Canon Confirms Development of High Megapixel Camera

Yes. Hence my point that DSLR sales will continue to erode. It's not about mirrorless. It's about smart phones and the microwave generation(s). When 90% of people are only looking to do things like low grade duckface selfies on instgram, they don't give a flying poop about quality DSLRs. Canon is not going make its nut on a 5D3 even being the cash cow of the pro world. If Canon had some sense, they would get in bed with an Apple or Nokia or LG and create cameras and/or lenses for other companies' smart phones. That said I believe from a quality of work perspective and prestige, they will continue to make strides at being competitive in the pro arena.
 
Upvote 0
privatebydesign said:
1/ It is a logical estimation guessed at from some numbers we do know. For instance EF lens sales 100,000,000, number of people that buy a DSLR and don't buy a lens outside the kit lens/es, 97%. That means the total number of all DSLR sales of all EOS cameras ever that goes to pros, semi pros and keen amateurs (the groups that almost certainly would get another lens) is around 3,000,000. If you thought every single one of those pros, semi pros and amateurs was to buy a 52MP camera (and we know they won't) and only they had no choice but that one 52MP camera, since 1987 they would only have sold 3,000,000, or around 100,000 a year. Give a camera a marketing shelf life of 5 years puts sales at an absolute maximum of 500,000 units if every single person who bought a seperate EF lens was to buy one.

The 100,000,000 number is for Canon EF lenses, it ignores all other EF lenses. I have no idea if that number is significant.
 
Upvote 0
kphoto99 said:
privatebydesign said:
1/ It is a logical estimation guessed at from some numbers we do know. For instance EF lens sales 100,000,000, number of people that buy a DSLR and don't buy a lens outside the kit lens/es, 97%. That means the total number of all DSLR sales of all EOS cameras ever that goes to pros, semi pros and keen amateurs (the groups that almost certainly would get another lens) is around 3,000,000. If you thought every single one of those pros, semi pros and amateurs was to buy a 52MP camera (and we know they won't) and only they had no choice but that one 52MP camera, since 1987 they would only have sold 3,000,000, or around 100,000 a year. Give a camera a marketing shelf life of 5 years puts sales at an absolute maximum of 500,000 units if every single person who bought a seperate EF lens was to buy one.

The 100,000,000 number is for Canon EF lenses, it ignores all other EF lenses. I have no idea if that number is significant.

I was rounding, the EF munber, which I assumed to include all TS-E, MP-E and certainly EF-S and EF-M, but the core numbers can't be far wrong. The market for a 54MP camera might be very vocal, but it is comparatively tiny.
 
Upvote 0
dilbert said:
privatebydesign said:
And of that 1,000,000 unit downgrade how many 52MP cameras do you think Canon expect to sell to fill the void?

That may depend on how well it reviews at websites like dpreview, etc.

If Canon do nothing to make the camera more modern than just put an upsized 7DII sensor in a similar but different body, I can't see it reviewing any better than Canon's more recent cameras: i.e. people will go "Yawn, what is Canon doing these days?!"

You are delusional if you believe reviews drive anything much, they mainly serve to reinforce our preconceptions, and history has shown that that is what they do, revisit some of them and you will see what utter garbage many of them are.

But, to believe reviews open up a market is even more delusional, a review might sway some impressionable fool from one brand to another when they should be concentrating on their own needs and experience, however it will not create new purchasers into the DSLR buying pool.
 
Upvote 0
privatebydesign said:
1/ It is a logical estimation guessed at from some numbers we do know. For instance EF lens sales 100,000,000, number of people that buy a DSLR and don't buy a lens outside the kit lens/es, 97%.

Unfortunately, specific camera sales data are not available - so we end up guessing a lot.

Still, I venture you underestimate the high end market with a wide margin. Also your method does not allow for the much higher numbers of lenses Canon currently sells (10 to 12 mio. units /year 2011-13)

The difference between Nikon and Canon Interchangable/DSLR sales trajectory 2012/13 probably tells us something about how much sales a strong new DSLR offering can bring to one of the leading camera companies: 2012/13 Canon DSLR sales fell from 9.2 mio. units to 8.0 mio. units. meanwhile Nikon Interchangable/DSLR sales jumped from 4.7 mio. to 7.0 mio units. A whopping ~50% increase for Nikon. I do not know how that was distributed between their models - but I believe its unlikely that the D800 did not make up a sizeable part of those 2.3 mio. additional Interchangable/DSLR units (while also contributing to Nikon's baseline sales).

