Re: colorsync-users Digest, Vol 19, Issue 49
I love you all. THANK YOU for being so generous with your knowledge! Happy Holidays!! : ) - Jeff Jeff Nova Chief Executive Officer Colorhythm https://colorhythm.com Mobile: +1 510-710-9590 Main: +1 415-399-9921 On Mon, Dec 5, 2022 at 12:00 PM <colorsync-users-request@lists.apple.com> wrote:
Send colorsync-users mailing list submissions to colorsync-users@lists.apple.com
To subscribe or unsubscribe via the World Wide Web, visit https://lists.apple.com/mailman/listinfo/colorsync-users or, via email, send a message with subject or body 'help' to colorsync-users-request@lists.apple.com
You can reach the person managing the list at colorsync-users-owner@lists.apple.com
When replying, please edit your Subject line so it is more specific than "Re: Contents of colorsync-users digest..."
Today's Topics:
1. White Balance data (Jeff Nova) 2. Re: White Balance data (Iliah Borg) 3. Re: White Balance data (Ben Goren) 4. Re: White Balance data (Iliah Borg) 5. Re: White Balance data (Ben Goren) 6. Re: White Balance data (Iliah Borg) 7. Re: White Balance data (Ben Goren)
----------------------------------------------------------------------
Message: 1 Date: Sun, 4 Dec 2022 22:22:27 -0800 From: Jeff Nova <jnova@colorhythm.com> To: colorsync-users@lists.apple.com Subject: White Balance data Message-ID: <5CA915DA-B0D0-4C97-AEF9-EDB37D79057A@colorhythm.com> Content-Type: text/plain; charset=utf-8
I?ve run into some data where the white balance is expressed in a manner befuddling to me:
1.46399. 1 1.52999
Is 6500K ?! There is a tint (4.2 for what that?s worth) applied to the white balance as well, which may confound this data.
Anyone have an idea of what these numbers might be? XYZ and xyY both seem wrong??
Thanks! Jeff
------------------------------
Message: 2 Date: Mon, 5 Dec 2022 06:07:48 -0500 From: Iliah Borg <iliah.i.borg@gmail.com> To: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <828dcbbd-2184-6cde-0d94-a100a56c1e5e@gmail.com> Content-Type: text/plain; charset=UTF-8; format=flowed
Hi Jeff,
Those are white balance multipliers, as the cameras write them. These multipliers equalize colour channel sensitivities with the aim that after multiplication raw values become equal on the gray.
The idea of expressing white balance as CCT + tint is a film paradigm, and it is not the best for digital (not directly measured data, difficult to calibrate, too many conversions back and forth).
On 12/5/22 01:22, Jeff Nova via colorsync-users wrote:
I?ve run into some data where the white balance is expressed in a manner befuddling to me:
1.46399. 1 1.52999
Is 6500K ?! There is a tint (4.2 for what that?s worth) applied to the white balance as well, which may confound this data.
Anyone have an idea of what these numbers might be? XYZ and xyY both seem wrong??
Thanks! Jeff _______________________________________________ Do not post admin requests to the list. They will be ignored. colorsync-users mailing list (colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription:
https://lists.apple.com/mailman/options/colorsync-users/iliah.i.borg%40gmail...
This email sent to iliah.i.borg@gmail.com
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
------------------------------
Message: 3 Date: Mon, 5 Dec 2022 07:35:31 -0700 From: Ben Goren <ben@trumpetpower.com> To: Iliah Borg <iliah.i.borg@gmail.com> Cc: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <7BA560F8-8884-4FA8-BC27-DBB571E35930@trumpetpower.com> Content-Type: text/plain; charset=utf-8
Just to elaborate a bit, they?re linear multipliers meant to be applied to the raw data recorded by the camera. Take every red pixel, multiply it by 1.46399. Leave the green pixels as they are.
They?ll be very specific to the camera model. The spectral sensitivity of a Canon 5DII is very similar to, but not identical to, that of a Canon 5DIII.
And, of course, they correspond to the reference illuminant. You mention that this is for 6500K; the values would be different (but very close) for 6501K, for D65, for an actual illuminant with a measured CCT of 6500K, and for an actual illuminant with a measured CCT of 6500K in non-neutral surroundings.
