Why can't the 70D be 40MP in regular mode?
- Thread starter IceAgeDX
- Start date
- Status
?IceAgeDX said:
do you mean in hdr mode or with bracketing & combining the photos? If so, I've definitely had individual photos much larger than 40mp once I've combined 5-7 raw photos (AE bracketed w/ ML) from the 18mp sensors. Not sure on the HDR on this camera since it seems to work in jpeg only.
If you're not talking about HDR or AE bracketed photos & your talking about the split pixels generating high mp photos, I doubt it. But if so, honestly your question is beyond my experience & I'll let the other responses that'll start coming in address your question....
Upvote
0
They probably use the 40MP information, but just use it to improve IQ. Better iso performance, noise reduction, ... It would be a kind of waste to only use these 20 extra MP to prepare a shot (AF) but not to analyze the info taken and to compare it with the other 20 MP. So input is imo definitely 40MP but they only give you a "cleaner" 20MP output, which is more then enough for me!
Upvote
0
No, I don't think that 40MP images will be possible, mostly because, individually, the pixels are not square. As far as I know, (almost) all current digital cameras use square pixels, presumably because it simplifies image post processing. If anybody knows of exceptions (after the Nikon D1X), I'd love to hear about them
Upvote
0
IceAgeDX said:
Huh? Of the Canon 1Dx, 5D3 and 6D, the 1Dx, the most expensive model, has the fewest pixels. Of the Nikon D4, D800 and D600, the D4 has the fewest pixels. Does that mean that Canon and Nikon are "stuck on this anti-megapixel circlejerk"?
Upvote
0
Bob Howland said:
I wasn't trying to argue that it should be able to and that/square pixel seems like a good reason why not. Though if it's just one pixel that just collects an abstract single piece of color information I suppose the shape wouldn't matter... if it were a triangle and it read the color as R23G33B92 it could just assign that to a square shaped pixel in the picture file right?
Upvote
0
neuroanatomist said:
From my other post
"Though if it's just one pixel that just collects an abstract single piece of color information I suppose the shape wouldn't matter... if it were a triangle and it read the color as R23G33B92 it could just assign that to a square shaped pixel in the picture file right?"
Upvote
0
The two "half-pixels" share the same front-lens so they probe the same spatial info and therefore are not increasing the spatial resolution. In order to double e.g. the horizonal resolution they would have to double the little frontlenses too but then they could not use it in the same way for phase contrast autofocus...
Upvote
0
L
Loswr
Guest
IceAgeDX said:
It's not about shape. The half-pixels all have the same orientation, so your hypothetical 40 MP image would be 10944 x 3648 pixels.
As also stated, there's one microlenses over those two sub pixels, so you'd be mucking up the aspect ratio and bloating the file size without adding real information content.
KacperP said:
They do. They're not separate pixels, but two sub pixels under each microlens/color filter.
Upvote
0
I'm not sure if this helps, but its a single pixel that is sub divided. It was not intended to act as two separate pixels, due to the geometry.
As others noted, the Bayer Filter covers the entire pixel with one color.
Here is a schematic diagram of the photosite from Canon's Patent Note that the tops of the two sub pixels are slightly angled so they pick up light from different directions. This geometry allows for the phase detection AF, but they need to be combined to get light from all directions.
As others noted, the Bayer Filter covers the entire pixel with one color.
Here is a schematic diagram of the photosite from Canon's Patent Note that the tops of the two sub pixels are slightly angled so they pick up light from different directions. This geometry allows for the phase detection AF, but they need to be combined to get light from all directions.
Upvote
0
IceAgeDX said:
Yes, there are.
And as already noted, these 20 million tiny lenses determine the image resolution - even though there's a 40mp sensor underneath.
The thing is, this 40mp sensor would have likely had quite poor image quality if it wasn't doing the 2->1 merging of signals.
Upvote
0
- Status