Based on this I speculate a succesful 5DIV (whatever megapix it will get) could potentially help boost Canon sales a lot more than just a 100.000 units /year. :P

If Canon ever releases numbers for the best selling DSLR ever - the 5DII - or Nikon does so for the D800/810 we would know much more.
 
Upvote 0
Sometimes I think applying figures and statistics to an otherwise organic subject can be a bit confounding.

There's much talk about resolution and the ability of a lens to resolve it but from practical experience I can wholeheartedly say it's a load of crap.

I've had a Pentax 645z on loan for a few weeks. I've got 30 year old glass on it and it resolves just as well, if not better than any of my Canon glass with little or no distortion or CA when wide open. I've always maintained that beyond a certain point of development and design the importance of super duper statistically perfect glass falls away.

It's just my opinion, but it's always been about the sensor for me. Sure, the mk1 version on the 24-70 was a bit of a dog compared to the mkii. But we're now at a point where the lenses are pretty much bang on.

I struggle to get over the image quality from small sensors with a high mp count on. Really I do. I bet you that if Canon does dump a 52mp 35mm Canon designed sensor on us (The Sony MF sensor is 51.9mp btw) then I say dollars to donuts the majority will rip it to shreds on here because it'll be noisy and have poor dynamic range.

That's unless they've really come up with something special. It's now mostly about the sensor size and the quality of the pixels on it. There's nothing wrong with having a 35mm sensor but the pixels have to be good ones not just an all you can eat buffet with as many as possible on.
 
Upvote 0
Quality Reviews make a lot of hay. There are great technical reviews like dpreview and digital picture. Then are great (and funny) real world use video reviews from guys like Digital Rev TV on youtube. For Pete's sake Kai Wong has 1,000,000 youtube followers on that channel.

Reviews play a huge part of the Camera World, otherwise these companies and rental places wouldn't bother sending these guys gear to test and play with.

Maybe to some folks they don't amount to a hill of beans, but they are most people's chances to see how gear works in various situations before they buy. I didn't have affinity for Sony products and thought if I went to video I'd ML hack a 5D3...

After extensive research and tons of various reviews I'd read and watched, I want a Sony A7s and a Shogun. Lots of power in reviews
 
Upvote 0
Lawliet said:
lo lite said:
The closest resolutions based on 1920 multiples would be 9600 x 6400 which amounts to 61.4MP or 7680 x 5120 which are just 39.3MP. Another half way decent resolution would be 8640 x 5760 which amounts to 49.8MP. But why has it to be such a silly number like 52MP? I don't get this.
If you fill a 135 sized sensor with the pixels of the 70D/7D2 you get about 52MP.
Multiples of 1920 are only beneficial if you can't read out the whole sensor - i.e. if you have to bin sensor cells. Considering that binning only increases the amount of collected light but leads to a lower color sampling density then a full readout the other option would be preferable.

It's hard to do full read out of a sensor with tons of MP though. The A7S has only 10MP for instance.
 
Upvote 0
unfocused said:
Such a silly comment is hardly worth responding to. Canon understands their market extremely well. There is ample evidence of that for anyone paying the least bit of attention. Understanding and meeting the market demand is the only thing that counts. Just because they may not find it profitable to produce exactly what you may want does not mean they don't understand a lot more than you can imagine.

And what of Canon marketing saying that the RX100 was a silly idea that would have zero sales?

What about when Canon marketing couldn't imagine anyone would want manual exposure control for video?

What about when Canon marketing decided that AutoISO was some astonishing high-end feature that deserved to be dribbled out over more than a decade (and even still the 7D2 and 1DX don't quite give AutoISO as well as other brand's Rebel level stuff does in some cases, and it's like a ten cent, ten minute feature to implement)?

etc.
 
Upvote 0
PureClassA said:
So then my next question would be "Can Canon not employ a similar design? Is such a design patented by Sony?" "If they can, what's stopping them from doing this if a solution has already be discovered? Would putting the ADC right on the chip be some radically expensive redesign?"

