What’s next from Canon?

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
1,105
543
Pretty much any man who brings his tablet into the, ahem, "library" has way more than seven minutes to watch a video. Sometimes as much as 20 minutes here... :geek:

(Please note this post is in no way a comment on where this thread has come to!)
I'm a quick worker in the, ahem, library and rarely spend more than 5 minutes there.
Beyond that, I can read the text transcript of most videos in about 1/3 the time it takes to watch a video. Why waste time when I can be learning three times as much in the same number of minutes?
 
  • Like
Reactions: YuengLinger

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
1,105
543
That 40% Less is still a twice more vs 5D IV files. :) Twice longer rendering time, etc.
Rendering time is usually based on the *uncompressed* size of an image. Most image editors/viewers work with the raster image with full values for each pixel.
 
  • Like
Reactions: SecureGSM

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
1,105
543
Somebody forwarded me some chatter today that Canon is considering replacing the regular EOS bodies with the R mount system. This would now be the second time that Canon screws me over with a new mount.
If true I'd like to sell my gear and save up for a Leica. What would my stuff approximately still be worth at this point?

Canon 5D Mark II with grip
50L
24-105 4L
135L
200 2.8L
bunch of 430EXII and 5series flashes
The R mount is also part of the EOS system.

The flashes will work seamlessly.

Beyond that, Canon appears to plan on selling both R mount and EF/EF-S mount lenses and cameras, along with EF-M mount lenses, for the foreseeable future.
Using an EF to R adapter is about as painless as it gets. There are no degrading optics involved. Both systems are based on the same core protocol. With all electronic connections between the lens and camera body for every EOS lens and body ever made, there are no real mechanical issues to overcome. The adapters are simply a spacer that passes through the camera/lens communications while adding an extra control ring that can be programmed to do pretty much whatever you want it to do.
 
Last edited:

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
1,105
543
i think you have to call these something different, out of focus is a different issue, a lens is designed to focus at a set back distance for its type, a broken or mis-performing lens is not this issue this issue is about the mirror whether other problems can be solved by the same solution does not make them the same problem as it may not appear on two different cameras where as a lens fault would and it would not affect resale as its a natural phenomenon of SLR cameras
AFMA can be used to correct for "all of the above."
 

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
1,105
543
Why do you keep conflating that this is the upgrade for the 5DMIV not the 5DS/R?

This upcoming camera has always been portrayed as the replacement for the high megapixel camera, not the general purpose pro camera. The 5DMV and the equivelant RII will more than likely come outjust before christmas next year, or in spring of 2021. I personally think it does show the interest in the R system (& mirrorless systems as a whole) as people are doing mental gymnastics trying to justify buying this when it comes out early next year instead of simply waiting for the RII, which will most likely be released at the same time and on the same rough release interval as a (possibly final) 5DMV.
No one is conflating any such thing. In the previous discussion the possibility was raised that the upcoming 80+ MP R body would be a mirrorless equivalent for both the 5Ds and the 5D Mark IV with no plans by Canon to release another 35-40MP R body for those wishing to move from the 5D mark IV to the R mount. Kind of like the 1D X replaced both the high speed APS-H 1D Mark IV and the slower FF higher resolution 1Ds Mark III. That's the context of the previous comments you're responding to.
 

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
1,105
543
Maybe I haven't understood debayering correctly: For me it was giving e.g. 10 Megapixel (2.5 Red, 5 Green, 2.5 Blue, millions) sensor the 10 full color Megapixels in the image by extrapolating the color signal of all pixels to their neigbours to give them some cull color information.
Some color adjustment is surely necessary, but I think this is much easier than calculating full color image pixels from single color detector pixels.
All "pixels" in a raw image file are single linear monochromatic luminance values. There's no more color information there than if you put a red, green, or blue filter in front of a frame of B&W film.

It changes the relative tonal values of different wavelengths and combinations of wavelengths of light, but you can't derive color information from them until you compare the results from using multiple filters of different colors. (By the way, that's also how our trichromatic human vision systems work to create color in our perception of visible light. There's nothing intrinsic about the color of a particular wavelength of light, there's only the perception of color by the brain interpreting the signals from our trichromatic retinas with three different types of cones. Other species may have visual response to wavelengths humans can't see, or may not have visual response to wavelengths of electromagnetic radiation that humans can see.)

