Is Canon going to do a Video Camera based on DSLR tech?
As already reported, Canon has recently shown a prototype of a replacement for the XH-G1s that records to flash cards rather than to HDV tape. This camera is expected to be rolled out at NAB in April, but already there's been discussion's on the boards about what else Canon might (or should) do.
I think a lot of people expected Canon to do something with the DSLR technology. There were hopes they'd place one of those large chips in a video camera, providing the shallow DOF performance those large chips provide, with better functionality than we're getting out of the DSLR bodies. Seems like a no brainer.
Maybe.
On the face of it, producing a camera like this makes sense, particularly when you look at the popularity of the Canon 7D and 5D Mark II. But when you start getting into the details, things get trickier.
Lenses
Canon's current video camera's - in their professional line - use 1/3" imagers. It's a big step up from 1/3" to the APS-C chip in the Canon 7D and the 35mm chip in the Canon 5D Mark II. Canon does also make lenses for 1/2" and 2/3" imagers, but even the 2/3" lens would barely cover the APS-C chip.
That probably means that Canon will have to produce a new range of video lenses. Alternatively, they could expect users to stick with conventional still camera lenses, but most SLR lenses - particularly zooms - don't operate like a good video lens. They'd have to offer at least one or two lenses for such a camera.
Too much resolution
The other problem Canon has with these chips is too much resolution. They have to throw away a lot of that resolution to get down to an HD image, and at the moment, the way they are doing that seems to be resulting in aliasing issues. To make a really good video camera - and a $7,000 camera has to be good - they have to find a way to better scale the image down, OR save all the data and let users deal with scaling later, OR produce a new large chip that has a lower resolution.
The reality is, Canon has to solve these problems, because while they can (mostly) gloss over the problems the current DSLR cameras have - because they are SLR cameras - it's a harder sell in a multiple thousand dollar video camera. None of these issues are insurmountable, and I'm sure Canon can solve them; if they think enough people would buy the solution. It's just that it's maybe not so much of a no brainer as you might think, and maybe that's why we won't see a video camera built on the DSLR imagers for some time.
I think a lot of people expected Canon to do something with the DSLR technology. There were hopes they'd place one of those large chips in a video camera, providing the shallow DOF performance those large chips provide, with better functionality than we're getting out of the DSLR bodies. Seems like a no brainer.
Maybe.
On the face of it, producing a camera like this makes sense, particularly when you look at the popularity of the Canon 7D and 5D Mark II. But when you start getting into the details, things get trickier.
Lenses
Canon's current video camera's - in their professional line - use 1/3" imagers. It's a big step up from 1/3" to the APS-C chip in the Canon 7D and the 35mm chip in the Canon 5D Mark II. Canon does also make lenses for 1/2" and 2/3" imagers, but even the 2/3" lens would barely cover the APS-C chip.
That probably means that Canon will have to produce a new range of video lenses. Alternatively, they could expect users to stick with conventional still camera lenses, but most SLR lenses - particularly zooms - don't operate like a good video lens. They'd have to offer at least one or two lenses for such a camera.
Too much resolution
The other problem Canon has with these chips is too much resolution. They have to throw away a lot of that resolution to get down to an HD image, and at the moment, the way they are doing that seems to be resulting in aliasing issues. To make a really good video camera - and a $7,000 camera has to be good - they have to find a way to better scale the image down, OR save all the data and let users deal with scaling later, OR produce a new large chip that has a lower resolution.
The reality is, Canon has to solve these problems, because while they can (mostly) gloss over the problems the current DSLR cameras have - because they are SLR cameras - it's a harder sell in a multiple thousand dollar video camera. None of these issues are insurmountable, and I'm sure Canon can solve them; if they think enough people would buy the solution. It's just that it's maybe not so much of a no brainer as you might think, and maybe that's why we won't see a video camera built on the DSLR imagers for some time.
Comments