It's impossible to create enough circuity on the sensor using 500nm fabs. The circuits come out too big and don't fit. They either need to start making DSLR sensors on their newer P&S fabs or build new fabs (very pricey) and both things cost money, money they don't wanna spend it seems.
 
Upvote 0
Lee Jay said:
dilbert said:
Lee Jay said:
dilbert said:
Lee Jay said:
lintoni said:
There's a lot of talk about this camera being a 5DIV. In Canon Rumors post from November, it's stated that the new high megapixel camera will come in a new line, above the 5D line, which makes sense to me. The 5D3 is pretty much the ideal all-rounder camera and I can't see Canon messing with what has been a winning formula.

http://www.canonrumors.com/2014/11/another-50mp-ff-dslr-mention-cr2/

Such a camera would be even more of an "all-rounder" giving the reach of a 7DII, the landscape potential of medium format

You're wrong about that. The DR and noise characteristics of a scaled up 7DII sensor would not put it on a par with other landscape cameras.

I'm still waiting to see my first real-world landscape image where the amount of DR you can get from a Canon sensor at base ISO is insufficient. All I ever see is contrived scenes or scenes where compressing all that DR into the final image makes it look like crap.

Here's an easy one for you: go and shoot a sunset (facing the sun) where you've got interesting things in the foreground that are in shadow. And yes, compressing DR into a final image does

I did that once, and I didn't have enough DR. However, a little math showed that I needed 30 stops of DR for that shot. So 1 or 2 extra stops from a Sony sensor would have done me exactly no good. Even bracketing didn't get the job done.

I've never yet found a shot that needed just a little bit more base ISO DR than I can get from a Canon sensor, but not too much for a Sony sensor.

Just go to a forest where the sun is dappling in and you hit a scenario that needs more DR than Canon delivers but where Exmor DR is just enough.
 
Upvote 0
LetTheRightLensIn said:
It's hard to do full read out of a sensor with tons of MP though. The A7S has only 10MP for instance.

The NX1 reads it's 28MP at 240fps - even with 52M dual pixels that would allow for 60fps, not bad and obviously within the realm of the possible. Or, if you limit phase detection to a really dense (like 750*500) grid of cross type sensors, you'd get 4k/LV/FPAF at up to 120fps.
 
Upvote 0
wockawocka said:
Sometimes I think applying figures and statistics to an otherwise organic subject can be a bit confounding.

There's much talk about resolution and the ability of a lens to resolve it but from practical experience I can wholeheartedly say it's a load of crap.

I've had a Pentax 645z on loan for a few weeks. I've got 30 year old glass on it and it resolves just as well, if not better than any of my Canon glass with little or no distortion or CA when wide open. I've always maintained that beyond a certain point of development and design the importance of super duper statistically perfect glass falls away.

It's just my opinion, but it's always been about the sensor for me. Sure, the mk1 version on the 24-70 was a bit of a dog compared to the mkii. But we're now at a point where the lenses are pretty much bang on.

I struggle to get over the image quality from small sensors with a high mp count on. Really I do. I bet you that if Canon does dump a 52mp 35mm Canon designed sensor on us (The Sony MF sensor is 51.9mp btw) then I say dollars to donuts the majority will rip it to shreds on here because it'll be noisy and have poor dynamic range.

That's unless they've really come up with something special. It's now mostly about the sensor size and the quality of the pixels on it. There's nothing wrong with having a 35mm sensor but the pixels have to be good ones not just an all you can eat buffet with as many as possible on.


Regarding the quality of a lens, I'd be curious to know what aperture you shoot at. Beyond a certain aperture, pretty much every lens is going to produce similar results, because they are all diffraction limited. If you shoot every lens at f/16, regardless of format, your circle of confusion is (barring a particularly bad lens) going to be wholly diffraction limited.


I do agree that for the most part, modern lenses, from almost any manufacturer these days, are more than good enough. Most people won't have any problem with most lenses.


For the discerning photographer, the improvements in recent Canon and Zeiss lenses offer meaningful benefits, and further improvements could be realized that some photographers (who knows how many, but I'd say enough to warrant continued improvements) will recognize. One of the biggest improvements that can still be realized is corner and edge performance. Lenses perform superbly in the center, not all lenses perform well in the peripheries. A lot of improvements in lens quality, from Canon, Zeiss, even Sigma, in recent lens releases have had to do with corner performance. Prior generations had HORRIBLE CA and blurring in the corners, while the new generations perform quite well to exceptionally well (i.e. Otus) in the corners.