If you have a photo of a reference chart (a 24-patch ColorChecker is perfect), you can easily determine the exact figures for that scene. Use the raw developer to dump the linear raw data ? no exposure, no gamma, no white balance, no nothing except demosaicing. Create an ICC profile from the chart and ask the chart what RGB values correspond with L*=100, a*=0, b*=0; this will be a reverse lookup. Basic high school algebra will tell you what multipliers you need to make that value be R=100%, G=100%, B=100%. This will, incidentally, also give you the perfect exposure adjustment.
Note that most people who have little to no experience shooting raw inevitably wind up underexposing their images, often dramatically so. Digital cameras have an overabundance of dynamic range, so the raw converters have no trouble silently and automatically covering up the underexposure. But don?t be surprised if your image of the ColorChecker is so dark that you can?t even find where the chart is on the image. In that case, if at all possible, adjust your exposure accordingly and re-shoot the scene.
Iliah?s RawDigger analysis software is an essential tool for anybody doing serious raw photography, and the Web site for RawDigger is possibly the best guide to raw exposure techniques (and related subjects) you?ll find anywhere. If you?re going to do anything like what I described above, you?ll want to use RawDigger to extract the values from the ColorChecker to create a GCATS file that you?d hand to your ICC profiler. There are other ways to do that sort of thing ? but ?
Cheers,
b&
On Dec 5, 2022, at 4:07 AM, Iliah Borg via colorsync-users < colorsync-users@lists.apple.com> wrote:
Hi Jeff,
Those are white balance multipliers, as the cameras write them. These multipliers equalize colour channel sensitivities with the aim that after multiplication raw values become equal on the gray.
The idea of expressing white balance as CCT + tint is a film paradigm, and it is not the best for digital (not directly measured data, difficult to calibrate, too many conversions back and forth).
On 12/5/22 01:22, Jeff Nova via colorsync-users wrote:
I?ve run into some data where the white balance is expressed in a manner befuddling to me: 1.46399. 1 1.52999 Is 6500K ?! There is a tint (4.2 for what that?s worth) applied to the white balance as well, which may confound this data. Anyone have an idea of what these numbers might be? XYZ and xyY both seem wrong?? Thanks! Jeff _______________________________________________ Do not post admin requests to the list. They will be ignored. colorsync-users mailing list (colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription:
https://lists.apple.com/mailman/options/colorsync-users/iliah.i.borg%40gmail...
This email sent to iliah.i.borg@gmail.com
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com _______________________________________________ Do not post admin requests to the list. They will be ignored. colorsync-users mailing list (colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription:
https://lists.apple.com/mailman/options/colorsync-users/ben%40trumpetpower.c...
This email sent to ben@trumpetpower.com
------------------------------
Message: 4 Date: Mon, 5 Dec 2022 11:36:35 -0500 From: Iliah Borg <iliah.i.borg@gmail.com> To: Ben Goren <ben@trumpetpower.com> Cc: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <136a1dd1-2001-a558-d918-8940dd72c212@gmail.com> Content-Type: text/plain; charset=UTF-8; format=flowed
Hi Ben,
They?ll be very specific to the camera model.
In Canon's case, they will be specific to individual camera sample, too (sometimes they even vary with the sensor temperature); and the table is stored in raw files.
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
------------------------------
Message: 5 Date: Mon, 5 Dec 2022 09:41:10 -0700 From: Ben Goren <ben@trumpetpower.com> To: Iliah Borg <iliah.i.borg@gmail.com> Cc: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <B32F253A-2030-41A7-B327-6908754B47E8@trumpetpower.com> Content-Type: text/plain; charset=utf-8
I was actually thinking of asking about that ? but not variation with respect to sensor temperature but rather with age.
Has anybody done any recent measurements of years-old cameras to see how much they?re changing? I mean, of course they?re changing; entropy is inescapable. But are the changes measurable, and significant if so? Any guidelines on how often one might need to re-profile a device?
b&
On Dec 5, 2022, at 9:36 AM, Iliah Borg <iliah.i.borg@gmail.com> wrote:
Hi Ben,
They?ll be very specific to the camera model.
In Canon's case, they will be specific to individual camera sample, too (sometimes they even vary with the sensor temperature); and the table is stored in raw files.
