December 04, 2016, 01:20:13 AM

Author Topic: Canon Doing Market Research on Medium Format?  (Read 16978 times)

dgatwood

  • 1D Mark IV
  • ******
  • Posts: 924
  • 300D, 400D, 6D
Re: Canon Doing Market Research on Medium Format?
« Reply #60 on: August 07, 2014, 02:07:17 PM »
However maintaining a constant back focus they could "region" the sensor to FF and you could use existing FF lenses.

I suspect you would have the same problems with FF lenses on an MF body that you'd have with EF-S lenses on a FF body.  They're designed under the assumption that the mirror will be a particular size, and MF would require a bigger mirror.  Judging by the very small difference between adapted lenses that hang at infinity and Canon lenses that don't, I suspect that a change in mirror height of only a millimeter or two would be enough to cause problems.

You could work around this with a specially designed teleconverter attachment that spreads and refocuses the light over the larger surface, with a mirror that doesn't flip straight up, or with a mirrorless design, of course, but each approach has disadvantages, either in IQ, battery life, or mechanical complexity.

canon rumors FORUM

Re: Canon Doing Market Research on Medium Format?
« Reply #60 on: August 07, 2014, 02:07:17 PM »

hajiaru

  • Power Shot G7X
  • **
  • Posts: 16
Re: Canon Doing Market Research on Medium Format?
« Reply #61 on: August 07, 2014, 04:34:43 PM »
Nikon D900 will have 50mp

jeffa4444

  • 1D Mark IV
  • ******
  • Posts: 960
Re: Canon Doing Market Research on Medium Format?
« Reply #62 on: August 08, 2014, 07:17:36 AM »
In order for there to be a noticeable image quality difference, you have to double the diagonal. So you'd be going from 43mm to 85mm, give or take.

I don't think the lenses would be a huge issue. They could rework some existing designs to throw a larger image circle. A "normal" lens would be 85mm. The job would be easier if they went with a mirrorless design. I doubt the 85 f/1.2 would throw an 85mm image circle, as designed, but I'll bet an 85 f/2 wouldn't be a difficult engineering challenge.

They hardest part would be the business case. a 36 x 48mm sensor would be pointless. The image quality difference would be imperceptible. You'd need a 55 x 70mm sensor to make things worth the effort. But then costs would be too high.

It seems unlikely to me.

Not altogether true you have to view it as "a complete system" and you have to decide what is important a. dynamic range or b.resolution. Sony has managed to improve resolution on its CMOS sensors (include any camera with a Sony sensor) by decreasing the pixel pitch but to compensate has better controlled noise particularly dark current noise which improves apparent dynamic range i.e. that range is truly useable. On a 36X24mm sensor the current happy medium is around 5um however the cinematography favorite the Alexa is around 8um because they want better dynamic range & color slightly at the expense of resolution and that image is projected far larger than most photographs.

The relationship between pixel pitch and line pairs per mm is also crutial ideally you want to match the two so you could be doubling this between small & large pixels hence a "complete system" would aim to maintain the relationship and make improvements to other elements like processing, glass types, light wells, micro lenses, fill factors etc.

The other factor to keep in mind is MF will have less depth of field than FF so auto focus will be far more critical and for bigger lenses speed will be the challenge for AF in sports & other fast moving subjects. Bigger glass means more expensive glass and I agree logic would say do away with the mirror to shorten the back focus and reduce the size.
Canon 5DS, Canon 6D, 16-35 f4L IS USM, 17-40 f4L USM, 28 f2.8, 24-70mm f4L IS USM, 24-105 f4L IS USM, 100mm f2.8L IS USM, 70-200 f4L USM, 70-300 f4-5.6 IS USM, 50 f1.8 STM, 100-400mm f4.5-5.6L IS USM II, 1.4EX III,  Canon G7 X, EOS 760D EF-S 10-18mm f4.5-5.6 IS STM

Stephen Melvin

  • EOS M3
  • ****
  • Posts: 236
Re: Canon Doing Market Research on Medium Format?
« Reply #63 on: August 08, 2014, 09:59:21 AM »

In order for there to be a noticeable image quality difference, you have to double the diagonal. So you'd be going from 43mm to 85mm, give or take.

I don't think the lenses would be a huge issue. They could rework some existing designs to throw a larger image circle. A "normal" lens would be 85mm. The job would be easier if they went with a mirrorless design. I doubt the 85 f/1.2 would throw an 85mm image circle, as designed, but I'll bet an 85 f/2 wouldn't be a difficult engineering challenge.

They hardest part would be the business case. a 36 x 48mm sensor would be pointless. The image quality difference would be imperceptible. You'd need a 55 x 70mm sensor to make things worth the effort. But then costs would be too high.

