Andrew Reid

  We only really tested the dynamic range from the middle down into the blacks. We didn't do a latitude test or overexpose the scene. Since we only had 2-3 hours for this test, that was an oversight. I'd really prefer next time to do a real world test and a comparison with the Alexa, since that takes the issue of 4K, 5K out of the already complex task of comparing the cameras.

  A real world test is definitely an aim. Just need to find the time.

That's a good question Rungunshoot.   I will probably add that comparison to the Nikon D5200 review.   So far it is really nice in low light. It will be cleaner (no moire) and more detailed. At ISO 3200 with Speed Booster the NEX 5N might get you more in the shadows though.

    Above: the Epic with anamorphic lens In the second and final part of this mini-shootout with the Epic and Blackmagic Cinema Camera, we take a look at the resolution chart.

What makes you think the markup isn't that much Paul? I found the 128GB version of that Toshiba SSD with matching model number on amazon for $79.   http://www.amazon.com/Style-Original-Toshiba-Micro-Thnsnc128gmmj/dp/B009TF135S   The Red Mag version is $1250   http://www.red.com/store/products/redmag-ssd/750-0026   I'm a little baffled as to what it is about the Red version that is supposably giving it such a high rate of reliability under constant use over the standard drive.

2000 lines of resolution you mean. 4K is a measure horizontally.   It's 4096 x 2160.   Saw my first 4K TV in a high street store yesterday.... 19,000 euros. Anyone want one!?

Neither does the Nikon unless you invest in Nikon glass!

  Click the image to view at 100%   Now considering the GH2 is almost as sharp as a Canon C300 in 1080p with the hack (EOSHD Unified 99 applied for this test), the Nikon is pushing to within 98% of the resolution on the GH2 shot.   However it has the advantage of a more gradual fall off in tone, a smoother looking image, less noise at high ISOs and a larger sensor.   This was shot at ISO 800, F2.8, 1/30.   Lens on the GH2 - Samyang 35mm F1.4. Lens on the Nikon - E Series 50mm F1.8 (pancake)   On APS-C, 50mm gives the same field of view of 35mm on Micro Four Thirds.   The rolling shutter seems similar to other DSLRs on the D5200. For some reason the Nikon seems to have worse rolling shutter on the live view display than on the actual recording.   As you can see from the codec info about the GH2 is pushing upwards of 90Mbit and the Nikon is at 22Mbit (seems a constant bitrate codec).

Locking post as I feel the topic is not constructive.   Certainly interested to hear more from Sean though.

I don't care how it reflects on me.   Talk talk talk... Reputations... Brown nosing... Is for people who give a shit about what others think of their narcissistic self-brand image.   I have no interest in what you think moebius22, and frankly if you can't support the site owner in a difficult situation like this it reflects badly on you, not me.

Technically almost no camera has zero aliasing or moire. It's always lurking on a test chart somewhere.   The D5200 rates very highly where this is concerned, it has virtually nothing noticeable similar to the 5D Mark III.   720p mode is worse though.

Well it seems Leang just got bitch slapped and well deserved it is too.   I find it amazing how much disrespect goes round on the internet from people literally with no justification, aimed at people who have shared a wealth of knowledge and experience such as BurnetRhodes here.   On subject of Bloom - This started because his responses to my 1D C observations were rude, prickly, pompous, belittling putdowns and I will not stand for that in front of a crowd of 60,000 followers. I was debating with absolutely no malice at all but he took it personally. If he can't deal with criticism, even criticism of his camera(!) then he has an major ego issue.

Yes lenses - forgot to mention in the post. The Epic had an EF mount. We used the Canon 85mm F1.2L on that and a Leica R 50mm F2.0 on the Blackmagic to match the field of view, roughly, across varying sensor sizes. I say roughly because the Epic has a different readout area depending on the resolution selected. 5K is more like APS-H, 4K is Super 35mm and most commonly used, 3K and 2K are smaller crops.   We did not use the lenses wide open for the resolution chart stuff. A minimum was F2.8. The Leica is plenty sharp at F2.8 but I did notice some drop off in the corners so won't be using the corners in Part 2. I'll be judging the resolution from the centre of the chart.