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
------------------------------
Message: 6 Date: Mon, 5 Dec 2022 12:32:48 -0500 From: Iliah Borg <iliah.i.borg@gmail.com> To: Ben Goren <ben@trumpetpower.com> Cc: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <50d5a009-06ff-5c20-1ef6-c0ba9fa85bdf@gmail.com> Content-Type: text/plain; charset=UTF-8; format=flowed
Dear Ben,
OK, re-measured my Oly E-P2 (the oldest one I kept measurements from) after giving it a thorough sensor cleaning. Over 512 by 512 center region the measurements are the same as they were. 64 by 64 region - 0.2% difference between today's and May, 2010.
On 12/5/22 11:41, Ben Goren wrote:
I was actually thinking of asking about that ? but not variation with respect to sensor temperature but rather with age.
Has anybody done any recent measurements of years-old cameras to see how much they?re changing? I mean, of course they?re changing; entropy is inescapable. But are the changes measurable, and significant if so? Any guidelines on how often one might need to re-profile a device?
b&
On Dec 5, 2022, at 9:36 AM, Iliah Borg <iliah.i.borg@gmail.com> wrote:
Hi Ben,
They?ll be very specific to the camera model.
In Canon's case, they will be specific to individual camera sample, too (sometimes they even vary with the sensor temperature); and the table is stored in raw files.
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
------------------------------
Message: 7 Date: Mon, 5 Dec 2022 10:40:09 -0700 From: Ben Goren <ben@trumpetpower.com> To: Iliah Borg <iliah.i.borg@gmail.com> Cc: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <AC90AAB9-0E60-415C-A82C-2A3A3E123C52@trumpetpower.com> Content-Type: text/plain; charset=utf-8
Thanks ? that?s good to know! That kind of stability is impressive and reassuring ? and I wouldn?t even be surprised if it?s within the margin of error of your measurement system.
My guess is that the CMOS / CCD arrays should be exceptionally stable; my biggest concern is the stability of the color filters. But the mirror box of a camera is almost always effectively as dark as dark gets in the everyday world ? if the color filters are as stable as, say, the pigments used for archival prints, then something else will kill the camera before color accuracy suffers.
b&
On Dec 5, 2022, at 10:32 AM, Iliah Borg <iliah.i.borg@gmail.com> wrote:
Dear Ben,
OK, re-measured my Oly E-P2 (the oldest one I kept measurements from) after giving it a thorough sensor cleaning. Over 512 by 512 center region the measurements are the same as they were. 64 by 64 region - 0.2% difference between today's and May, 2010.
On 12/5/22 11:41, Ben Goren wrote:
I was actually thinking of asking about that ? but not variation with respect to sensor temperature but rather with age. Has anybody done any recent measurements of years-old cameras to see how much they?re changing? I mean, of course they?re changing; entropy is inescapable. But are the changes measurable, and significant if so? Any guidelines on how often one might need to re-profile a device? b&
On Dec 5, 2022, at 9:36 AM, Iliah Borg <iliah.i.borg@gmail.com> wrote:
Hi Ben,
They?ll be very specific to the camera model.
In Canon's case, they will be specific to individual camera sample, too (sometimes they even vary with the sensor temperature); and the table is stored in raw files.
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org <http://www.libraw.org/> www.rawdigger.com <http://www.rawdigger.com/> www.fastrawviewer.com <http://www.fastrawviewer.com/>
------------------------------
Subject: Digest Footer
_______________________________________________ colorsync-users mailing list colorsync-users@lists.apple.com https://lists.apple.com/mailman/listinfo/colorsync-users
------------------------------
End of colorsync-users Digest, Vol 19, Issue 49 ***********************************************
Just to be certain, it sounds like there is no possible route to CCT from these numbers, true? - J Jeff Nova Chief Executive Officer Colorhythm https://colorhythm.com Mobile: +1 510-710-9590 Main: +1 415-399-9921 On Mon, Dec 5, 2022 at 3:30 PM Jeff Nova <jnova@colorhythm.com> wrote:
I love you all. THANK YOU for being so generous with your knowledge! Happy Holidays!! : )
- Jeff
Jeff Nova Chief Executive Officer
Colorhythm https://colorhythm.com Mobile: +1 510-710-9590 Main: +1 415-399-9921
On Mon, Dec 5, 2022 at 12:00 PM <colorsync-users-request@lists.apple.com> wrote:
Send colorsync-users mailing list submissions to colorsync-users@lists.apple.com
To subscribe or unsubscribe via the World Wide Web, visit https://lists.apple.com/mailman/listinfo/colorsync-users or, via email, send a message with subject or body 'help' to colorsync-users-request@lists.apple.com
You can reach the person managing the list at colorsync-users-owner@lists.apple.com
When replying, please edit your Subject line so it is more specific than "Re: Contents of colorsync-users digest..."
