Use Camera Profiles.. not possible?

[seanmcfoto]

Sean,
Yes.... it works exactly as you described.

Once you've run the calibration script on a RAW file in ACR, then, all you have to do is IMPORT that same RAW file into Lightroom. The Camera Calibrate setting come with it the sidecare XMP file. Cool!
p
[a href=\"index.php?act=findpost&pid=102983\"][{POST_SNAPBACK}][/a]

Thanks,
I figured it would, but not having access etc, couldn't check.
Sean.

[Kuryan Thomas]

Maybe. But is it better than no profile?
[a href=\"index.php?act=findpost&pid=102682\"][{POST_SNAPBACK}][/a]

It's my understanding from Bruce Fraser's book that ACR, and therefore probably Lightroom, does have profiles. In fact, it uses two profiles and moves between them as you shift the white balance in the image.

In other words, it is color managed - it just doesn't allow user-defined profiles.

There is some discussion of this in Andrew Rodney's book. I'll have to go re-read it.

[digitaldog]

Thomas Knoll has already rejected using camera profiles. It's come up a few times and the answer is always the same: Profiles only work for fixed lighting conditions so it won't be included.
[a href=\"index.php?act=findpost&pid=102538\"][{POST_SNAPBACK}][/a]

Not really. He's rejected 3rd party ICC profiles of which almost all have issues. He uses two profiles in Photoshop Camera Raw plug-in  (BTW, the product I'm told is no longer called ACR). These are not ICC profiles but camera profiles built based on his own software and needs for the rendering pipeline.

There's really little need for mucking around with profiles in these products anyway, we have the calibration tab for tweaking the existing profiles which, if the cameras are consistent from batch to batch, should need little tweaking (more than is necessary with the provided tools).

[digitaldog]

I'm not personally apt to listen to the aesthetic opinions of a man whose company has a website that looks like some 15 year old's Geocities nightmare circa 1994, nor to those of any person who is in the business of selling camera profiles or camera profiling software/targets. If profiles make you happy, then by all means, buy all the targets and software you can afford. 90% of the world can get by with a gray card, and that's why Adobe doesn't choose to support the ICC method in this particular instance. If you're doing scientific rather than artistic work with your camera, then this might not be the solution for you, but then again, neither would an off-the-shelf camera.
[{POST_SNAPBACK}][/a]

We sir are in total agreement!

I also would ask Mike (and anyone else) to define "accurate color" (oh boy, here he goes again).

Scene referred is accurate and often ugly. Few want or need it. Here's what the ICC has to say about this:

[a href=\"http://www.color.org/ICC_white_paper_20_Digital_photography_color_management_basics.pdf]http://www.color.org/ICC_white_paper_20_Di...ment_basics.pdf[/url]

Read it, embrace the idea of accurate color being a big ambiguous term used mostly by those in marketing. If they say accurate, is that colorimetrically correct scene referred color? Do you the image creator like it? End of story.

The question should be, is anyone using LR or Photoshop Camera Raw plug-in  having an inability with produce the desired color rendering using the tools provided (before even mucking around with profiles)?

[bjanes]

We sir are in total agreement!

I also would ask Mike (and anyone else) to define "accurate color" (oh boy, here he goes again).

Scene referred is accurate and often ugly. Few want or need it. Here's what the ICC has to say about this:

http://www.color.org/ICC_white_paper_20_Di...ment_basics.pdf

Read it, embrace the idea of accurate color being a big ambiguous term used mostly by those in marketing. If they say accurate, is that colorimetrically correct scene referred color? Do you the image creator like it? End of story.
[a href=\"index.php?act=findpost&pid=103230\"][{POST_SNAPBACK}][/a]

Andrew,

White paper #20 is anonymous, but the style of writing is similar to yours. Is this a coincidence? At any rate, it is well done.

Bill

[61Dynamic]

He uses two profiles in Photoshop Camera Raw plug-in  (BTW, the product I'm told is no longer called ACR).
[a href=\"index.php?act=findpost&pid=103228\"][{POST_SNAPBACK}][/a]

So... what are they going to call it now?

If it follows the same bad renaming process of Lightroom mabye, "Adobe Photoshop Lightroom Camera Raw" or APLCR for short. Rolls of the tongue. :D

[digitaldog]

Andrew,

White paper #20 is anonymous, but the style of writing is similar to yours. Is this a coincidence? At any rate, it is well done.

