Pages: 1 2 3 [4] 5 6   Go Down

Author Topic: D810 support?  (Read 30773 times)

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: D810 support?
« Reply #60 on: July 20, 2014, 04:09:44 pm »

Depends on how one defines colour space. Mathematically and from the point of view of physics, no, it is not a space because the metrics is not defined.
That's where I'm coming from. How are the metrics defined so the color space is defined? Again, I'm told this is not completely precise, there are good and not as good choices, but no single correct conversion.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: D810 support?
« Reply #61 on: July 20, 2014, 04:18:46 pm »

> How are the metrics defined so the color space is defined?

Commonly, in colour the metrics is defined if the conversion to LAB/XYZ/Munsell is defined.

> there are good and not as good choices, but no single correct conversion.
A number of manufacturers have conversion matrices and twists included right in the raw files' metadata. Some are even calibrated to particular sensor sample, and additionally take into account shooting conditions, like sensor temperature. Is that conversion data is always metrically good / appeals to my taste? - No. So you are quite right, the answer is definitely "no single correct conversion exists" and more importantly - indeed, "there are good and not as good choices".
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: D810 support?
« Reply #62 on: July 20, 2014, 04:23:02 pm »

Commonly, in colour the metrics is defined if the conversion to LAB/XYZ/Munsell is defined.
To LAB/XYZ/Munsell from what? I guess the answer is, it depends. So thanks, it seems you're confirming what I thought to be the case.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: D810 support?
« Reply #63 on: July 20, 2014, 04:25:53 pm »

Lab/XYZ/Munsell from RGB.
I'm making 2 points: in case of D810 as well as in case of many other cameras, neither raw format, nor colour transform are offering any excuse for delayed camera support in raw converters.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: D810 support?
« Reply #64 on: July 20, 2014, 04:30:00 pm »

Lab/XYZ/Munsell from RGB.
I'm making 2 points: in case of D810 as well as in case of many other cameras, neither raw format, nor colour transform are offering any excuse for delayed camera support in raw converters.
What RGB would that be is the question (which has nothing to do with an excuse for delayed camera support in raw converters)?
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: D810 support?
« Reply #65 on: July 21, 2014, 02:11:31 am »

> What RGB would that be is the question
Camera RGB. Once the metrics is introduced (that's what profiling in fact is) the conversion is possible. This metrics is of course an approximation, like it is with any physical objects. "Correct" profiles exist only in theory, be it a printer or a sensor. One of the reasons is that the physical sensor is rather far from ideal, meaning it is not a flat field - different pixels of the same colour have slightly different responses depending on their location and manufacturing tolerances (same as different nozzles do not produce the exact same droplets, and paper is not perfectly uniform too). If the target is shot, then rotated and shot again, colour transforms calculated from both shots render noticeable differences (larger than 2 dE). However if one compares different modern profiling methods, the difference in colour transforms is below 4 dE, enough for practical purposes. It means we have mostly only good choices, given we know what we are doing. The difference between sensor samples reaches 6%, plus we have different lenses which affect the colour registration too. So if one wants a good profile he better makes it himself.
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: D810 support?
« Reply #66 on: July 21, 2014, 02:30:11 am »

Hi

A question to Iliah: Do you need a representative camera that you can shoot to be able to support it? With "correct colour".

Your discussion of tolerances is quite revealing. Thanks a lot!

Best regards
Erik




> What RGB would that be is the question
Camera RGB. Once the metrics is introduced (that's what profiling in fact is) the conversion is possible. This metrics is of course an approximation, like it is with any physical objects. "Correct" profiles exist only in theory, be it a printer or a sensor. One of the reasons is that the physical sensor is rather far from ideal, meaning it is not a flat field - different pixels of the same colour have slightly different responses depending on their location and manufacturing tolerances (same as different nozzles do not produce the exact same droplets, and paper is not perfectly uniform too). If the target is shot, then rotated and shot again, colour transforms calculated from both shots render noticeable differences (larger than 2 dE). However if one compares different modern profiling methods, the difference in colour transforms is below 4 dE, enough for practical purposes. It means we have mostly only good choices, given we know what we are doing. The difference between sensor samples reaches 6%, plus we have different lenses which affect the colour registration too. So if one wants a good profile he better makes it himself.
Logged
Erik Kaffehr
 

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: D810 support?
« Reply #67 on: July 21, 2014, 02:47:23 am »

Dear Erik,

> Do you need a representative camera that you can shoot to be able to support it? With "correct colour".