Today's Topics:
1. White Balance data (Jeff Nova) 2. Re: White Balance data (Iliah Borg) 3. Re: White Balance data (Ben Goren) 4. Re: White Balance data (Iliah Borg) 5. Re: White Balance data (Ben Goren) 6. Re: White Balance data (Iliah Borg) 7. Re: White Balance data (Ben Goren)
----------------------------------------------------------------------
Message: 1 Date: Sun, 4 Dec 2022 22:22:27 -0800 From: Jeff Nova <jnova@colorhythm.com> To: colorsync-users@lists.apple.com Subject: White Balance data Message-ID: <5CA915DA-B0D0-4C97-AEF9-EDB37D79057A@colorhythm.com> Content-Type: text/plain; charset=utf-8
I?ve run into some data where the white balance is expressed in a manner befuddling to me:
1.46399. 1 1.52999
Is 6500K ?! There is a tint (4.2 for what that?s worth) applied to the white balance as well, which may confound this data.
Anyone have an idea of what these numbers might be? XYZ and xyY both seem wrong??
Thanks! Jeff
------------------------------
Message: 2 Date: Mon, 5 Dec 2022 06:07:48 -0500 From: Iliah Borg <iliah.i.borg@gmail.com> To: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <828dcbbd-2184-6cde-0d94-a100a56c1e5e@gmail.com> Content-Type: text/plain; charset=UTF-8; format=flowed
Hi Jeff,
Those are white balance multipliers, as the cameras write them. These multipliers equalize colour channel sensitivities with the aim that after multiplication raw values become equal on the gray.
The idea of expressing white balance as CCT + tint is a film paradigm, and it is not the best for digital (not directly measured data, difficult to calibrate, too many conversions back and forth).
On 12/5/22 01:22, Jeff Nova via colorsync-users wrote:
I?ve run into some data where the white balance is expressed in a manner befuddling to me:
1.46399. 1 1.52999
Is 6500K ?! There is a tint (4.2 for what that?s worth) applied to the white balance as well, which may confound this data.
Anyone have an idea of what these numbers might be? XYZ and xyY both seem wrong??
Thanks! Jeff _______________________________________________ Do not post admin requests to the list. They will be ignored. colorsync-users mailing list (colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription:
https://lists.apple.com/mailman/options/colorsync-users/iliah.i.borg%40gmail...
This email sent to iliah.i.borg@gmail.com
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
------------------------------
Message: 3 Date: Mon, 5 Dec 2022 07:35:31 -0700 From: Ben Goren <ben@trumpetpower.com> To: Iliah Borg <iliah.i.borg@gmail.com> Cc: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <7BA560F8-8884-4FA8-BC27-DBB571E35930@trumpetpower.com> Content-Type: text/plain; charset=utf-8
Just to elaborate a bit, they?re linear multipliers meant to be applied to the raw data recorded by the camera. Take every red pixel, multiply it by 1.46399. Leave the green pixels as they are.
They?ll be very specific to the camera model. The spectral sensitivity of a Canon 5DII is very similar to, but not identical to, that of a Canon 5DIII.
And, of course, they correspond to the reference illuminant. You mention that this is for 6500K; the values would be different (but very close) for 6501K, for D65, for an actual illuminant with a measured CCT of 6500K, and for an actual illuminant with a measured CCT of 6500K in non-neutral surroundings.
If you have a photo of a reference chart (a 24-patch ColorChecker is perfect), you can easily determine the exact figures for that scene. Use the raw developer to dump the linear raw data ? no exposure, no gamma, no white balance, no nothing except demosaicing. Create an ICC profile from the chart and ask the chart what RGB values correspond with L*=100, a*=0, b*=0; this will be a reverse lookup. Basic high school algebra will tell you what multipliers you need to make that value be R=100%, G=100%, B=100%. This will, incidentally, also give you the perfect exposure adjustment.
