Danyyyel

I think that this video is a bit misleading because there is zero motion. It would be better to see some more motion to get a better idea of VP9 encoding with 4k on normal footage (movies) compared to paint drying videos.

So this is the future of 4k, because all those advocating 4k will have to learn to use those filters to smooth the sharpness of those 4k sensors, more so on human beings. Tiffen must be rubbing its hand in anticipation of the number of sales coming its way.

No its the nature of promist filter which has very fine particle that will diffuse the light a little bit. So more of it and there will be more diffusion effect. Shane hulburt has a nice article on the different filters on his blog.

Let me guess, familly/holiday shooter... no no, cat shooter, that's it cat shooter.

So true, I cannot comprehend that on a creative site people are advocating copyright theft. Some seem so entitle by them self that they think that there holiday video have to have the music that they want (some multi million dollars production) and that the 20 views from there family and friends is going to magically make a big promo for the artist. Others it is right to pillage the work of fellow artist because either way the money is not going to them but to the big corporation. So lets steal more from them so that they even get less from what they were getting. I comprehend that the implementation, like errors from a bot etc can cause problem but the copyright is something fundamental to protect the rights of artist. If your ambition is just doing some cat, holiday, family and even wedding videos, I can understand that the concept of the sweat and tears that go into the creative work is a bit abstract. But to all those that aspire to create and that the word film maker and not cat maker is appropriate, then copyright should be defended at all cost. Some people on this discussion are just lazy, if you are a genuine creator, either go out and meet and collaborate with other artist. Or take some time and go on the site that have been listed or the web and search for the music and artist and either buy or find venues of collaboration. If your work is of some value, I am sure that you will find other talented people to work with you. With the combination of digital, internet and globalisation it is harder and harder to live out of your art, so if even artist start to question the fundamentals and pillage other fellow artist work in not so long we will have on one side the big corporation producing mostly just mass market shit for only money and on the other side doctors/accountant/lawyers cat shooting film makers and all the artist will be dishwashers/cleaners, because there will be no more way to make a living out of your art.

Yeah, I know about the timelapse, perhaps you are right about the 2.7 crop, but a lot of it is also clearly the full frame look, with the super shallow depth of field.

Very very nice video, looks very sharp to my eyes. If you couple this type of sharpness with 13 stop of DR and amazing low light this is a very very good image. Now for $ 6500, it is not the best investment of you do mainly video, but the people that are going to buy this camera as primarily for photo journalism/sport/wedding are going to get a very very good video camera now.

64 db dynamic range, less than 11 stop.

As Matt has said before, I think your best bet is to get a Nikon D5300. Very good all-rounder with no moire aliasing, best DR and low light in the dslr cameras. It might only loose in terms of resolution but only a little. For example the amount of brick wall in your video could turn into a mess for most dslr and the blackmagic cameras. In terms of DR, I don't think that even the 12 stop of the Nikon or 13 of the blackmagic would suffice. You should either light and Nd the window (movie style) or change your shooting hours when the contrast is much less. The latter is better because using light for example you would need hollywood style budget (many thousand watt of HMI ) to be able to counter the sun. Now blowing out the window in itself is not bad aesthetic. A little better rolloff would be nicer, so you can get a little promist filter (not too much) and the higher DR of the d5300 could be good. In aesthetic, sometime less is good, like resolution is overstated, like you don't want too see every pores of the skin of a beautiful woman.

I think you have to take the overall dxomark numbers with a little pinch of salt. The numbers are very close if you take the 3 individual values dxomark measures like colour accuracy, DR and low light but the D800 kills the Dragon in one aspect not measured. It has twice the resolution. Now it is true that 36 mega pixel is not needed every-time, but if you have to measure all the aspect of the sensor resolution (the easiest one to measure) should also be taken in. The second thing is the ergonomic/shooting photos. What about the autofocus and shooting for example in portrait mode. So it is not as straightforward as it seems. Dxomark measures only the sensor, and a photo camera is much more than just a sensor.

This is a very good thing that DXomark is starting to measure the video camera sensor. It was long time we needed a proper scientific way to measure performance. It will pacify the forums, like it did in the photo world and debunk some of the fanboys. For example, how will the Red and their fanboys take on the Dragon sensor numbers in DXOmark. On one side it is now the overall highest sensor ranking, but on the other side it has only 14.8 stop Dynamic range. Where are the 16-18 that Red has been barking over as the super feature of the Dragon. 14.8 is still very good but not much more compared to their main competitor Arri Alexa.

Whoa that is fantastic, if they are getting that much out of a D5100 so they could get even higher frame rate from the latest Nikon expeed camera. Imagine a 120 fps Nikon D3300 $ 600 Camera LOL. Another thing is that they are getting better knowledge of the code by the day and things like RAW is a possibility (would have prefered 10 bit high bitrate codec). 14 stop DR for $ 800/$1100 camera, that would be awesome. I don't know if the Nikon D5200/D7100 sd card interface would have the speed for at least 1080p raw.