Just as with a red filter in front of B&W film, more than only red light (that is actually yellow-orange centered on about 590nm) passes through the "red" filter on a Bayer mask in front of a digital sensor. It's just that the further from 590nm the light gets, the more it is reduced by the filter. There's a lot of overlap of what wavelengths of light get through each of the three filter colors in a Bayer array. Some short wavelength "blue" light gets through the long wavelength "red" filter. A LOT of "green" light makes it through the "blue" and "red" filters, a LOT of "red" light makes it past the "green" filters , and there is a second shelf of more than minimal response to short wavelength "blue" light that makes it past the "red" filter. And as we have pointed out earlier, the three colors of the Bayer filter array in the vast majority of digital cameras are in no way equivalent to the "RED", "GREEN", and "BLUE" we use in our RGB color reproduction systems. Then there's the whole issue of human response to different types of lighting "normalizing" the colors we perceive of certain things.
 
Last edited:

SecureGSM

2 x 5D IV
Feb 26, 2017
1,340
322
Rendering time is usually based on the *uncompressed* size of an image. Most image editors/viewers work with the raster image with full values for each pixel.
True. My bad. 40% of the file size reduction will unlikely reduce post production efforts.
I may consider a hi res camera as a second body though. This may turn out to be a reasonable arrangement.
 
  • Like
Reactions: Michael Clark

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
1,105
543
If I look at the Zeiss planar which has a very good reputation, it has macro at f/2 so this is possible. The EF f/2 100mm is an "old" lens introduced 10/1991. If I look at the design of the EF-M 32 there should be some progress in (1) new glass types, (2) new aspherical lens possibilities, (3) new methods of optics calculation. You are right, it is not easy, but I think it would be possible @ 1200-1400 EUR on RF mount - maybe to much for some hypotheticel target group.
How much does the EF 100mm f/2 cost? How much does the Zeiss Planar 110mm f/2 cost? How long has that design been around, by the way? (Hint: it's been around a lot longer than 1991.)

But beyond that, the Zeiss Planar series is not nearly as well corrected for field curvature as the Batis and Otis series are. The Planar was a well corrected lens in 1896 when it was introduced. It's a slight variation of a standard double Gauss design, same as the EF 50mm f/1.8, EF 50mm f/1.2 L, EF 85mm f/1.2 L / II and a ton of other prime lenses in the 35-85mm range (for 135 format cameras). The Zeiss 110/2 is a medium format lens with a field of view about the same as 50mm cropped to a square in the 135 format.
 
Last edited:

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
1,105
543
I'm in the U.S.A. and using the Canon USA site.

The points values don't make a lot of sense. There must be some other method they're using to calculate points. For example, I got a full 8 points for my old Canon EF-S 10-22mm that I bought with my Rebel XTi! Yet as you correctly stated, neither my 16-35L or 24-70L (both original versions) qualified for a single point. I also got 8 points for my original 70-200 F4L IS, but the F2.8 version doesn't get a single point. I have no idea what methods they're using to assign points.
The lenses that you're not getting any points for have all been discontinued and usually replaced by newer versions. The older lenses that you are still getting points for are still in the current catalog and have never been updated (other than the EF 70-200mm f/4 L IS that was only replaced by the EF 70-200mm f/4 L IS II a year ago, so the meter is ticking on that one now.)
 

Cryve

EOS 80D
Jul 4, 2018
108
70
Germany
What kind of target are you focusing on? The AF target should be flat and exactly parallel to the camera's sensor. If you're trying to AF on a tilted target, it's probably just a case of the area of sensitivity being larger than the size of the AF "point" displayed on your screen.
Pretty much anything i focus on has this backfocus. For my tests on a tripod i focused on Things that had other objekts behind them, so i could see if the object behind was actually sharper than what i had focused on.

So i focused on leafes in trees (and the leafes behind the leaf i focused on were sharper)
Or on leafes laying in the grass (and the grass behind was sharper)

or on packing with text on it that i brought with me, and the grass or whatever behind it was sharper.

Here some Pictures, you can check them out for yourselfes: https://drive.google.com/open?id=1jeuy_KH-nR2vXjRAxulw0JquY8lajWII
 

AaronT

EOS 80D
Jan 5, 2013
163
251
"
I'm not picking an argument with you, just making a statement. There is nothing wrong with overkill, might not be for everyone but if you only want 1 camera to do it all then a high MP camera can."

