I have had many conversations lately with professionals about shooting with the Canon 5D. Most directors wanted to shoot with it just because it was the big buzz of the moment but as everyone knows it’s not so easy to master. There are many contingencies about filming with the 5D like the rolling shutter or simply the fact that you always have to work in full manual mode. Not to mention the fact that the optics are not easy to use unless you have a full rig of follow focus and an assistant to pull it. So most of you will say why even go in this direction? Well to me this is all a stepping stone. DSLR makers have just shown that yes it’s possible. And the lambda user is happy because he doesn’t have to carry around two cameras to shoot stills and video. But what good does it bring to us, professional users?
To me what is interesting is the fact that the signal is captured by one single chip. Why? I explain….. In the days of film, yeah I know to some of us it seems like prehistoric, the film stock was able to record much more dinamic range than video. This is why it is still used today in 35mm. There is much more freedom to correct and to manipulate the image in post production. Todays video cameras capture about 11 Fstops of light dynamic range. In photography many are now trying themselves at HRD through bracketing (exposing the same picture at different values and later merging the underexposed and overexposed shots) the problem in film and video is that the subject is in movement so there is no time to make 3 consecutive shots of the same frame. But what if 3 of those DSLR chips where to be assembled in the same way our 3CCD cameras are. One CMOS would be capturing the correct exposure and the two others would get an overexposed and underexposed record of the same scene, at the very same moment in time. The three signals would then be assembled together into a MHDRi file. Same as HDRi files except that it would be a 24p or 25p,30p version and based on a floating integer file format enabling to record the entire spectrum of data now captured.
Well the people at Soviet Montage have now tried it. and it works. Gizmodo website states that they have used 2 Canon 5D MarkII DSLR’s coupled together with a beam splitter and varied the exposure on both cameras. I suspect that for now they have rendered the images one by one in batch. But the result is stunning. Have a look.
So now imagine a full camera with 3 CMOS chips coupled together. That is what I think is coming up in the near future.
But the RED comunity is not far behind, they have also released some footage in what they call HDRx taken by a new mode in the EPIC camera.
So this is what I see is going to be the next big step. A big (because I don’t like small cameras camera capable of recording everything the eye can see in terms of light from the darkest shadows to the brightest sun without ever burning or dodging the picture. Huge possibilities in terms of post production color correction including, tone mapping. So be on the watchout for the next HD422 Motion HDRi cameras and if someonr asks you who told you about it just point them out to my blog.
I’d like to know what you think about this so don’t hesitate to leave a comment, I will respond to all of them personally.
Cet article est également disponible en: Anglais