[Coiseam]

The Canon does look more gentle. Look at the crop of the woman's eye. Notice that there is less detail in the skin. The Sony is capturing more detail and thus more imperfections. You get slightly softer skin with less contrast with the Canon, and that isn't a bad thing for a portrait photographer. But that goes back to my original point. Why is the 50MP Canon sensor (no AA filter) capturing less detail than the 42MP Sony? Is Canon applying some smoothing to the RAW files? Noise Reduction? Sensor/image processor producing less contrast? The differences are irrelevant for real world use, but its odd that the higher MP Canon is showing less detail and contrast.

As far as the authors opinion on which is most accurate, he seems to be going by his own personal opinion. The Best Cameras for Color Reproduction, Ranked Color is obviously subjective in terms of what is most pleasing to the viewer. What is most accurate and what is most pleasing might not be the same.

I have a feeling that here Sony amplify shadows and contrast and that's the reason why Sony pics seem to look more sharp. The woman's eye colour - I feel that Sony's brown eye colour reproduction is just too dark to match a reality (just a feeling, but I have many brown eyed people around) whereas Canon is simply natural.

My personal preference is to have from the beginning as natural (i.e. close to what I see) RAW as possible. Then I can add sharpness, contrast, amplify colours and so on in post processing. It seems to me that that is easier than doing the opposite - trying to get a natural look from in-camera amplified RAWs. In case of Sony I would try to decrease contrast (and may be lower shadows), but perhaps that would cause overall loss of contrast (not only skin tones but background, too) and that would require working with colour channels and changing contrast in reds only (kind of that).

I can compare Pentax and Fuji RAWs - Fuji have this additional contrast and deep shadows from the beginning. It's not a bad thing - but just for why?

---

Last edited by Coiseam; 07-08-2019 at 12:30 PM.

[Dan Rentea]

If you look at Pentax K-1 files or Nikon files I think you will see the same results. Sharper images with more detail.

If you look at images taken with cameras that don't have AA filter, then is normal to look similar in terms of details. If you look at images taken with cameras with AA filter vs. images taken with cameras without AA filter, then a difference should be visible in terms of details, even if those differences are less obvious than you think.

You have an example in the above link with the woman being sharper on Sony image but in the man image the Canon image looks sharper. How do you explain that?

Never mind, here is an image at full resolution taken with 5D Mark IV, D610 and A7 III (in this order). I downloaded it from a discussion between me and my friends on our private group. The Canon image is larger due to the extra 6mp. The settings for each image are: ISO 100, 1/500s, f2.2, 35mm. Auto white balance on all 3 cameras. Images were taken from a tripod, the distance between images being under 20s. On Canon is ISO 125 because I forgot at that time to change my settings from auto ISO to manual ISO.

On Canon and Sony we used Canon 35mm f2 IS USM (different lenses with the same focal lenght) and on Nikon we used the Nikon 35mm f1.8G. I hope it can be downloaded.

Canon-5-D-Mark-IV-Nikon-D610-Sony-A7-III ? imgbb.com


Color accuracy or not, the Sony image looks the worst in terms of skin tones to my eyes. I've shot enough with A7 III and some other Sony cameras and the only things that I really like on Sony are eye af and silent shutter. But the EVF, the ergonomics, the colors, the touchscreen and the menu makes me skip Sony each time I have the option to buy one at a good price. And now that Canon has entered in the mirrorless game and I've shot quite a lot with EOS R, Sony looks less and less apealing. And the reason that Canon looks more apealing when comes to mirrorless is the new RF lenses. For me are more important than the body and Canon does make exceptional lenses.

[Coiseam]

Color accuracy or not, the Sony image looks the worst in terms of skin tones to my eyes.

The same. But at the same time Sony has the least yellowish greens. If only we were reptiles...

[Dan Rentea]

Another possible difference when comes to sharpness in the files may come from the program used to develop the RAW files. On Canon DPP I have the Picture Style called Fine Details if I want to simulate the lack of AA filter. Compared to Lightroom, the difference in details is visible. On top is the image where Fine details Picture Style from Canon DPP was applied and on the bottom is the same file opened with Lightroom. Nothing else was applied on the files.

The skin tones looks a lot better on the image opened with Canon DPP and the details on the skin are also more visible than the ones from the image opened in Lightroom. Again, I hope that the image can be downloadable from the link. Maybe that's why Sony images looks sharper... Who knows?!

DPP-vs-Lightroom ? imgbb.com

---

Last edited by Dan Rentea; 07-08-2019 at 01:17 PM.

[Rondec]

I have a lot of trouble truly seeing the difference either looking at Imaging Resource Files or DP Reviews widget ( Studio shot comparison: Digital Photography Review

There may be a hint of more softness with Canon files, but when resized they are pretty close and I think with similar levels of sharpening you couldn't tell the difference.

[surfar]

So the variable "kit lens" will be announced very soon.

Price indications alone is guessed at U$ 899.


When it comes bundled with the camera body,I'm expecting close to U$ 2000,which is attractive to people who want to try FF mirrorless.


No other camera maker will have this sort of combination at such a reasonable price.

[Rondec]

Seems like a waste.

How much better will images really be with an entry level full frame MILC and super zoom compared to something like the KP and an 18-270 or 300? Maybe a little bit better...

[surfar]

Maybe a little bit better...

Your mind can assume anything it likes!

[Rondec]

Your mind can assume anything it likes!

Nice. I can think of better ways to spend 2000 dollars for someone who is serious about photography. But each to his own.

[surfar]

I can think of better ways to spend 2000 dollars for someone who is serious about photography.

Yes, of course.You know everything the 2K buys in photography...don't you?

[Rondec]

Yes, of course.You know everything the 2K buys in photography...don't you?

I have a fair idea.

[surfar]

I have a fair idea.

That's great!

I can think of better ways to spend 2000 dollars for someone who is serious about photography.

Tell me the better wayS for the candidates who will be considering M/L FF?

[Winder]

Another possible difference when comes to sharpness in the files may come from the program used to develop the RAW files. On Canon DPP I have the Picture Style called Fine Details if I want to simulate the lack of AA filter. Compared to Lightroom, the difference in details is visible. On top is the image where Fine details Picture Style from Canon DPP was applied and on the bottom is the same file opened with Lightroom. Nothing else was applied on the files.The skin tones looks a lot better on the image opened with Canon DPP and the details on the skin are also more visible than the ones from the image opened in Lightroom. Again, I hope that the image can be downloadable from the link. Maybe that's why Sony images looks sharper... Who knows?!

That might be true, but the reviews I have linked above and some others I have seen have been from existing Canon users who were using LR to process both.

Never mind, here is an image at full resolution taken with 5D Mark IV, D610 and A7 III (in this order). I downloaded it from a discussion between me and my friends on our private group.

I know the original capture was full resolution, but the 15MB file in the link is obviously not the full resolution of of each image. It is 3 images that have been processed and compressed into a single JPEG. It would be impossible to tell anything about the starting RAW files from that.

[Dan Rentea]

That might be true, but the reviews I have linked above and some others I have seen have been from existing Canon users who were using LR to process both.

Maybe the algorithms for Canon files are not as good in Lightroom?! I was a big fan of Lightroom until I discovered by accident the Canon software (Canon DPP). It's been more than one year since I stopped using Lightroom or Camera RAW to edit images because the difference is quite visibile between these editing software. I use Canon DPP software for basic edit and if the file needs some retouching in Photoshop, I export the files in TIFF format from Canon DPP.

I know the original capture was full resolution, but the 15MB file in the link is obviously not the full resolution of of each image. It is 3 images that have been processed and compressed into a single JPEG. It would be impossible to tell anything about the starting RAW files from that.

Winder, we are a group of close friends shooting with different systems. Some of us are using Canon, some Sony, some Nikon, some are using 2 systems. We go out as often as we can and from time to time we do these tests as a "treatment" for GAS (gear acquisition syndrome), not to convince anyone about anything.

The images were in landscape orientation and we transformed them in portraits by applying 2x3 ratio instead of the original ratio (3x2) in order to remove the parts from the images that weren't interesting to us, that's all. There wasn't any resize of the images and it doesn't matter if you zoom in a landscape image or a portrait image as long as the images weren't resized.

All the images were RAW and converted to JPEG in their own software (for Canon we used Canon DPP, for Nikon we used Capture NX-D, and for Sony we used Capture One Pro). We haven't touched the images by adding contrast, saturation, etc. to them. The only one who edited one of the images from that test was my nephew (the boy from the image) who became interested in photography and editing and he used my Photoshop action to edit himself. Below is the image edited by him (I've taken it from his Facebook page).

So, if you ask me, as long as we used the original software to convert the RAW files into JPEG files, our results are closer to reality than the ones using one software for all the images. Lightroom is good as a software for the photographers who have to edit images in batch and not for getting the most out of the RAW files. And as I showed you in the second crop from Lightroom and Canon DPP, it seems that Lightroom does reduce the sharpness in Canon raw files and because of that some think that Canon is applying noise reduction or smothening in their raw files. We may still have the RAW files from that test and when I'll get back from holiday I'll look for them on my hard drive. If not, we will go out again in one week and we can do other tests.




---------- Post added 07-09-19 at 08:07 AM ----------

Seems like a waste.

How much better will images really be with an entry level full frame MILC and super zoom compared to something like the KP and an 18-270 or 300? Maybe a little bit better...

I don't think that people who buy Canon RP are looking for best image quality and durability. They will probably be interested in the Canon RP with that super zoom lens because this combo gives them:

- a light combo (Canon RP weights 485g with card and battery)
- a versatile lens for the ones who don't want to invest in another lens
- an articulating touchscreen LCD
- 4k video capabilities
- the idea of having a full frame sensor for low light situations
- EVF (it could be a plus for some and to others it may be the reason for not to buy it)

And there is another major factor that you have to take into consideration. Canon is present all over (internet or store shelves), while Pentax is hard to find even online if you know nothing about cameras in general. I know I wouldn't buy this Canon new lens and if I buy RP I would probably use it on vacations with the RF 35mm f1.8 lens or with EF 16-35mm f4 lens. It hasn't been available for renting yet, but for traveling I'm considering getting the RP just for the size, weight and for the fact that I can use my lenses without compromises.

Speaking of internet presence, even if almost everyone on internet speaks only about how better Sony is compared to competition, Canon somehow managed to increase their market share in 2018. They may know the market better than the influencers out there. It seems that Nikon and Sony have the direct battle for market share...

Canon 40.5 % (+ 3.9)
Nikon 19.1 % (- 2.7)
Sony 17.7 % (- 0.7)
Fujifilm: 5.1 % (+ 1.3)
Olympus: 2.8 % (+ 0.1)


And the link to the article Digital camera market down 22% for 2018, the latest global market share: Canon 40.5 %, Nikon 19.1 %, Sony 17.7 % - Nikon Rumors

[biz-engineer]

Tell me the better wayS for the candidates who will be considering M/L FF?

Who said mirrorless is a must have, or prove that mirrorless camera take better images. For me, going from 700 x 36Mp CIPA shots on a K1 to 250 x 20Mp CIPA shots in a RP and having to buy a new set of RF lenses, isn't progress, it's about get less for more money.