I’ve written in other posts that I don’t like to concentrate on vaporware like the RED Scarlet. But I think rumors can be fun and give an interesting glimpse towards the possible future. Even if the products never appear, it’s cool to see what some companies are thinking.
There’s a post on CanonRumors.com about Canon’s possible answer to Panasonic’s AF100. Please read the whole post here so I don’t have to copy the original author’s entire article (and check out the CanonRumors site and forum. Fun stuff.)
Done reading? Good. Here’s my take:
The first rumor is entirely plausible and Canon really needs to push this as their Pro video cameras are still very lacking compared to the competition. 2 specs that I have trouble believing though, are 12 stops of dynamic range and up to 1080p60 output. Right now there are very few video camera systems, at any price level, that can manage these 2 things. The only system I’ve used that can do this is the Cineflex (Sony 1500) with Hypergamma turned on, running to an HDCAM SR deck. This is not a cheap combo ($500k-$1M). It could be debated if we’re even getting 12 stops from that system. Another possibility could be the ARRI Alexa, but I don’t know if it can do 1080p60 output.
The second rumor is interesting and I think the original author and his video friend may be missing the meat of the rumor. 3 chip cameras were originally designed using a prism that separates out red, green and blue to different sensors. This usually results in an image with better color rendition than a single chip. The problem is that you can’t use cinema or still lenses on a 3 chip camera without a relay, because all the colors are at different wavelengths and need to travel different paths to get to their sensors. Thus the Bayer-pattern single chip sensor emerged. This is what almost all single chip cameras are, video or still.
20 stops of dynamic range, 3 chip camera? Sounds like Canon has designed a 3 chip HDR camera. All 3 chips get the same info from the prism system (or maybe it’s a mirror system), but each are designed for different levels of brightness, like a bracketed HDR photo. 3 sensors makes more sense than a single chip running really fast because the all the bracketed images would have to be captured at the same moment in time. Since the 3 sensors are working at the same time, their outputs can be merged and processed to create an HDR image in real-time. Creating an HDR video. This is just my speculation, but it doesn’t seem too far fetched. The biggest technical hurdles are probably the real-time processing required to pull this off (I don’t want a RAW system where I have to reprocess all this in post, that’s why I hate the RED workflow), and designing a prism system that will work with 35mm lenses instead of a single built-in proprietary lens.
An HDR digital cinema camera. That’s just cool!