Note that most people who have little to no experience shooting raw inevitably wind up underexposing their images, often dramatically so. Digital cameras have an overabundance of dynamic range, so the raw converters have no trouble silently and automatically covering up the underexposure. But don?t be surprised if your image of the ColorChecker is so dark that you can?t even find where the chart is on the image. In that case, if at all possible, adjust your exposure accordingly and re-shoot the scene.
Iliah?s RawDigger analysis software is an essential tool for anybody doing serious raw photography, and the Web site for RawDigger is possibly the best guide to raw exposure techniques (and related subjects) you?ll find anywhere. If you?re going to do anything like what I described above, you?ll want to use RawDigger to extract the values from the ColorChecker to create a GCATS file that you?d hand to your ICC profiler. There are other ways to do that sort of thing ? but ?
Cheers,
b&
On Dec 5, 2022, at 4:07 AM, Iliah Borg via colorsync-users < colorsync-users@lists.apple.com> wrote:
Hi Jeff,
Those are white balance multipliers, as the cameras write them. These multipliers equalize colour channel sensitivities with the aim that after multiplication raw values become equal on the gray.
The idea of expressing white balance as CCT + tint is a film paradigm, and it is not the best for digital (not directly measured data, difficult to calibrate, too many conversions back and forth).
On 12/5/22 01:22, Jeff Nova via colorsync-users wrote:
I?ve run into some data where the white balance is expressed in a manner befuddling to me: 1.46399. 1 1.52999 Is 6500K ?! There is a tint (4.2 for what that?s worth) applied to the white balance as well, which may confound this data. Anyone have an idea of what these numbers might be? XYZ and xyY both seem wrong?? Thanks! Jeff _______________________________________________ Do not post admin requests to the list. They will be ignored. colorsync-users mailing list (colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription:
https://lists.apple.com/mailman/options/colorsync-users/iliah.i.borg%40gmail...
This email sent to iliah.i.borg@gmail.com
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com _______________________________________________ Do not post admin requests to the list. They will be ignored. colorsync-users mailing list (colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription:
https://lists.apple.com/mailman/options/colorsync-users/ben%40trumpetpower.c...
This email sent to ben@trumpetpower.com
------------------------------
Message: 4 Date: Mon, 5 Dec 2022 11:36:35 -0500 From: Iliah Borg <iliah.i.borg@gmail.com> To: Ben Goren <ben@trumpetpower.com> Cc: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <136a1dd1-2001-a558-d918-8940dd72c212@gmail.com> Content-Type: text/plain; charset=UTF-8; format=flowed
Hi Ben,
They?ll be very specific to the camera model.
In Canon's case, they will be specific to individual camera sample, too (sometimes they even vary with the sensor temperature); and the table is stored in raw files.
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
------------------------------
Message: 5 Date: Mon, 5 Dec 2022 09:41:10 -0700 From: Ben Goren <ben@trumpetpower.com> To: Iliah Borg <iliah.i.borg@gmail.com> Cc: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <B32F253A-2030-41A7-B327-6908754B47E8@trumpetpower.com> Content-Type: text/plain; charset=utf-8
I was actually thinking of asking about that ? but not variation with respect to sensor temperature but rather with age.
Has anybody done any recent measurements of years-old cameras to see how much they?re changing? I mean, of course they?re changing; entropy is inescapable. But are the changes measurable, and significant if so? Any guidelines on how often one might need to re-profile a device?
b&
On Dec 5, 2022, at 9:36 AM, Iliah Borg <iliah.i.borg@gmail.com> wrote:
Hi Ben,
They?ll be very specific to the camera model.
In Canon's case, they will be specific to individual camera sample, too (sometimes they even vary with the sensor temperature); and the table is stored in raw files.