Bill
[{POST_SNAPBACK}][/a]

I wrote it with Jack Holm of HP.

This subject will hopefully be discussed in far greater detail at PMA:

[a href=\"https://pma07.bdmetrics.com/portal/ViewSession.aspx?id=1390720]https://pma07.bdmetrics.com/portal/ViewSess...aspx?id=1390720[/url]

[NikosR]

I'm not a colour scientist and maybe missing all the fine nuances and technical detalis, but for me this whole issue can be reduced to the following questions:

Is 'colorimetrically accurate' or 'objective (read unadjusted for psycho-visual adaptation)' colour representation of the scene important for your work?

Do you have full control over both the scene lighting conditions AND the display conditions of the final photo product?

If the answer to both these questions is yes then input ICC profiles might offer a solution, provided the ICC profile is correct and accurate. This will depend on profiling under the exact same light as for the scene and profiling in a correct way.

If the answer to either of these questions is no, or profiling cannot be accurately performed for whatever reason, I think that input ICC profiles is a waste of time.

It's all too easy to say that if you're colour managed then be so all the way, but I'm of the opinion that colour managing the camera input does not really address the same problem as screen and output profiling is.


As for my particular needs, the notion of having a separate profile for my 'shade-under-the-big-green-tree-in-my-garden' lighting conditions and for my 'direct-sunlight-at-12am-in-June-near-my-ochra-painted-wall' lighting conditions sounds absolutely ludicrous.

[seanmcfoto]

Not really. He's rejected 3rd party ICC profiles of which almost all have issues. He uses two profiles in Photoshop Camera Raw plug-in  (BTW, the product I'm told is no longer called ACR). These are not ICC profiles but camera profiles built based on his own software and needs for the rendering pipeline.

There's really little need for mucking around with profiles in these products anyway, we have the calibration tab for tweaking the existing profiles which, if the cameras are consistent from batch to batch, should need little tweaking (more than is necessary with the provided tools).
[a href=\"index.php?act=findpost&pid=103228\"][{POST_SNAPBACK}][/a]

Thanks for the clarification Andrew. I did mean ICC profiles, but on review it looks broader than that.

[pss]

I'm not a colour scientist and maybe missing all the fine nuances and technical detalis, but for me this whole issue can be reduced to the following questions:

Is 'colorimetrically accurate' or 'objective (read unadjusted for psycho-visual adaptation)' colour representation of the scene important for your work?

Do you have full control over both the scene lighting conditions AND the display conditions of the final photo product?

If the answer to both these questions is yes then input ICC profiles might offer a solution, provided the ICC profile is correct and accurate. This will depend on profiling under the exact same light as for the scene and profiling in a correct way.

If the answer to either of these questions is no, or profiling cannot be accurately performed for whatever reason, I think that input ICC profiles is a waste of time.

It's all too easy to say that if you're colour managed then be so all the way, but I'm of the opinion that colour managing the camera input does not really address the same problem as screen and output profiling is.
As for my particular needs, the notion of having a separate profile for my 'shade-under-the-big-green-tree-in-my-garden' lighting conditions and for my 'direct-sunlight-at-12am-in-June-near-my-ochra-painted-wall' lighting conditions sounds absolutely ludicrous.
[a href=\"index.php?act=findpost&pid=103484\"][{POST_SNAPBACK}][/a]

you are absolutely right.....color is extremely subjective and with different display mediums it is downright ridiculous to talk about the correct color....because even if you match it to a swatch, everybody still sees it differently!
i have seen art directors match my pics (the dresses in the photos) with pantone swatches..they have never asked me for  anything like that at the shoot.....
there is no way to calibrate film, and the final presentation is up to the printer so they would do the swatches.....
for fine art the most important thing is how the artist sees the color...everything else does not matter......
what i like about IP were the different profiles for different lightconditions.....

btw: i treid to download the IP demo last night....i really know my way around the software, but hte install did not work, when it finally did, it could not find the printer, then it could not activate the printer.....all this just for the demo, not even dealing with the usb plug......i really wanted to see the difference to my custom profiles on FAP on my 4800....i guess i will give it a try again sometime and see...

[eronald]

What I find entertaining is that Thomas Knoll, and all the Lightroom crowd tell me that ICC profiles aren't appropriate for cameras, and then just about everyone on this forum adds that I'm an idiot.