I need at least 2 cameras to be reasonably sure I do not have an "outlying" sample instead of a representative. It is quite a headache actually to borrow them. Without having a representative camera in hands one can try to use Imaging Resource ColorChecker SG target shots, and perform an image matching trick, that is calculate an RGB-to-RGB transform from raw RGB and RGB obtained from "native" raw converter (Nikon Capture, Canon DPP,...). This usually gives a decent approximation.
Logged

Lupin

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 61
Re: D810 support?
« Reply #68 on: July 21, 2014, 06:59:06 am »

Strange that C1Pro v7.2.3 doesn't include D810 support.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: D810 support?
« Reply #69 on: July 21, 2014, 09:36:53 am »

> What RGB would that be is the question
Camera RGB.
How is camera RGB defined?
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: D810 support?
« Reply #70 on: July 21, 2014, 10:59:49 am »

How is camera RGB defined?
Camera RGB is a generic name for sensor output.
Logged

Vladimirovich

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1311
Re: D810 support?
« Reply #71 on: July 21, 2014, 11:18:24 am »

The current behavior isn't acceptable IMHO
it is acceptable because you (you personally) do nothing serious to change it and the situation is supported by those who make raw converters (and those who make sure that dcraw code lives - that way mr. Coffin himself is probably the #1 reason why) because they just provide support for the new cameras and they do that in acceptable timeframe for 99 out 100 purchasers who are using non OOC JPG/non OEM raw converters... there is ~zero pressure of any kind on camera manufacturers to provide the fully documented access to their raw files in whatever format they like... just be a realist.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: D810 support?
« Reply #72 on: July 21, 2014, 11:33:05 am »

Camera RGB is a generic name for sensor output.
OK but how is that defined, what flavor of RGB is it exactly, how does one raw converter company vs. another get to that recipe? Or is this completely understood by each and not ambigous? That's the root of my question.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: D810 support?
« Reply #73 on: July 21, 2014, 11:35:06 am »

it is acceptable because you (you personally) do nothing serious to change it and the situation is supported by those who make raw converters (and those who make sure that dcraw code lives - that way mr. Coffin himself is probably the #1 reason why) because they just provide support for the new cameras and they do that in acceptable timeframe for 99 out 100 purchasers who are using non OOC JPG/non OEM raw converters... there is ~zero pressure of any kind on camera manufacturers to provide the fully documented access to their raw files in whatever format they like... just be a realist.
If you'd read a bit of the posts here and elsewhere, you'll see that it isn't acceptable to others besides myself. But it doesn't matter at this point, you simply don't care about the burdens of other's facing their inability to use the data as they so desire for absolutely no justifiable reason.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: D810 support?
« Reply #74 on: July 21, 2014, 11:39:19 am »

> how is that defined

By the response necessary. Colour filter arrays in front of the sensor are not arbitrary filters.
 
> what flavor of RGB is it exactly

This question is more or less like "what flavour is scanner RGB". I guess you know the answer. Filter flavour.
 
> how does one raw converter company vs. another get to that recipe?

Three ways I mentioned above.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: D810 support?
« Reply #75 on: July 21, 2014, 11:44:11 am »

> how is that defined
By the response necessary. Colour filter arrays in front of the sensor are not arbitrary filters.
I didn't think they were arbitrary, I'm wondering how their behavior so to speak, is defined before conversion to the internal RGB color space used for processing.
 
Quote
> what flavor of RGB is it exactly
This question is more or less like "what flavour is scanner RGB". I guess you know the answer. Filter flavour.
True, in terms of taking a target of known color values, scanning it which can vary drastically, then building a profile. I thought a bayer array was a bit more complicated than that to define.
 
Quote
> how does one raw converter company vs. another get to that recipe?
Three ways I mentioned above.
OK, so if I understand you, that varies and that goes back to the statement that there are good and less good ways to do this, but assumptions have to be made along the way. Is that fair?
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Vladimirovich

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1311
Re: D810 support?
« Reply #76 on: July 21, 2014, 11:56:05 am »

If you'd read a bit of the posts here and elsewhere, you'll see that it isn't acceptable to others besides myself.

that's just noise - manufacturers pay zero attention, because they know that even you will buy their cameras in the end.