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
------------------------------
Message: 6 Date: Mon, 5 Dec 2022 12:32:48 -0500 From: Iliah Borg <iliah.i.borg@gmail.com> To: Ben Goren <ben@trumpetpower.com> Cc: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <50d5a009-06ff-5c20-1ef6-c0ba9fa85bdf@gmail.com> Content-Type: text/plain; charset=UTF-8; format=flowed
Dear Ben,
OK, re-measured my Oly E-P2 (the oldest one I kept measurements from) after giving it a thorough sensor cleaning. Over 512 by 512 center region the measurements are the same as they were. 64 by 64 region - 0.2% difference between today's and May, 2010.
On 12/5/22 11:41, Ben Goren wrote:
I was actually thinking of asking about that ? but not variation with respect to sensor temperature but rather with age.
Has anybody done any recent measurements of years-old cameras to see how much they?re changing? I mean, of course they?re changing; entropy is inescapable. But are the changes measurable, and significant if so? Any guidelines on how often one might need to re-profile a device?
b&
On Dec 5, 2022, at 9:36 AM, Iliah Borg <iliah.i.borg@gmail.com> wrote:
Hi Ben,
They?ll be very specific to the camera model.
In Canon's case, they will be specific to individual camera sample, too (sometimes they even vary with the sensor temperature); and the table is stored in raw files.
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
------------------------------
Message: 7 Date: Mon, 5 Dec 2022 10:40:09 -0700 From: Ben Goren <ben@trumpetpower.com> To: Iliah Borg <iliah.i.borg@gmail.com> Cc: colorsync-users@lists.apple.com Subject: Re: White Balance data Message-ID: <AC90AAB9-0E60-415C-A82C-2A3A3E123C52@trumpetpower.com> Content-Type: text/plain; charset=utf-8
Thanks ? that?s good to know! That kind of stability is impressive and reassuring ? and I wouldn?t even be surprised if it?s within the margin of error of your measurement system.
My guess is that the CMOS / CCD arrays should be exceptionally stable; my biggest concern is the stability of the color filters. But the mirror box of a camera is almost always effectively as dark as dark gets in the everyday world ? if the color filters are as stable as, say, the pigments used for archival prints, then something else will kill the camera before color accuracy suffers.
b&
On Dec 5, 2022, at 10:32 AM, Iliah Borg <iliah.i.borg@gmail.com> wrote:
Dear Ben,
OK, re-measured my Oly E-P2 (the oldest one I kept measurements from) after giving it a thorough sensor cleaning. Over 512 by 512 center region the measurements are the same as they were. 64 by 64 region - 0.2% difference between today's and May, 2010.
On 12/5/22 11:41, Ben Goren wrote:
I was actually thinking of asking about that ? but not variation with respect to sensor temperature but rather with age. Has anybody done any recent measurements of years-old cameras to see how much they?re changing? I mean, of course they?re changing; entropy is inescapable. But are the changes measurable, and significant if so? Any guidelines on how often one might need to re-profile a device? b&
On Dec 5, 2022, at 9:36 AM, Iliah Borg <iliah.i.borg@gmail.com> wrote:
Hi Ben,
They?ll be very specific to the camera model.
In Canon's case, they will be specific to individual camera sample, too (sometimes they even vary with the sensor temperature); and the table is stored in raw files.
