Look closer at the Specs again. Arriraw is Max. 2888 x 1620. The higher output is for scaled monitor/EVF.
Besides, if you can always re-render CGI at any resolution you want, no? There's hardly any up-convert issue here at all to 4K, unless I'm missing something big.
Rendering motion CGI in higher resolution than 2K is very time-consuming and therefore expensive.
Example; The Amazing Spider-Man where they had to cut back and make compromises to get the rendertime down low enough to support a large production. The Text under the second video in this article explains
what they had to do.
And this was only for a 2K DI, even if the Live-shots where shot stereo in 5K.
.....as to be expected the render times swelled to 30 hours and beyond.
Adding in the indirect lighting to the sub-surface calculation our renders took another big hit and reached well beyond 50 hours on medium or closer framed shots.
So we relented and turned off the extra indirect SSS contribution. Hovering around 20 hours a frame the render times were still unrealistic given scope and number of shots.
........“After a few weeks our R&D team not only cut our rendering time in half to just over 10 hours a frame,” says Bialek
Originally Posted by billdag
The same thing goes for 4K sensors. Most of them also have a bayer pattern so the TOTAL pixel count for them would be 4096 x 2160 = 8,847,360. It's good to see that there are now starting to appear sensors that are capable of recording the full 4K resolution.
The 4K TV's, of course, have 3840 x 2160 = 8,294,400 pixels of each color (red, green, blue) for a total of 24,883,200 pixels!!
That is why only the cameras with higher than 4K sensors have high enough resolution and good enough quality for 4K movies, because they are able to resolve more than 4K.
The only cameras at the moment that are good enough for real quality 4K movies are;
The RED Epic and Scarlet with 14 megapixel, which in September will get the Dragon sensor upgrade with a 19 Megapixel sensor and 16+ stops of latitude . (Red One-MX has the same sensor as the original Epic and Scarlet but slightly lower output resolution).
And Sony F65 with 20 megapixel (effective sensor recording area of 17.6 megapixel) and Sony F55 which also have the same sensor.
But don't fall for the Sony hype that the F65 is a 8K camera, that is clearly not possible without some pixel shifting.
The double (or more) resolution of the intended 4K finish format makes it possible to downsample for better quality motion images.
Ideally the resolution should be four times the intended end format, but that might take a couple of more years with faster in camera processing power to capture that much data.
Not only is the article outdated and even was full of misinformation at the time of publishing. It is based on a previous video lecture by John Galt.
The video lecture was a campaign from John Galt to discredit Bayer pattern CMOS sensors and an argument against the necessity for higher than 2K resolution like 4K and beyond.
Sony motion camera division was also in on the anti CMOS Bayer pattern campaign, even though their stills camera division manufactured Bayer CMOS sensors and cameras.
But after RED entered the arena for Digital Motion cameras with Bayer pattern CMOS sensors in 2007, Sony shifted and released the F65 in the beginning of 2012, five years after RED launched their first 4K sensor camera and a year after RED launched their first 5K camera.
.Edited by coolscan - 7/14/13 at 4:23pm