Rumor: Canon to move into Medium Format?

I shot Medium Format stills (film) for the first time about a year ago... have since been dreaming about the day it becomes Medium Format video.

Hope it comes true.
 

With everyone using cellphones, and dslr market being saturated in some ways, seems like a good move, as enthusiasts have to go to farther extremes to get away from slr parents and the average instagramer.

Anyway, I am still hoping for a real video camera in the size of a stills camera. Been since 2008 and even with huge price premiums that hasn't happened yet. Sad. The IQ is still improving in good ways, but tools for using the things are not being developed quickly enough (ie live waveforms built in).
 
Curious what video features such a camera might incorporate.

You want to shoot the equivalent of 65mm film capture? Those huge lenses, the struggle for light, fighting the shallow DOF. And to make it worth while, it'd have to capture at 8k, with the expensive high speed memory and storage problems that go with it. You want this?

S35, yes. 65mm, no thanks. It's OK for stills if you like that sort of thing (I'm a large format shooter myself, and think medium format is too small to worry with, for stills). But video? I'm not interested in much if any bigger than S35.
 
Who's to say it would have video capabilities and, if it did, would it use the entire sensor for it? I don't know. As I said, I'd be curious to know.

btw... Large format is great. I've shot tons of 4x5 and 8x10 Sinar stuff over the years.
 
I think it's just a made-up rumor to keep Canon fans excited. The D800 route makes a lot more sense. I don't think Pentax is making loads of money with its medium format business.
 
have since been dreaming about the day it becomes Medium Format video.

The DOF would be ultra-shallow unless you shoot everything at f/8.0 or smaller. ( this will also mean a lot more light if you don't want paper-thin DOF )
 
The thing about sensor size, DoF and lowlight sensitivity, is that everything cancels out if you look deep enough.

Like: the 5D2 allowed you to push ISO one stop higher than the 7D, but it also required you to close down your aperture around one stop to have similar DoF.
As long as the number of pixels is constant (which is completely different from what you used to have back in the film days), that is always approximately true. If the larger sensor has more Mpix too, then things can get a bit more complicated, but in general it ends up winning (it can give you a sharper image if you want it, or it can be downsampled to give you something very similar to what the small-sensor camera would give you).

What the bigger sensor gives you is more options:
* I can find a 50mm f/1.2 for full frame, but not a 25mm f/0.6 for m43, or a 12.5mm f/0.3 for 2/3". Then again, try to pull focus with a 50mm f/1.2 on full frame...
* If you're going to use the same lenses in both systems, then the more you close their aperture the sharper they will be (in general), so the one with the bigger sensor will most likely give you sharper images (http://www.similaar.com/foto/lenstestsff/lenstestsff.html)

I don't completely aggree with everything stated here, but if you have some time to spare...
http://www.clarkvision.com/articles/does.pixel.size.matter/
 
Last edited:
* If you're going to use the same lenses in both systems, then the more you close their aperture the sharper they will be (in general), so the one with the bigger sensor will most likely give you sharper images

...Which leads to the question, "Should we always try and use lenses designed for the format we are shooting with?"

i.e. There is a significant difference in image sharpness when using a premium m4/3 lens on a m4/3 camera, versus using a premium FF lens on a m4/3 camera.
 
The thing about sensor size, DoF and lowlight sensitivity, is that everything cancels out if you look deep enough.
That is so, so true and so misunderstood. It took quite a while to dawn on me, but as far as lowlight performance and dof go, then what it comes down to is that the critical factor is the diameter of the front element. If you want better lowlight performance and shallower dof, then for the same angle of view etc then sensor size becomes irrelevant BY ITSELF.

Assume a given sensor size and a lens of (say) 100mm with max aperture of f2. If you quadruple the area of the sensor, you'll expect the native sensitivity to improve by two stops. To keep the same angle of view, you have to double the focal length (so 200mm). But keep the same diameter of lens, and optical theory says by definition that it must now be f4. So increasing the sensor size ONLY improves lowlight performance if you increase the lens diameter at the same time - in the above example, to keep it at f2, you'd need to double the diameter.