On the other hand, the C1 Leica crowd amongst which working photographers say thank you for my M8 profiles and send me money for them *after* testing them under various real-world lighting conditions.

Of course, payment is the sincerest and most acceptable form of flattery, but I still wonder why something that cannot and should not be done works so well in practice

Or maybe it just works well for idiots who do not use Adobe raw converters

Edmund

[BFoto]

Unless you have a higher-end camera with higher-quality manufacturing tolerances the only way to get accurate color from any raw developer is to build a custom profile of your own.

Great, got a simple workflow to achieve this for us dummies. Would love to have both the daylight and tungston presets in the calibration tool specific for my 5D.

[digitaldog]

Unless you have a higher-end camera with higher-quality manufacturing tolerances the only way to get accurate color from any raw developer is to build a custom profile of your own.

[a href=\"index.php?act=findpost&pid=111239\"][{POST_SNAPBACK}][/a]

What's accurate color? And how does a profile ensure that?

[61Dynamic]

*cough* ...pleasant color... *cough*

[digitaldog]

*cough* ...pleasant color... *cough*
[a href=\"index.php?act=findpost&pid=111248\"][{POST_SNAPBACK}][/a]

Haven't had any issues producing pleasant colors San's ICC profiles for years.

[eronald]

Haven't had any issues producing pleasant colors San's ICC profiles for years.
[a href=\"index.php?act=findpost&pid=111309\"][{POST_SNAPBACK}][/a]

Me neither. But camera profiles are an easy way to "package" a choice of rendering and deliver it to a third party. At  least, in the way they are used by the C1 converter.

Edmund

[digitaldog]

Me neither. But camera profiles are an easy way to "package" a choice of rendering and deliver it to a third party. At  least, in the way they are used by the C1 converter.

Edmund
[a href=\"index.php?act=findpost&pid=111442\"][{POST_SNAPBACK}][/a]

At DIMA, I sat in on the ICC meeting where the discussion of using ICC profiles with edits for rendering was discussed. Thomas Knoll dismissed this and I agree based on his arguments here. For one, the lack of a perceptual rendering intent is an issue. For another, we have plenty of rendering control in our converters. The ONLY thing this brings to the table is that you could try to apply the same rendering using multiple raw converters but in the end, they wouldn't match since you have to deal with bringing the data into an output referred from scene referred rendering so there goes the match. ICC profiles for rendering tweaks are poorly designed tools with poor options and add a huge layer of complexity and expense for the user. Profiles have traditionally been used to define device behavior. A capture device as complex and used in as many variable conditions as a camera makes finger-printing these devices difficult to say the least. That's why they are useful in fixed situations (think copy work). To use that model as a rendering option IMHO is a solution in search of a problem. As Thomas said about this idea at the open DIMA session on ICC profiles for his task "when all you know is a hammer, everything looks like a nail". We have far better, faster, cheaper rendering tools than ICC profiles.

[eronald]

At DIMA, I sat in on the ICC meeting where the discussion of using ICC profiles with edits for rendering was discussed. T

We have far better, faster, cheaper rendering tools than ICC profiles.
[a href=\"index.php?act=findpost&pid=111470\"][{POST_SNAPBACK}][/a]

Show me !

Do we really have better tools for editing *raw* ?

At present the advantage of the profile is that you feed it to the Raw converter (which is not PS) and, while you edit, the converter previews the rendered image, incorporating both the Raw settings edits and the "camera profile" edits, and (maybe even ?) the output profile gamut compression. In Photoshop ACR this is not possible as you first need to convert and then apply your curves.

I won't defend the "science" of camera profiles coz' there ain't one. I will defend the profile workflow, which can be applied to many things that are not strictu sensu profiles ...

I feel very uncomfortable, obviously, arguing these workflow issues with people vastly more able than I am, but given the success I seem to have with camera profiles I am starting to suspect that they actually serve some function, where before I  assumed they were only perceived to serve one.

Edmund

[digitaldog]

In Photoshop ACR this is not possible as you first need to convert and then apply your curves.
[a href=\"index.php?act=findpost&pid=111942\"][{POST_SNAPBACK}][/a]

What? You lost me.

[digitaldog]

Also, what gives you the idea ACR doesn't use profiles? It does (two).