The fact that two separate sensors with the same MP but otherwise different AA configurations, downstream electronics, lenses, etc yielded identical extinction resolutions tells us that Nyquist is the limiting factor. There should have been a difference yet there wasn't meaning one factor sits above the others.
Is the Canon EOS 5DS series to be replaced by a mirrorless camera? [CR1]
- Thread starter Canon Rumors Guy
- Start date
Lensrentals measurements of MTFs on the 5DSR, with the AA-filter neutralised, and 5DS, with the AAfilter on, under the same conditions https://www.lensrentals.com/blog/2015/06/canon-5ds-and-5ds-r-initial-resolution-tests/; and AP's measurements on the Sony RX1R II with the AA-filter off, standard and high under the same conditions are controlled measurements https://www.amateurphotographer.co.uk/reviews/compacts/sony-rx1r-ii-review/7
As stated in the article that was not a test for extinction resolution, but for resolution at MTF50. I've already pointed this out. Did you not read the article? Do you not understand the difference between MTF50 and extinction resolution?
BTW, you can sharpen any image and increase the measured MTF50 point.
That's a very different LPF design. I think it's an interesting result, but also trumped by the fact that two 24mp sensors with/without a traditional LPF show the same extinction resolution.
We can go round and round forever. If the 5DsR (or D800E) truly showed higher resolution (traditionally MTF10), as opposed to merely higher sharpness (MTF50), it would be easy to spot in some of the many test images at DPReview and Imaging Resource. But it's just not there. The images are very difficult to discern even at 100% and 200%, and impossible after a light sharpening pass on the AA filtered versions.
An AA filter means resistance to aliasing, moire, and higher sharpening settings when you do sharpen. No AA filter means some aliasing, occasional moire, and lower sharpening settings. In the case of high resolution sensors like the D800/D800E and 5Ds/5DsR you're never going to see aliasing or sharpness differences in print even if you don't correct them. The only thing that will stand out in print if you don't clean it up is moire.
Do you have a report of the testing you are referring to? I’m curious about the methodology and am happy to learn something. If there is a link above I apologize; I can’t readily find it.
My off-the-cuff guess is that they shoot increasingly close together line pairs, and determine at which point they can no longer register a distinction. Maybe it’s a better methodology, because there are holes in that one. I just find it very interesting that a filter which exists specifically to introduce blur would not pay a resolution penalty for that blur.
Edit: typo correction
Last edited:
If the AA-filter has all the advantages of less aliasing and Moire, and there is no loss of resolution because increased sharpening regains the the resolution, why do just about all the camera manufacturers for their high mpixel sensors either remove the AA-filter or give you a choice of not using it? It would be a no-brainer not to remove the AA- or low-pass filter if you are right.
But:
Canon claims the removal (cancelling) increases resolution:
https://cpn.canon-europe.com/content/product/cameras/eos_5ds_r.do
"the inclusion of a low-pass cancellation filter that allows for maximum resolution from the sensor."
As does Sony:
https://www.sony.com/electronics/cyber-shot-compact-cameras/dsc-rx1rm2
Low pass filter “Off (to prioritize image resolution)”
Nikon
https://www.nikonimgsupport.com/eu/BV_article?articleNo=000006377&configured=1&lang=en_GB
“remove the OLPF with high resolution cameras to take advantage of the increased image sharpness.”
So, are these manufacturers all lying to us?
the thought that the 5Ds and sr are outdated probably has to do with their dynamic range, which so many people have been focussing on lately.
The 5Ds Bodies are missing some of Canon's newer tech like on-chip ADC, Touchscreen, 4k or Dual Pixel Autofocus right? I don't feel like you can blame people for thinking the Bodies are getting old when all these little differences in comparison to the more recent releases are adding up to give just that impression.
Tired about a lot of misleading reports about Canon EOS 5Ds R camera body. I understood Canon is going to have both EOS 5Ds and 5Ds R into a single camera because it was not necessary to have the 5Ds with the low-pass filter with high MP sensor. It is NOT obsolete and had no problem with the filter front of the sensor. It is one of the BEST camera designed. I realized several people wrote the articles did not even own that camera. They provided a FAKE description on the websites.
Canon EOS 5Ds R Mark II is in the test stage with new higher MP sensor with improving features. We have to wait until 2019 to see more EF lenses and new EF based camera bodies included EOS 5Ds R Mark II as a second generation camera. According to the rumor, the second generation may have 60.1 MP or higher for a new full-frame sensor. I think I saw other website showed Canon designed 100 plus MP in 35mm sized sensor.
A higher MP sensor requires a lot of time for data transferring.
Canon needs faster chips to manage higher data speed before the photographer can make the next shot.
Canon developed four different lenses system. The full-frame EF lenses, APS-C based EF-S lenses and EF-M and now RF.
I did not understand why Canon choose "R" and "RF" labels for a new system because the "R" suffix already used in 1960 Canon system. Canon tends confusion with their history because I was there when I got my first camera, Canonflex RM with R lenses. The R lenses became FL then FD and FDn. The FD and FDn considered being the world best lenses with built like a tank because it made in metal. The EF lenses developed based from FD/FDn lens construction plus autofocus and replaced metal with plastic in 1989. The L series EF came with metal and plastic parts combined. Canon modified the lens from 42mm flange to focal length to 44mm for EF lens system.
The bottom line, Canon should not use R and RF labels at all. Secondly, too many different lens models. Hopefully, they will correct by remove "R" out and replace with a new suffix for the mirrorless system.
Canon EOS 5Ds R Mark II is in the test stage with new higher MP sensor with improving features. We have to wait until 2019 to see more EF lenses and new EF based camera bodies included EOS 5Ds R Mark II as a second generation camera. According to the rumor, the second generation may have 60.1 MP or higher for a new full-frame sensor. I think I saw other website showed Canon designed 100 plus MP in 35mm sized sensor.
A higher MP sensor requires a lot of time for data transferring.
Canon needs faster chips to manage higher data speed before the photographer can make the next shot.
Canon developed four different lenses system. The full-frame EF lenses, APS-C based EF-S lenses and EF-M and now RF.
I did not understand why Canon choose "R" and "RF" labels for a new system because the "R" suffix already used in 1960 Canon system. Canon tends confusion with their history because I was there when I got my first camera, Canonflex RM with R lenses. The R lenses became FL then FD and FDn. The FD and FDn considered being the world best lenses with built like a tank because it made in metal. The EF lenses developed based from FD/FDn lens construction plus autofocus and replaced metal with plastic in 1989. The L series EF came with metal and plastic parts combined. Canon modified the lens from 42mm flange to focal length to 44mm for EF lens system.
The bottom line, Canon should not use R and RF labels at all. Secondly, too many different lens models. Hopefully, they will correct by remove "R" out and replace with a new suffix for the mirrorless system.
Getting old and being outdated are two different things. Sure they lack some of the newer features, but they are still capable modern cameras. The 5Ds(R) is only 3.5 years old!
Rebranding after production of at least one camera and 3 lenses would to me seem to introduce far more confusion.