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org <http://www.libraw.org/> www.rawdigger.com <http://www.rawdigger.com/> www.fastrawviewer.com <http://www.fastrawviewer.com/>
------------------------------
Subject: Digest Footer
_______________________________________________ colorsync-users mailing list colorsync-users@lists.apple.com https://lists.apple.com/mailman/listinfo/colorsync-users
------------------------------
End of colorsync-users Digest, Vol 19, Issue 49 ***********************************************
There is, Adobe DNG SDK suggests one way to approximate it, and there are other ways, too. With Canon one can do interpolation as the raw files contain data like this: 10900, {2.92571f, 1.0f, 1.25183f, 1.0f}, 10000, {2.86034f, 1.0f, 1.28160f, 1.0f}, 8300, {2.70185f, 1.0f, 1.36533f, 1.0f}, 7000, {2.53465f, 1.0f, 1.46705f, 1.0f}, 6000, {2.36490f, 1.0f, 1.57538f, 1.0f}, 5600, {2.28062f, 1.0f, 1.63578f, 1.0f}, 5200, {2.18337f, 1.0f, 1.70383f, 1.0f}, 4700, {2.06452f, 1.0f, 1.81883f, 1.0f}, 4200, {1.92120f, 1.0f, 1.96923f, 1.0f}, 3800, {1.78397f, 1.0f, 2.11570f, 1.0f}, 3500, {1.67320f, 1.0f, 2.26549f, 1.0f}, 3200, {1.54683f, 1.0f, 2.45564f, 1.0f}, 3000, {1.46286f, 1.0f, 2.63239f, 1.0f}, 2800, {1.36533f, 1.0f, 2.79781f, 1.0f}, 2400, {1.16232f, 1.0f, 3.21003f, 1.0f} On 12/5/22 23:00, Jeff Nova via colorsync-users wrote:
Just to be certain, it sounds like there is no possible route to CCT from these numbers, true? - J
Jeff Nova Chief Executive Officer
Colorhythm https://colorhythm.com Mobile: +1 510-710-9590 Main: +1 415-399-9921
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com
It also, of course, depends on what the goal is. Jeff, are you trying to figure out the CCT based on known multipliers, or trying to determine multipliers based on a known CCT, or … ? And, regardless, for what purpose? That is, once you know CCT / multipliers, then what? I ask because neither figure is likely to be especially useful in isolation. If the goal is to apply white balancing to an image, either number will get you in the ballpark and likely “close enough” for any casual usage — but neither will be “good enough” for color-critical work. I’ll also toss out there that, if you know a camera’s spectral response, you can, in theory, get a not-miserable spectral measurement of a scene’s illuminant from a photograph of a ColorChecker. You certainly won’t be able to identify Fraunhofer lines or anything like that; however, you should absolutely be able to determine the type of illuminant (actual daylight, incandescent, flash, fluorescent, etc.). And, for illuminants with variable spectra (especially daylight and incandescent), determining the precise color temperature (and thus exactly characterizing the spectrum) is also achievable. This should be extendable; you won’t get the precise spikes of a fluorescent source, but, if you know that the source is one of a known set to pick from, picking the right one should be trivial. I haven’t tried it, but, in principle, you should also be able to suss out ratios of mixed-lighting environments, too ... b&
On Dec 5, 2022, at 9:08 PM, Iliah Borg via colorsync-users <colorsync-users@lists.apple.com> wrote:
There is, Adobe DNG SDK suggests one way to approximate it, and there are other ways, too.
With Canon one can do interpolation as the raw files contain data like this:
10900, {2.92571f, 1.0f, 1.25183f, 1.0f}, 10000, {2.86034f, 1.0f, 1.28160f, 1.0f}, 8300, {2.70185f, 1.0f, 1.36533f, 1.0f}, 7000, {2.53465f, 1.0f, 1.46705f, 1.0f}, 6000, {2.36490f, 1.0f, 1.57538f, 1.0f}, 5600, {2.28062f, 1.0f, 1.63578f, 1.0f}, 5200, {2.18337f, 1.0f, 1.70383f, 1.0f}, 4700, {2.06452f, 1.0f, 1.81883f, 1.0f}, 4200, {1.92120f, 1.0f, 1.96923f, 1.0f}, 3800, {1.78397f, 1.0f, 2.11570f, 1.0f}, 3500, {1.67320f, 1.0f, 2.26549f, 1.0f}, 3200, {1.54683f, 1.0f, 2.45564f, 1.0f}, 3000, {1.46286f, 1.0f, 2.63239f, 1.0f}, 2800, {1.36533f, 1.0f, 2.79781f, 1.0f}, 2400, {1.16232f, 1.0f, 3.21003f, 1.0f}
On 12/5/22 23:00, Jeff Nova via colorsync-users wrote:
Just to be certain, it sounds like there is no possible route to CCT from these numbers, true? - J Jeff Nova Chief Executive Officer Colorhythm https://colorhythm.com Mobile: +1 510-710-9590 Main: +1 415-399-9921
-- Best regards, Iliah Borg LibRaw, LLC www.libraw.org www.rawdigger.com www.fastrawviewer.com _______________________________________________ Do not post admin requests to the list. They will be ignored. colorsync-users mailing list (colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/colorsync-users/ben%40trumpetpower.c...
This email sent to ben@trumpetpower.com
participants (3)
-
Ben Goren
-
Iliah Borg
-
Jeff Nova