a really nice image at 12,800 ISO. Towards the end of the Nofilmschool link you’ll see what I mean. And even then they didnt grade the blacks to be as dard as the 5D showing smoother shading in the chart on the right and her cheek shadow much brighter and smoother, also much sharper looking at her eyes and the chart.

            nofilmschool.com/2012/05/markii-vs-markiii-vs-

            “5DMk3 less aliasing and moire” Because of its soft image. If the D800 was that soft you’d expect to see the same.

            I would pick a 5DMk2 or 3 over the BMCC for the workflow issues alone. Storage issues alone. Battery issues alone. Weather proofing alone…..Trade all that for one f-stop?

            Here is the link to the D600 making of film and check out the whip pans at the very start. Almost no rolling shutter considering how fast the camera is moving. Most rolling shutter tests dont move half as fast as they are in this car.

            vimeo.com/49443048

            Also I noticed DXOMark.com has rated the D600 as its third highest score leaving 6 Nikons in the top ten. Looks to have the same stills dynamic range as the D800 with 14.2 evs compared to the D800’s 14.4 evs. Should be interesting as ev’s are pretty close if not the same as f-stops I think.

            I like our chat. PlanetMitch we need a forum area here where we can all share what we learn and post questions etc. Like Creative cow. Also more news posts please, doesnt need to be gear and new products. Could be technique based sometimes or an interview a day. I come here instead of reading a newspaper (ickk!!).

          2. Author
            planetMatt

            5dmkIII has more NR and is softer than the D800 – I have agreed with this since the tests first came out. However, the D800 doesn’t just have noise like grain, it has codec artifacts like all DSLR’s (but it is helpful to record off the HDMI). Highlights got rough and pasty too on the D800.

            The reality of it is this: I own the 5dmkII and have thousands and thousands of shots on it, hundreds of hours of video on it, I’ve used tons of DSLR’s. (PS: it is NOT weatherproofed really at all, for basic moisture but that is it. Not like a D series).

            If we are only talking about 1 stop of dynamic range I guess why would I pay $3,000 over my 5dmkII?

            The differences are:

            1. Around 2.5 stops of added dynamic range from a FLAT profile on the 5dmkII

            2. 680 lines vs 2,500 of resolution

            3. Compressed h.264 4:2:0 vs RAW 12 BIT

            4. Aliasing and moire hell vs a clean image

            5. No audio monitoring vs full audio support, monitoring

            6. Downscaled, non HD video monitoring vs full HD video monitoring

            7. No software vs over $1,500 of free, professional software included

            Again, the images speak for themselves.

        2. TONYA

          NOTE TO NIKON :how are the nikons fit for the job ?

          mynikonlife.com.au/pro/pro-news/nikon-d4-d800—first-dslr-cameras-fit-for-broadcast

          the very last line, last page of Alans report on the D800 reads ” This camera cannot be recommended for serious programme-making.”

          tech.ebu.ch/docs/tech/tech3335_s02.pdf

          DID NIKON NOT READ THE LAST LINE ???

          but nikon website reads “Nikon’s pro range reaches new heights this week as the D4 and D800 have opened an important door into the professional video broadcasting market. Both models have become the first DSLR cameras to pass the very elaborate European Broadcasting Union (EBU) test, commonly referred to as the BBC Test.

          These tests were conducted by Alan Roberts, who spent most of his career in research and development at the BBC and wrote the new EBU guidelines and testing procedures. Alan commented that both the D4 and D800 exhibited the lowest amount of rolling shutter he has ever seen in a CMOS equipped camera. He also commented that the exposure latitudes of the D4 were truly remarkable at around 13 stops; a figure only bettered by top grade LS rated cameras.

          ( And im a major nikon shooter, d3, d3x etc etc ) but GH3 for video me thinks next.. or a sony something..

  9. Simon

    I’ve never seen rough or pasty highlights on the D800 on anything I’ve shot or elsewhere. This clip is worth checking out which is a nice demonstration of dynamic range and detail.

    As for noise I thought it was the other way around and this quote is from the nofilmschool article.

    “As far as the noise patterns go, the Nikon certainly looks the most like film grain. To me, noise is noise, but when it comes to removing it, a clean pattern with defined noise can actually be easier to remove than a blotchy pattern that already has hardware noise reduction applied.”

    Weatherproofing I assumed the Canons were the same as the Nikons so my mistake there but I have seen a 7D with a hose spraying water test on youtube that was sprayed for an hour I think, and it worked fine after that.

    2. 680 lines (5D) vs 2,500 lines.

    I think you mean vs 1366 (horizontal lines) instead of 2,500 and that 1366 is manufacturers spec as 1080 is the manufacturers spec for a 5DMk2 so until the BMCC is independently tested we wont know but it will be lower than 1366.

    Worth noting that 1366 (RAW) is 30 minutes per SSD so I cant imagine a shoot where you dont need at least 4 extra SSD’s. Thats for the shoot, then you get back to the office, log your footage and maybe delete a quarter of what you shot that your sure you dont need. Then edit the project but even after the clients signed off on it you cant format the remaining 3 SSD’s because your client may want to make a change 6 months later using a different take, so your going to have to copy 3 SSD’s to standard hard drives because they are cheaper than SSD’s before you can use the SSD’s again.

    In Prores and DNxHD its 1080p

    4. Aliasing and moire hell vs a clean image.

    A little harsh considering the beautiful work thats been done with 5DMk2’s. I’ve worked with a lot of 5DMk2 footage and cant say moire is noticeable very often. It’s something you need to look for. When you choose a camera for its shallow depth of field its rare to see a brick wall in the distance in focus enough to be an issue. Plus that mosiac screen filter you can get for the 5D looks amazing.

    Also DSLR’s can edit (trim clips down) and delete footage in camera. I’m assuming the Canons can trim footage as well as the Nikon’s. Its a great feature that never gets a mention. Where as the BMCC cant even delete a clip to save space. Maybe in the future it may be able to but if the could they would already I would have thought.

Leave a Comment