But it doesn't matter at this point, you simply don't care about the burdens of other's facing their inability to use the data as they so desire for absolutely no justifiable reason.

I do not buy a new camera (not that I am buying a lot of cameras, like some people here) until there is at least one raw converter that I can use & want to use for it - the rest is just emotions that will die out now even for most people who complained (now that ACR RC was made available last week)... so for me this is a non issue, but I want to try to represent the opposite view as a matter of principle.

Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: D810 support?
« Reply #77 on: July 21, 2014, 12:03:39 pm »

that's just noise - manufacturers pay zero attention, because they know that even you will buy their cameras in the end.
Fine, we'll agree to disagree and I'll continue to believe based on your posts, you just like those manufacturers don't really care about photographers and their issues or this industry.
Quote
I do not buy a new camera (not that I am buying a lot of cameras, like some people here) until there is at least one raw converter that I can use & want to use for it - the rest is just emotions that will die out now even for most people who complained (now that ACR RC was made available last week)... so for me this is a non issue, but I want to try to represent the opposite view as a matter of principle.
Yes, it's all about you and of course everyone who buys a camera before the converter they hope to use is just a fool. I wonder what Nikon or Canon or the camera suppliers would say about that. Like you, they don't give a s^%t,.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: D810 support?
« Reply #78 on: July 21, 2014, 02:20:41 pm »

> how their behavior so to speak, is defined before conversion to the internal RGB color space
The guides for creating CFAs are based on a trade-off between tolerable metrics ambiguity (including metameric failures) and low noise. To have low noise and better colour response stability in shadows the filters deviate from L-I, and also they are more transparent meaning lower separation. We are now on the third generation of CFAs, first being very strict and thus cameras were more noisy, the second was too relaxed and resulted in mushy and plastic colours, now it is in-between, which is possible due to lower read noise. Incidentally larger pixels allow for stricter CFAs due to "lower" noise, and so are larger sensors (one of the reasons semi-MF sensors are more joy to me). There is a disconnect between filter design and colour transforms design - good colour transforms are based on experiments and not on design data. It is like it used to be with lenses, resolution was determined through tests and not based on calculations, and even now the calculated MTFs can't be trusted much. Internal space is not always RGB; and even if it is an RGB it is often that quite a lot of calculations is performed over the camera RGB.

Scanner filters were dealing with 4 main types of dyes, slide, neg, photopaper cmy, and press cmyk. Those can't be separated well enough out of the box with the same set of filters. But profiling takes a very good care of the diversity making it possible to reproduce all types of originals with a single profile made from a combined target which contains different emulsion types. The situation with Bayer filters is similar, given we normalize the target by a flat field before profiling, and we profile using targets encompassing 10 stops (2 shots of SG 4 stops apart with a black trap /SpyderCUBE/ included in the scene is a good way, but the first shot needs to be exposed so that raw values for the white patch are about 227 /that corresponds to 242 in gamma 2.2/); or have an adequate setup with a monochromator. More, for most of the practical purposes even colour separation we have with a film is quite adequate to cover for a wide variety of light and scenery. On a side note, digital photography introduced a false notion that filters on the lens are not necessary anymore.

Good and bad ways do exist, but there is a lot of information allowing to avoid bad ways. Assumptions currently (since 1999 at least) are not necessary because the math. models and calculation methods we have are quite good. Ironically, not so good for printing.
Logged

Vladimirovich

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1311
Re: D810 support?
« Reply #79 on: July 21, 2014, 04:18:51 pm »

of course everyone who buys a camera before the converter they hope to use is just a fool.

no, there are

1) curious people willing to play using tools available and wait for support or people like recently joined here IB who can create their own quality profiles
2) OOC JPG shooters or those willing to do this and wait for some non OEM raw converter of their choice
3) OEM raw converter users or those who are OK to use it and wait for some non OEM raw converter of their choice

and that covers 99.5% early adopters... so for a mass user it is not a problem and so it is not a problem for manufacturers... and the fools are people who earn money using cameras and who sell their old cameras to afford a brand new one w/o waiting for it to be supported by tools used in their "workflow", those are fools... I can hardly imagine the need to sell in a rush for example D800/800e to buy D810 if your $$$ do not allow to buy it just to play... in most cases that just GAS w/o $$$ to support it.
Logged
Pages: 1 2 3 [4] 5 6   Go Up