This is what Cory Rich replied to someone on the comment below. If this is true that the full frame is also razor sharp then this is a very very good camera. The D4 was already measured at 13 stop Dynamic Range in the EBU (BBC) test and the low light was awesome. The only downside was the resolution. If it is fixed then it has very very good video. Hi Bruce, Actually we (the entire crew) were blown away by the D4s video quality; it's MUCH improved over the D4! I hate to admit it but the shot at 0:11 was operator error (I just shot it out of focus on a 24mm f/1.4 lens... we were working fast :-) and in the rush to slam in a few clips from Ireland, our last location, the clip somehow made the edit. But the video footage in FX and 2.7 crop mode is RAZOR sharp! And the high iso is game changing. The motor-cross shots that we did at night are at 6400 iso - 60p and they look stunning in full hd. Anyhow, it's a badass camera. Ill be buying two! Hope this helps. Corey Rich

I think the odyssey does, but I don't know if it is not too costly. I find it strange that they did not put 60p recording in 1080p for the Ninja. With all the camera coming out with 60p. They seem fixed on the Canon now, with the 5d and C100. They will be missing some market with all the cheap Nikon's coming out with clean hdmi. Now what we can hope is that the Nikon hack can bring higher bitrate to the 60p. That would be very nice.

I think you should try what I wrote before on not over stressing the codec (more so if you don't have an external recorder). You never know because your first video with the d5300 regardless of the g6 looked sharper than the last two one. I think you were shooting wide open and part of the image was blurred and did not stress too much the codec. One thing to take into account is that some codec are geared toward detail and other toward other factors like subtle gradient, noise, even motion etc. This is from the experience of the different gh2 hack with diftwood and lpowell. So just over stressing the codec might not be a simple answer because with the external recorder these Nikon's might be different than the typical Canon with fix bitrate. One you tested them with wide lens with deep dof and lot of detail they just turn to mush. So one test would perhaps be to just shoot something with a lot of detail but perhaps 1/4 of the screen and a very plain background. So that the codec bitrate is not over stressed but still able to render than part of the image the sharpest it can, at least without using an external recorder that would not rely on the internal codec at all and would show the best result.

One thing to take into account is that some codec are geared toward detail and other toward other factors like subtle gradient, noise, even motion etc. This is from the experience of the different gh2 hack with diftwood and lpowell. So just over stressing the codec might not be a simple answer because with the external recorder these Nikon's might be different than the typical Canon with fix bitrate. One you tested them with wide lens with deep dof and lot of detail they just turn to mush. So one test would perhaps be to just shoot something with a lot of detail but perhaps 1/4 of the screen and a very plain background. So that the codec bitrate is not over stressed but still able to render than part of the image the sharpest it can, at least without using an external recorder that would not rely on the internal codec at all and would show the best result.

It also tend to stress the codec. It use to relevant for the canons but for the Nikon's it would be good to test with an external recorder to take out the codec variable.

At first the narration bothered me a bit and then slowly it just worked out. It had a very woody Allen feel to it and could have been a little nicer with more related imagery, but was still very nice.

I don't know if it is the shots, but the D5300 does not seem that soft compared to the G6. In the last one it even looked sharper. But it could be due to what was in focus.

This is very nice, they are making big progress reverse engineering the Nikon firmware and with the example of magic lantern for Canon, big things could result. I don't know how the SD interface of the Nikons like the D5200/D5300/D7100 would sustain data rate for raw. If they are able to do it, it would be immense because of the dynamic range that would put them much higher than the Canons for me. I would personally prefer a 10 bit log high bitrate codec but I think that it might be more complicated than RAW.

You know Dynamic range is not dependent of just Full frame. It is a set of technologies to treat the signal to noise ration and the full well capacity of the pixel. A very concrete example is that all the latest Nikon/Sony apsc sensor score 2+ stop of Dynamic range (about 14 stop) compared to the 5dmark 3 (11.6 stop) in raw. Even in video the D7100/D5200 where at least a stop above the 5D3. The more filmic look is more to do about the capacity to blur the background so that the lack of composition/framing and lower production value is more apparent than when blurring a bit more the background. Perhaps my problem is that I think more in the movie style of shots than the DW/BBC (when I was young in the eighties, they use to pass short video of people, nature, city scape with some music) that Philip Bloom has popularized.

I think I am loosing my time discussing with you. You started by physics, then it was standard and now to make your point it is just your eyes esthetics. So be it, listening to you, it is as if the Cine 35mm/Apsc can only yield deep focus camcorder style look with no blurred background and deep focus. I think my needs is more of a practical view of moving subject that needs to be in focus and the fact that in low light situation if two cameras like the 5dmark3 and D5300 are about the same quality ISO, it will be much easier to shoot at 2.8 to 1.4 with an apsc with the Nikon vs the Canon. If you have to close the Canon by more than a stop for it to be focusable then you would need to shoot at higher ISO. The same argument can be said about 24p, yes it was the minimum to save on film and get good motion. But in the end 9/10 people would tell you that 24p is one of the main ingredient to the suspension of disbelief or movie Look.

The Nikon cameras can be discused at length, but the Nikon lens is surely one of the best to get for low cost filming because they can be adapted on nearly any format. So should you change camera system in video where autofocus is less used you can choose a whole set from zoom's to primes from more than half a century that will cover a vast range of focal length and can be had for very low price used. Most of them will also cover full frame.

In fact you are confirming what I was saying. We are talking about cinematography and not photography here!!!!!!!!!!! The standart for cinematography is Cine 35mm, which actually is about the same as APSC. So yes the standard is APSC in most motion picture film for more than half a century. As a professional photographer, I can tell you that even if photo and video share many many traits, in practice there are some big difference. Photography is an instant, video is time and continuity. For example if you take 4 photos for a portrait and that one is focused, you photo session is good. But if you are filming someone and that he is in focus only 1/4 of the time you filmed him, your video will be bad.