Sure, most lenses are good enough for a majority of photographers. However, there are nuances and complexities when it comes to lens design that will continue to warrant improvement in lenses for years to come. Canon's wider angle lenses, particularly their zooms, were (and really still are in some cases) in desperate need of an update, not for resolving power in the center, but for overall IQ and resolving power at the periphery. I think Canon has been succeeding immensely in that area, with the 24-70 IIs, the 16-35 f/4, etc. The Zeiss Otus line is utterly incredible, and should be the icon of lens quality for a decade or more to come. Every manufacturer should strive over the long term to approach the quality of the Otus line.
 
Upvote 0
"Sure, most lenses are good enough for a majority of photographers. However, there are nuances and complexities when it comes to lens design that will continue to warrant improvement in lenses for years to come. Canon's wider angle lenses, particularly their zooms, were (and really still are in some cases) in desperate need of an update, not for resolving power in the center, but for overall IQ and resolving power at the periphery. I think Canon has been succeeding immensely in that area, with the 24-70 IIs, the 16-35 f/4, etc. The Zeiss Otus line is utterly incredible, and should be the icon of lens quality for a decade or more to come. Every manufacturer should strive over the long term to approach the quality of the Otus line." - jrista

+1000 Amen. And I think they are already making stride towards that Otus level. You see it with Sigma ART at 1/4th the price. I think it's also why Sigma has in part delayed their 85mm ART to tweak it to the new Otus level as they did with their 35 and 55mm. Tweak as close to a $4k lens in a $1K package as possible.

It looks like the new 100-400 Mk2 has also fixed a great deal of corner sharpness. This is probably where things will continue to go.
 
Upvote 0
My 85mm L started finding eyelashes at f/1.2 when I adjusted it for back focus.
Will the next EOS microadjust automatically with a little accessory plugged in the USB port?

An interesting finding on the local buy/sell site: Somebody wants to exchange his Fuji XT1 setup to either a Canon 5DIII or a Sony 7S. Obviously, the hype the Fuji flagship has received did not correspond to the actual hands-on experience.
 
Upvote 0
I like using old glass, I have resurrected some manual lenses from the back of the cabinet - but- there is no question that lens design has improved in some respects. I love my old "Planar" (double Gauss) design normal lenses (AIS Nikkor 50mm f/1.2 and Mamiya-Sekor 55mm f/1.4), but they are full of aberrations wide open, though very nice indeed stopped to f/2.8. So, if I want "dreamy", head to the Nikkor at f/1.2. If I want "scary sharp with bokeh", I would be better off with the Sigma Art 50 at f/1.4 (or Otus, but that isn't happening).
 
Upvote 0
LetTheRightLensIn said:
It's hard to do full read out of a sensor with tons of MP though. The A7S has only 10MP for instance.

I'm not sure why you say that. The effort scales linearly in the number of pixels. If you have twice as many pixels, you just throw twice as much hardware at the problem. With a two-sided or stacked chip, I'd think that it would be possible to pack all the needed processing on the back side of the sensor die, with room to spare.

What makes high-MP sensors hard is yield. The more pixels you have, the higher your tolerance for stuck pixels has to be—particularly as the feature size approaches the limits of their lithography technology.
 
Upvote 0
dgatwood said:
What makes high-MP sensors hard is yield. The more pixels you have, the higher your tolerance for stuck pixels has to be—particularly as the feature size approaches the limits of their lithography technology.

Speaking of which, has anybody gathered any statistics on dead/stuck pixel rates across the different sensor manufacturers?
 
Upvote 0
dgatwood said:
dgatwood said:
What makes high-MP sensors hard is yield. The more pixels you have, the higher your tolerance for stuck pixels has to be—particularly as the feature size approaches the limits of their lithography technology.
Speaking of which, has anybody gathered any statistics on dead/stuck pixel rates across the different sensor manufacturers?
Good question. ::)
I am not an advocate of "the more megapixel best". ??? But I know that a 0.1% rate pixel killed in a 52 megapixel sensor would be much less problematic than 0.1% of dead pixel on a 12 megapixel sensor.
 
Upvote 0