It seems unlikely to me.
Not altogether true you have to view it as "a complete system" and you have to decide what is important a. dynamic range or b.resolution. Sony has managed to improve resolution on its CMOS sensors (include any camera with a Sony sensor) by decreasing the pixel pitch but to compensate has better controlled noise particularly dark current noise which improves apparent dynamic range i.e. that range is truly useable. On a 36X24mm sensor the current happy medium is around 5um however the cinematography favorite the Alexa is around 8um because they want better dynamic range & color slightly at the expense of resolution and that image is projected far larger than most photographs.

The relationship between pixel pitch and line pairs per mm is also crutial ideally you want to match the two so you could be doubling this between small & large pixels hence a "complete system" would aim to maintain the relationship and make improvements to other elements like processing, glass types, light wells, micro lenses, fill factors etc.

The other factor to keep in mind is MF will have less depth of field than FF so auto focus will be far more critical and for bigger lenses speed will be the challenge for AF in sports & other fast moving subjects. Bigger glass means more expensive glass and I agree logic would say do away with the mirror to shorten the back focus and reduce the size.

MF only has reduced DOF if the lenses are built proportionally. Currently, no MF system is capable of producing shallower DOF than a Canon system. It was true in the film days, and it's even more true now.

With the same level of technology (for example, putting a roll of Tri-X in a Nikon and another roll into a Bronica), you have to double the diagonal for there to be a noticeable difference in the image quality.

My wall is full of very large prints made from cameras from the 30D to the 5D, Mk II. I'd challenge anybody to tell which ones were made with the FF camera vs the APS cameras. There just isn't that big of a difference between APS-C and FF. I only switched because of the lenses.

moreorless

  • 1D Mark IV
  • ******
  • Posts: 751
Re: Canon Doing Market Research on Medium Format?
« Reply #64 on: August 08, 2014, 01:53:27 PM »
MF only has reduced DOF if the lenses are built proportionally. Currently, no MF system is capable of producing shallower DOF than a Canon system. It was true in the film days, and it's even more true now.

With the same level of technology (for example, putting a roll of Tri-X in a Nikon and another roll into a Bronica), you have to double the diagonal for there to be a noticeable difference in the image quality.

My wall is full of very large prints made from cameras from the 30D to the 5D, Mk II. I'd challenge anybody to tell which ones were made with the FF camera vs the APS cameras. There just isn't that big of a difference between APS-C and FF. I only switched because of the lenses.

In this respect I actually see more reason for Nikon to go MF than Canon, at present there at a disadvantage with DOF control as there seemingly unable to create F/1.2 lenses with AF due to the smaller size of the F mount where as Canon have shown they can create F/1 lenses if needed. Going MF would give them the chance to come up with something with similar of better DOF control.

I disagree with your point about needing to double the diagonal to see an improvement in image quality but I suspect a big factor will be whether 35mm lenses can be designed easily to make the best of 50 MP+ sensors or whether MF might actually make designs easier.

scyrene

  • 5DSR
  • *******
  • Posts: 1497
    • My Flickr feed
Re: Canon Doing Market Research on Medium Format?
« Reply #65 on: August 08, 2014, 02:25:56 PM »
In order for there to be a noticeable image quality difference, you have to double the diagonal. So you'd be going from 43mm to 85mm, give or take.

I dunno, the 645z (diagonal 55mm?) looks to have better high ISO capability than the A7s, which is about the cleanest full frame sensor at present. It won't be worth the massive price difference to most people, but it is noticeable.

I agree all this sounds unlikely though, anyway.
Current equipment: 5Ds, 5D mark III, 50D, 24-105L, MP-E, 70-200 f/2.8L II, 100L macro, 500L IS II; 1.4xIII + 2x III extenders; 600EX-RT.
Former equipment includes: 300D; EOS-M, EF-M 18-55, Samyang 14mm f/2.8, EF 35 f/2 IS, 70-200L f/4 non-IS, 85L II, Sigma 180 macro, 200L 2.8, 400L 5.6

lexptr

  • Power Shot G7X
  • **
  • Posts: 16
    • Website
Re: Canon Doing Market Research on Medium Format?
« Reply #66 on: August 10, 2014, 03:39:57 PM »
All that Cine and MF stuff is nice, but I afraid, they will lose focus on 35mm format. I.e. less products in that mortal-affordable niche, rare updates.
Canon 7D, Canon 50D, Canon EF-S 60mm f/2.8 Macro, Tokina 11-16mm f/2.8, Sigma 100-300mm f/4 EX DG HSM, Sigma 18-35mm f/1.8 DC HSM | Art, Sigma 1.4X Teleconverter EX APO DG, Canon Speedlite 430EX II.
My website: www.len-lex.com

canon rumors FORUM

Re: Canon Doing Market Research on Medium Format?
« Reply #66 on: August 10, 2014, 03:39:57 PM »