It is limited to Nikon and Leica R glass, maybe some medium format stuff.   No Canon FD, M42, EF, Leica M, Pentax, Minolta, Olympus OM. None of it can focus to infinity on the Nikon mount. You're in a similar boat to Sony Alpha owners I'm afraid.   Happily though there's a lot of goos Nikon glass around and you can buy the old AI/S stuff with manual aperture rings.

  Yes clean HDMI.   I'm still not convinced about the merits of HDMI on DSLRs yet though - it never seems to look any different to internal codec! Even though D800 was uncompressed the noise grain still looked blotchy and 4-2-2 didn't seem to make much difference either - not as if it is making full use of 10bit ProRes is it? I thought the hacked GH2 looked better!   I've recorded to Blackmagic HyperDeck Shuttle on the D5200 but only a couple of clips. It looks pretty good but again the internal codec seems to give similar results.

Let's stop that line of conversation here shall we... And get back onto the article.

Bought a Nikon D5200 yesterday - This thing is really surprising me... appears the new Toshiba sensor in the Nikon D5200 fixes moire. There's nothing like the problems here that plague the D600 and D800 full frame sensors. There's still a tiny bit of aliasing but no more than on the GH2. It is giving really nice Canon C300 approaching detail in 1080p, all on a $800 DSLR. You can dial it down really flat in-camera too, though I prefer to shoot graded in camera with a standard picture profile.   In producing footage for the review I have found it to be in the same league as the mighty GH3 for video so far. Here's a 100% crop of a 1080p frame grab shot in 25p mode. The picture profile on the GH3 was slightly flatter in-camera so ignore the contrast difference, just look at the detail. Really nice!  

You need different mounting frames as the screen aspect ratio is different between 5D2 and 60D.   Z-Finder itself can be used on both with correct frame.

I agree. I much prefer the Red codec to Cinema DNG. It is also better supported and more reliable than CineForm currently is.   These things take a lot of processing power to encode though, and a lot of time & skill to develop. I don't think Blackmagic are quite at the point of doing proprietary raw codecs yet on their camera line yet. This is their first attempt at a camera.   I'd like to see 4K ProRes though, and a stronger OLPF.

How does the Blackmagic Cinema Camera compare to an established film industry standard, the Red Epic? Along with ReWo at MBF Film Technik in Berlin we decided to find out.

The C300 is certainly priced for rental.   I prefer the image from the Blackmagic. To get a job done easily and fast - I'd probably choose the C300, for pure artistry the Blackmagic.   I'm not surprised the 5D has disappeared from rental... It never made much sense to rent one when they are so cheap to own! I bet there's a lot of owner operators still shooting their commercial work on one.

  Above: my Blackmagic Cinema camera with ReWo cage - a shootout with the Red Epic is coming soon BlackmagicUser.net recently did an interview with John Brawley, the cinematographer close to Blackmagic Design and main tester for the cinema camera. His experience exposing the camera goes along with what I am finding, that the Blackmagic Cinema Camera works best when you 'expose to the right'.

Nothing I have seen in video mode on any camera can match what the Olympus OM-D E-M5 can do, now even with legacy glass and adapters. It is certainly worth a look for handheld shoots even though the codec sucks and is 30p.   I'm not a huge fan of the OIS on Panasonic lenses for video. Sometimes it jumps and the Lumix lenses are a little bit too clinical looking for my tastes. You are far better off stabilising the camera with a shoulder mounted rig.

You know, the camera I'm most curious about in this test is the Sony F5.   I hope they pay attention to all the cameras equally as I am not sure any one person would be able to get the best out of such a wide variety of them, and some of Shane's info recently has been a little slap-dash... Especially saying stuff on twitter like the Canon Super 16mm lens covers the Blackmagic sensor perfectly with no vignette. I know a guy who actually went out to buy one based on that info and I even wrote an article about it!

8K is certainly useless for TVs... I'd even go so far as to say TVs are dying. The internet is the new TV and when I want to watch something where image quality is a factor I project it onto a 100 inch wall!   In the UK TV is still very much part of the living room but the next generation I feel have moved on from it.   Getting back to resolution... I think there's always going to be a measure of detail but it might not be pixels :)