As said, unless you increase the lens diameter then sensor size variation alone won't have any effect on low light ability. (Similar arguments for dof as well.)
What the bigger sensor gives you is more options:
Again, absolutely true.

With large sensor you can EITHER have something like a prime lens which is quite fast (to take advantage of the high ISO, give the lowlight performance and shallow dof) OR a zoom lens which may seem quite slow (say f4 or f5.6). But the latter may then be something similar in size/weight etc to the lenses currently used for 2/3" cameras, and although only f5.6 will still have the lowlight performance of an f2 lens on something like a 2/3" camera.

This assumes like for like comparisons, and at present you'd expect 2/3" to be 3 chip, large sensor to be single chip. With all else equal, 3 chip must have a sensitivity advantage of about a stop over a comparable single chip and that needs to be borne in mind.

One relevance is that 4K is becoming more and more the hot topic, it's expected to dominate this coming IBC. And so far all quality 4K cameras have been s35. People have been asking when 2/3" 4K might be seen, to keep the sort of lens options people are used to for sport. But the trouble is that 4K with a beam splitter and 3 chips may be less practical, a single 4K 2/3" chip will compromise sensitivity, so what to do?

Accept the arguments above, and it leads to conclusions that there's not that much point in 2/3" 4K unless it's practical in a 3 chip form. Better to just accept the fact that 4K means large sensor, and if you want a long zoom it will be something like f4-5.6 to keep the size down.
 
What can Canon do to grow the business? Hmmm, lets see...invest millions upon millions of R and D into a new format or actually innovate and merge features (rather than crippling their stills cameras from a video/forward thinking perspective).

Yep, Canon will go medium format.

Canon and Nikon today are like Ford and Chevy from decades ago not paying attention while those "weird" little companies called Honda, Toyota and Subaru started to innovate.
 
There are a lot of film sizes that fit under "medium format"

If Canon was really going to do this then I think it would probably be in 645. Which is 6cmX4.5cm. I wish it would have something like 6X7.

Also the amount of breathing in medium format is so high it would really distract. I would love an MF camera that came in under 4K
 
...Which leads to the question, "Should we always try and use lenses designed for the format we are shooting with?"

i.e. There is a significant difference in image sharpness when using a premium m4/3 lens on a m4/3 camera, versus using a premium FF lens on a m4/3 camera.

I agree with Bruce Watson, harddrive, and especially Samuel H.

In seeking the ideal sensor size, it turns out that the number-one factor is not depth of field but lens choice. That is to say, first ask yourself, of all the lenses in the world, which ones do I want to use most? Then, just see what imaging area they make, and pine for a camera with that sensor.

For me, it's weird to admit it, but the ideal sensor size is full-frame 35mm. And that's because the lenses that meet my needs for the least amount of money are standard 35mm lenses, like these wonderful lenses from Nikon. They breathe little, focus sharply, are small, and each is less than a thousand dollars. Really, the corresponding PL mount lens, from someone like Cooke, are one-hundred times as much. But the Cooke lenses aren't one-hundred times better. They may be 10% better, to the average viewer, maybe 50% better to the discerning eye, and maybe 100% better overall when you consider build quality, etc. But they aren't 1,000% better, even though 1,000% more money.
 
IMAX, I have a frame of it sitting before me as I type & from memory Imax came about by using 120mm Hassy film stock ( ? ) Getting all a bit hazy now. Anyway for those thinking about the "medium format" motion photography it is IMAX.
 
You want to shoot the equivalent of 65mm film capture? Those huge lenses, the struggle for light, fighting the shallow DOF. And to make it worth while, it'd have to capture at 8k, with the expensive high speed memory and storage problems that go with it. You want this?

S35, yes. 65mm, no thanks. It's OK for stills if you like that sort of thing (I'm a large format shooter myself, and think medium format is too small to worry with, for stills). But video? I'm not interested in much if any bigger than S35.

With the sensitivity these modern sensors are reaching, I feel like most of the concerns of shooting 65mm are somewhat fading away. With a 2000 ISO sensor, stopping down that 90mm Hassleblad to f/8 for a manageable depth of field is rather less of an issue than it used to be!

And look at the 65mm Digital prototype that Panavision have built - it's the size of an Alexa, and even lighter! At that point, I think you can safely shoot away in large format.

I'm super excited for the future.

My 6x7 medium format stills photography is still my favourite stills work that I've done.
 
The thing about sensor size, DoF and lowlight sensitivity, is that everything cancels out if you look deep enough.

Like: the 5D2 allowed you to push ISO one stop higher than the 7D, but it also required you to close down your aperture around one stop to have similar DoF.
As long as the number of pixels is constant (which is completely different from what you used to have back in the film days), that is always approximately true. If the larger sensor has more Mpix too, then things can get a bit more complicated, but in general it ends up winning (it can give you a sharper image if you want it, or it can be downsampled to give you something very similar to what the small-sensor camera would give you).

What the bigger sensor gives you is more options:
* I can find a 50mm f/1.2 for full frame, but not a 25mm f/0.6 for m43, or a 12.5mm f/0.3 for 2/3". Then again, try to pull focus with a 50mm f/1.2 on full frame...
* If you're going to use the same lenses in both systems, then the more you close their aperture the sharper they will be (in general), so the one with the bigger sensor will most likely give you sharper images (http://www.similaar.com/foto/lenstestsff/lenstestsff.html)

I don't completely aggree with everything stated here, but if you have some time to spare...
http://www.clarkvision.com/articles/does.pixel.size.matter/

Medium format is rarely chosen for shallow DoF or low light. It has more to do with film grain for tonality and composing wider shots with longer lenses. But what you mentioned is also true.
 
It makes sense that Canon would be looking at medium-format. They have been experimenting on how to make CMOS sensors ever larger. The limitations that current CCD medium-format backs are not suited for video capture due to their sensor size and low-speed capture (generally ~1 fps). A Canon CMOS sensor would definitely open up possibilities for video.

A recent example of a massive Canon sensor is the sensor they made for a telescope. 202mmx205mm (8 inch x 8 inch; 40x the size of full-frame 35mm), 1.6MP, 160μm photosite captures 60 fps video at 0.3 Lux (half the brightness of moonlight). Canon also showed a vastly different sensor that was 120MP for a APS-H sensor size, as well as a 50MP 6-color filter 'multiband' camera that can capture color beyond what the human eye can distinguish; they are experimenting with lots of different technology. With the recent aggressiveness of Sony's sensor business its easy to forget that Canon is technical innovator in this field.

One could imagine that Canon would want to further commercialize their technology, medium-format cameras have basically stagnated recently. Its a market Canon could easily dominate with the right camera.

vHu4Gn5.jpg


75P69RH.jpg
 
Medium format is rarely chosen for shallow DoF or low light. It has more to do with film grain for tonality and composing wider shots with longer lenses. But what you mentioned is also true.

I shoot MF for this reason.

You can get lower DOF in 135 with F.95 then any lens available in MF. You can also go wider with available 135 lens VS medium format glass.

The fastest MF lens available is a MAMIYA 80mm F1.9 which is equal to F1.2 in 135. Some awesome things about MF are chromatic abrasions and coma are better controlled, and there is a 3d pop to you pictures if you do use a shallow depth of filed. This is one that I shot wide open to give you a idea how shallow it gets.

http://www.flickr.com/photos/james-wilsey/9392005758/
http://www.flickr.com/photos/james-wilsey/9389231291/
 
Last edited:
I think it's just a made-up rumor to keep Canon fans excited. The D800 route makes a lot more sense. I don't think Pentax is making loads of money with its medium format business.

I think you're right about the first part. But just for fun:

I don't think halo / bragging rights products have to make loads of money. It's a problem for Pentax 'cause they don't have tons else going on, but Canon? Who cares, as long as it doesn't actually lose money. High margins are expected / accepted on low volume products like this. Red's stayed in the game that way, and even managed to do some killer R&D (Dragon's better than any sensor Canon has, for God's sake). What the heck? Go for it.
 
Back
Top