So how about shooting 10+ fps for sports/action/wildlife in motion?

No single camera can "do it all" as well as other cameras can for specific purposes. High MP cameras are specialized tools, just as very high frame rate cameras are. More general "all purpose" cameras balance the suitability for most use cases somewhere in between the extremes of the specialty cameras.
Okay Michael, I'm big enough to admit you got me on that one. Let me rephrase.
"There is nothing wrong with overkill, might not be for everyone but if you only want 1 camera to do it all then a high MP camera can. Except for shooting 10+ fps for sports/action/wildlife in motion." :geek: Then again, a Sony a7R IV hits 10 fps with 61 MP but I would never get one because of its sensor dust problems and ergonomics. Canon has me spoiled in that regard.
 

YuengLinger

EOS 5D MK IV
Dec 20, 2012
2,627
746
Southeastern USA
Dunno what you are talking about. My 5D2 has a single card slot and less than 4fps burst rate.
Okay. I should have consulted an archaeologist. But I do not believe Canon rates the current R as a pro body, so I don't think the RII will be either. A new name will be given to a pro FF mirrorless.
 

mb66energy

EOS 6D MK II
Dec 18, 2011
1,297
205
Germany
www.MichaelBockhorst.de
How much does the EF 100mm f/2 cost? How much does the Zeiss Planar 110mm f/2 cost? How long has that design been around, by the way? (Hint: it's been around a lot longer than 1991.)

But beyond that, the Zeiss Planar series is not nearly as well corrected for field curvature as the Batis and Otis series are. The Planar was a well corrected lens in 1896 when it was introduced. It's a slight variation of a standard double Gauss design, same as the EF 50mm f/1.8, EF 50mm f/1.2 L, EF 85mm f/1.2 L / II and a ton of other prime lenses in the 35-85mm range (for 135 format cameras). The Zeiss 110/2 is a medium format lens with a field of view about the same as 50mm cropped to a square in the 135 format.
Sorry, I was not clear about f/2 planar, I meant this one (EDIT: english version of lens description for convenience :):
which is a evolutionary developed Planar type lens and which is macro - the medium format f/2 110mm lens is no macro with 1:5 max. reproduction ratio.
And TDP comparison shows that the EF 100 L macro is a tad better in the center, but the ZEISS visibly better in the edges while comparing f/2.8 with f/2.0 on the same camera:
 
Last edited:

mb66energy

EOS 6D MK II
Dec 18, 2011
1,297
205
Germany
www.MichaelBockhorst.de
All "pixels" in a raw image file are single linear monochromatic luminance values. There's no more color information there than if you put a red, green, or blue filter in front of a frame of B&W film.
Raw files do code color information because the positions of the pixels are coded in the raw files and connected to r, g and b filter"lets". Shurely you need the information about which pixel/ADC value is filtered by which color (red, green blue) to reconstruct color information and therefore you need some piece of software. So the Bayer pattern is some approach to have alle three color filters at the same time but not at the same pixels to make "one shot color photography" possible.

That these filters have a broader transmission spectrum is critical to produce an image of a yellow Natrium emission line which has to excite the red and green filtered photosites. But this has IMO nothing to do with Bayer patterns or X-Trans patterns and applies to Foveon sensors too.
 

YuengLinger

EOS 5D MK IV
Dec 20, 2012
2,627
746
Southeastern USA
They do here in Canon Netherlands, according to my CPS account.
If you can figure out the CPS point system, you are a true genius!

The R is amazing as a portrait body, and a wonderful complement to a pro body when shooting events. I don't think the EVF performance of the R would put it at the top of the list of most pros as their primary body for sports, dance, or any event that involves more than slow, predictable motion. Two card slots would offer peace of mind; better weather sealing would too; and a higher burst rate is, in 2020, a must. IBIS would be great, but as few pro bodies have it yet, I wouldn't include that in the expected features of a body that paid photographers rely upon to keep customers happy while consistently performing at near the top of what the industry offers.

But, in fact, a pro could take a photo with a smartphone. So, smartphones are pro cameras, if that is your standard.
 
Last edited: