Pages: 1 2 3 [4]   Go Down

Author Topic: Canon 5Ds/R New review with file samples  (Read 18770 times)

fdisilvestro

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1862
    • Frank Disilvestro
Re: Canon 5Ds/R New review with file samples
« Reply #60 on: March 26, 2015, 01:52:09 pm »

Do you know of any way to set a custom curve as the 'default' in-camera jpeg shooting mode? Because, if you could set a linear curve going from the noise floor to full well saturation, that would give you a RAW histogram anyway.

It is not possible to get a RAW histogram once you transformed to jpeg, no matter what "custom" curve you apply to it. Raw histograms represent the luminance recorded in each channel, in jpeg there is an encoding to a color space (sRGB or Adobe RGB). When you see a clipped channel in jpeg you cannot tell if it is because of luminance or a color outside the gamut. Raw values are not limited by the gamut of a color space.

fdisilvestro

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1862
    • Frank Disilvestro
Re: Canon 5Ds/R New review with file samples
« Reply #61 on: March 26, 2015, 01:58:46 pm »

Sorry to veer the topic away from blending and DR ( ;D), but thanks Bart for the info on RawDigger. I have never used it, is there a link you could point me to (for Windows 7)? Is this a utility for just being able to read the header information, or can you actually either view the image itself or save it into a format that can then be read via PS or DPP? Thanks

Jim

RawDigger is a wonderful tool to analyse a RAW image. Yes you can view the image. It is available for windows 64 & 32 bits and Mac.
www.rawdigger.com

Edit: oops! I did not realise that Bart had already answered to this.

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Canon 5Ds/R New review with file samples
« Reply #62 on: March 26, 2015, 02:07:41 pm »

It is not possible to get a RAW histogram once you transformed to jpeg, no matter what "custom" curve you apply to it. Raw histograms represent the luminance recorded in each channel, in jpeg there is an encoding to a color space (sRGB or Adobe RGB). When you see a clipped channel in jpeg you cannot tell if it is because of luminance or a color outside the gamut. Raw values are not limited by the gamut of a color space.



You are generally right, but there are some workarounds. First, a small correction. The information recorded in the raw file does not bear any special relationship to luminance, which presupposes a spectral sensitivity that hardly ever (actually never, in my experience)  matches that on any raw channel.

With that quibble out of the way, if is possible to get the histograms of the JPEG image to approximate those of the raw image by setting the white balance of the camera to [1 1 1]. This is sometimes called "UniWB". It is not a perfect solution, but many find it workable.

Here is a discussion and some instructions for those interested in this approach.

http://blog.kasson.com/?page_id=2387

Jim

fdisilvestro

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1862
    • Frank Disilvestro
Re: Canon 5Ds/R New review with file samples
« Reply #63 on: March 26, 2015, 05:41:32 pm »

You are generally right, but there are some workarounds. First, a small correction. The information recorded in the raw file does not bear any special relationship to luminance, which presupposes a spectral sensitivity that hardly ever (actually never, in my experience)  matches that on any raw channel.

With that quibble out of the way, if is possible to get the histograms of the JPEG image to approximate those of the raw image by setting the white balance of the camera to [1 1 1]. This is sometimes called "UniWB". It is not a perfect solution, but many find it workable.

Here is a discussion and some instructions for those interested in this approach.

http://blog.kasson.com/?page_id=2387

Jim

Hi Jim,

Thanks for clarifying about luminance. I was actually referring to the intensity of light recorded by the sensor, which has no relation to color saturation.

In relation to uniWB, I know about it and use it regularly, but is (at least IMO) just a "better than nothing" approach.

The first attached file shows three channels that go from middle gray to full saturation in RGB and the corresponding histogram, which shows the channels near saturation. The second file shows the gamut warning in sRGB if the original image was in ProPhotoRGB. Somebody could take the wrong decision to reduce exposure to avoid those blown channels if restricted to a color-space encoded histogram.

Regards

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Canon 5Ds/R New review with file samples
« Reply #64 on: March 26, 2015, 06:18:38 pm »

Thanks for clarifying about luminance. I was actually referring to the intensity of light recorded by the sensor, which has no relation to color saturation.

I figured that was what you meant.

In relation to uniWB, I know about it and use it regularly, but is (at least IMO) just a "better than nothing" approach.

The first attached file shows three channels that go from middle gray to full saturation in RGB and the corresponding histogram, which shows the channels near saturation. The second file shows the gamut warning in sRGB if the original image was in ProPhotoRGB. Somebody could take the wrong decision to reduce exposure to avoid those blown channels if restricted to a color-space encoded histogram.


Your example is an excellent demonstration of the dangers of using a histogram after a transformation to an arbitrary color space to learn something about the characteristics of the image in the original color space. However, with the exception of some highly saturated subjects like flowers, that this is usually a problem when you're using UniWB and the in-camera histogram.

Nevertheless, if you say UniWB is inadequate for you to judge exposure in-camera, then who am I to say you're wrong?

I'm sorry if I wasted your time; it wasn't clear to me from your earlier post that you knew all about UniWB.

Jim

shadowblade

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2839
Re: Canon 5Ds/R New review with file samples
« Reply #65 on: March 26, 2015, 08:41:28 pm »

Thanks and I do not remember to have seen any of them before.

Btw. they make Peter Lik look naturalistic :)

I'm not sure why they're showing up as so heavily saturated and with a colour cast in Firefox (even with colour awareness turned on) when they look normal (on the saturated side) in Photoshop and in print.

My style definitely leans towards high saturation and contrast, though. Possibly due to that, I've noticed a definite geographical distribution sales - lots of sales to Mediterranean countries, Latin America and southern/southeast Asia, with very few sales in northern Europe, North America (outside of the US West Coast) and other English-speaking countries apart from Australia. They probably work better in a bright, tropical hotel or resort than in the darker, more subdued settings more popular in these places.
Logged

Hans Kruse

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2106
    • Hans Kruse Photography
Re: Canon 5Ds/R New review with file samples
« Reply #66 on: March 27, 2015, 04:31:44 am »

I'm not sure why they're showing up as so heavily saturated and with a colour cast in Firefox (even with colour awareness turned on) when they look normal (on the saturated side) in Photoshop and in print.

My style definitely leans towards high saturation and contrast, though. Possibly due to that, I've noticed a definite geographical distribution sales - lots of sales to Mediterranean countries, Latin America and southern/southeast Asia, with very few sales in northern Europe, North America (outside of the US West Coast) and other English-speaking countries apart from Australia. They probably work better in a bright, tropical hotel or resort than in the darker, more subdued settings more popular in these places.

I see the same in Safari. Firefox on the Mac is color managed well enough to show pictures the same way as Safari. Why not link to your website as I assume you have one (given print sales)?

I have also noticed the relationship between location and favoring highly saturated images. Asia comes to mind here. I have also seen it on my workshops to some degree. Some go totally overboard in saturation and contrast :)

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13985
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Canon 5Ds/R New review with file samples
« Reply #67 on: March 27, 2015, 09:14:51 am »

I see the same in Safari. Firefox on the Mac is color managed well enough to show pictures the same way as Safari. Why not link to your website as I assume you have one (given print sales)?

I have also noticed the relationship between location and favoring highly saturated images. Asia comes to mind here. I have also seen it on my workshops to some degree. Some go totally overboard in saturation and contrast.

I didn't know some people still dared to touch the saturation slider. ;)

Cheers,
Bernard

shadowblade

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2839
Re: Canon 5Ds/R New review with file samples
« Reply #68 on: March 27, 2015, 09:25:55 am »

I see the same in Safari. Firefox on the Mac is color managed well enough to show pictures the same way as Safari. Why not link to your website as I assume you have one (given print sales)?

I have also noticed the relationship between location and favoring highly saturated images. Asia comes to mind here. I have also seen it on my workshops to some degree. Some go totally overboard in saturation and contrast :)

The website's down at the moment, and has been for a few months - don't have enough time to get it up and running. Selling through a few online galleries/brokers at the moment.

As far as geographical distribution of sales go, I'd probably divide it between 'areas with strong Protestant Christian cultural influence' (favouring muted images) and 'almost everything else' (favouring saturated images). Which is probably understandable from an artistic-historical point of view. Black-and-whites go in their own category.

I prefer saturated colours and bold, contrasty black-and-whites myself. To me, landscapes evoke drama, passion and emotion. Saturated colours and strong contrast convey this well - I use them for the same reason I use long exposure or ultra-short exposures to give a piece a certain mood. Muted colours, low-contrast renditions and a realist or neoclassical aesthetic don't have the same effect.
Logged

spidermike

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 535
Re: Canon 5Ds/R New review with file samples
« Reply #69 on: March 27, 2015, 09:57:53 am »


As far as geographical distribution of sales go, I'd probably divide it between 'areas with strong Protestant Christian cultural influence' (favouring muted images) and 'almost everything else' (favouring saturated images). Which is probably understandable from an artistic-historical point of view. Black-and-whites go in their own category.

I prefer saturated colours and bold, contrasty black-and-whites myself. To me, landscapes evoke drama, passion and emotion. Saturated colours and strong contrast convey this well - I use them for the same reason I use long exposure or ultra-short exposures to give a piece a certain mood. Muted colours, low-contrast renditions and a realist or neoclassical aesthetic don't have the same effect.

I thing there is a regoinal difference regards colours and is something I have noticed in Asia. I agree that in the West there is a growing propensity for saturated colours but in Asia the colours can become downright gaudy - not just saturated by bright and garish.  I think it a a book by Michael Freemen that gave the intersting supposition that in the West we have been through the whole evolution of art as a representative medium, from merely a way of recording what happened with any sense of scale being secondary and perspective non-existent through to the painting-as-reality (for example the accurate representation of perspective or the analysis of colour by Turner). This has also pervaded our approach to photography. But in Asia art took a very different path and the accurate use of colour was far less important, so when they want to emphasise colour in a photo they see nothing wrong with whacking up all sorts of controls and it can look downright weird.
Logged

shadowblade

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2839
Re: Canon 5Ds/R New review with file samples
« Reply #70 on: March 27, 2015, 10:54:43 am »

I thing there is a regoinal difference regards colours and is something I have noticed in Asia. I agree that in the West there is a growing propensity for saturated colours but in Asia the colours can become downright gaudy - not just saturated by bright and garish.  I think it a a book by Michael Freemen that gave the intersting supposition that in the West we have been through the whole evolution of art as a representative medium, from merely a way of recording what happened with any sense of scale being secondary and perspective non-existent through to the painting-as-reality (for example the accurate representation of perspective or the analysis of colour by Turner). This has also pervaded our approach to photography. But in Asia art took a very different path and the accurate use of colour was far less important, so when they want to emphasise colour in a photo they see nothing wrong with whacking up all sorts of controls and it can look downright weird.

Not just in Asia. And, even there, India is quite different from southern China (both generally bold and colourful), which is very different again from northern China and Korea (more subdued), with southeast Asia displaying many influences from both southern China and India.

In the west, you can see a distinct difference between predominantly Catholic areas and predominantly Protestant areas. Paintings in Catholic areas - much of the western Mediterranean basin, including Italy and Spain - have tended to be bright, colourful and vivid, at least over the last 500 years. This seems to have also spread and influenced works and preferences in Latin America, the Philippines and other areas (pre-Spanish/pre-Catholic Filipino paintings are quite different from post-Spanish works). You may look at some of the 400-year-old works now and think they look a bit dull, but work backwards and take away 400 years of pigment fade, though air pollution and UV light, and you find that most of these works were, in fact, originally very colourful and strongly saturated - including many details and colourful ornamentations which have actually faded away completely over time. Conversely, works from Protestant-dominated areas have tended to be darker, with more muted tones. Compare and contrast the ceiling of the Sistine Chapel in Rome with St Paul's Cathedral in London. The former is bright and colourful, whereas the latter is much more muted, dark and subdued. Of course, there are many shades in between the two broad groups. Romanticist works, for example, tend towards darker shades, but use a lot of rich, saturated colours and a rendering designed to convey emotion (e.g. motion-blurred clouds or stormy seas) rather than the what-the-eye-sees approach of the realists and neoclassicists. Of course, this may also represent a north-south divide rather than one based on philosophical differences, although philosophical outlook certainly played a role. Northern areas, with long, dark winter nights, tended towards dark and muted tones, whereas brighter and sunnier southern areas tended towards bright and saturated tones.
Logged

AreBee

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 638
Re: Canon 5Ds/R New review with file samples
« Reply #71 on: April 06, 2015, 11:58:17 am »

Logged

dwswager

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1375
Re: Canon 5Ds/R New review with file samples
« Reply #72 on: April 06, 2015, 01:43:49 pm »

You are generally right, but there are some workarounds. First, a small correction. The information recorded in the raw file does not bear any special relationship to luminance, which presupposes a spectral sensitivity that hardly ever (actually never, in my experience)  matches that on any raw channel.

With that quibble out of the way, if is possible to get the histograms of the JPEG image to approximate those of the raw image by setting the white balance of the camera to [1 1 1]. This is sometimes called "UniWB". It is not a perfect solution, but many find it workable.

Here is a discussion and some instructions for those interested in this approach.

http://blog.kasson.com/?page_id=2387

Jim

I played with UniWB previously, but I'm just not that plussed up about nailing exposure in most field situations.  Don't get me wrong, I would love and think we should already have RAW histograms that show where the exposure is between the noise floor and saturation AND I don't advocate sloppieness in technique, but since I started shooting the D810 it isn't as big a deal as it once was.  What I'm saying is that as the sensors get better providing additional DR and better noise characteristics the whole thing becomes less of problem.  I still want to get the best capture I can and not rely on post to fix it later, but when I don't, I at least know with my current camera it usually can be fixed without a lot of detriment to the finished image.
Logged

Slobodan Blagojevic

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 18128
  • When everyone thinks the same, nobody thinks
    • My website
Re: Canon 5Ds/R New review with file samples
« Reply #73 on: August 22, 2015, 12:21:59 pm »

... Slobodan, it is perhaps wise to let the photographers choose the files that represent them.

Hi Keith,

I revisited this thread today, looking for something else, when I noticed your revised post.

It sounds as if I sneaked into his studio, went through his trash bin, and selected images that he wouldn't like to be represented with. What I did is I collected a few that he posted on LuLa before, obviously something he selected to represent him. There was certainly no malicious intent on my part to take them out of context or misrepresent it in any other way.
Pages: 1 2 3 [4]   Go Up