Luminous Landscape Forum

Raw & Post Processing, Printing => Colour Management => Topic started by: Ligament on September 12, 2014, 02:48:48 pm

Title: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: Ligament on September 12, 2014, 02:48:48 pm
Howdy all,

For those of you with experience with all three, or good knowledge of all three; which provides the most accurate results?

Thanks!
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: digitaldog on September 12, 2014, 02:52:55 pm
I'd be very careful with the term accurate in this context. What I can say is that a custom DNG camera profile can help (I've yet to see one I've built not improve what I consider pleasing color) compared to the canned profiles supplied for my Canon's. I've only used the Passport which is a great product in terms of the targets and design. The DataColor product seems to be a bit of a kludge in how it sets the LR controls (HSL). No experience with Qpcard. You don't need to make many of these custom profiles FWIW, just for each differing kind of illuminant (daylight, Fluorescent, tungsten).
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: deejjjaaaa on September 12, 2014, 03:14:08 pm
Howdy all,

For those of you with experience with all three, or good knowledge of all three; which provides the most accurate results?

Thanks!

hint - no vendor is using neither 3 to build any OEM camera profiles... why'd it be so  ::) ?
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: aaronchan on September 12, 2014, 03:40:53 pm
I have used both ColorChecker Passport and the QCard.
I like the product design of the Passport. It has a hard shell which can protect the patches.
Also it has some other creative patches which you can really use them in a real world photography job.

But for the quality, I found out that the QCard has more pleasing color to me.
Specially on the blue. It definitely create more "accurate" color than the Passport.
But this is according to my eyes only. I have never use any tool to compare and contrast both products.
The downside is the Qcard is kind of flimsy. It is covered by 2 card board and that's it!

For fine art copy work, I like to use the QCard. But for location work, it would be the other way around.

Aaron
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: deejjjaaaa on September 12, 2014, 04:07:32 pm
I have never use any tool to compare and contrast both products.



I was given the following suggestion once -> shoot both in one frame, build profile using QPCard, build profile using XRite, convert that same shot using both profiles (pay attention to what converter you are using - if ACR/LR make sure to use a proper conversion settings to avoid hidden curves & other corrections etc), bring into babelcolor patchtool, get dE(whatever)'s from the opposite (to what was used to build a profile) target for each conversion... compare... they key thing is to avoid testing on the same target that was used to build a profile - hence 2 targets in one frame, cross check
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: ErikKaffehr on September 12, 2014, 04:25:22 pm
Hi,

I would vote for Adobe DNG editor, which is also completely free.

The ColorChecker Passport software is the easiest to use. The DNG Profile editor is almost as easy to use but builds more complex profiles, it seems. The DNG Profile Editor also have tools to tweak the profile.

QPCard uses a different set of patches, that may or may not work better. It may also build profiles from the ground up, while DNG Profile Editor seems to just adjust an existing base profile.

I have tested all three products with my P45+ back and my Sony Alpha 99 SLT:

http://echophoto.dnsalias.net/ekr/index.php/photoarticles/79-p45-colour-rendition?start=3
http://echophoto.dnsalias.net/ekr/index.php/photoarticles/79-p45-colour-rendition?start=6
http://echophoto.dnsalias.net/ekr/index.php/photoarticles/79-p45-colour-rendition?start=5


At the end I stayed with Adobe DNG editor.

Best regards
Erik



Howdy all,

For those of you with experience with all three, or good knowledge of all three; which provides the most accurate results?

Thanks!
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: digitaldog on September 12, 2014, 04:27:14 pm
I would vote for Adobe DNG editor, which is also completely free.
So is the Passport software, you just need a target. IF you own a 24 patch Macbeth, it's free like the Adobe DNG editor. If you don't own a target, I'd suggest the Passport insetad of the MacBeth (more bang for the buck).
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: aaronchan on September 13, 2014, 12:21:02 am
So is the Passport software, you just need a target. IF you own a 24 patch Macbeth, it's free like the Adobe DNG editor. If you don't own a target, I'd suggest the Passport insetad of the MacBeth (more bang for the buck).

But I've heard the value of the classic (big) 24 patch colorchecker is different than the passport.
Both 24 color patches contain different values. Is that going to affect the DNG profile result?
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: brntoki on September 13, 2014, 03:15:24 am
So is the Passport software, you just need a target. IF you own a 24 patch Macbeth, it's free like the Adobe DNG editor. If you don't own a target, I'd suggest the Passport insetad of the MacBeth (more bang for the buck).

Why can't a correctly created digital file be displayed on a well calibrated monitor and be use for creating a DNG profile with Adobe's editor? Seems like it should work well.
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: ErikKaffehr on September 13, 2014, 04:39:38 am
Hi,

DNG Profile Editor makes that automatically, but you can adjust them manually. I would guess that the automated solution optimises the conversion matrices while manual editing makes more local adjustments.

In my tests, using an IT8 target for evaluation DNG Profile Editor did a very good job.

Best regards
Erik

Why can't a correctly created digital file be displayed on a well calibrated monitor and be use for creating a DNG profile with Adobe's editor? Seems like it should work well.
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: mac_paolo on September 13, 2014, 05:17:43 am
I tried both the old pre-digital ColorChecker and the more recent QpCard.
Despite the fact that the manufacturer of the latter seems to be dead (you now have to convert do ACR 5.4 compatible DNG to properly calibrate), I found the QpCard to provide the most pleasing results, especially with nasty artificial lightning conditions.

To my eyes (Dell U2711 with i1Display Pro accurate monthly profiling), QpCalibration provides the best color fidelity. The only annoyances are the fact that WB greatly differs from ACR in Tint setting and that you can't create dual illuminant profile, so be prepare to make a LOT of profiles with very different lighting conditions, though a generic "Sun" profiles work fine for the majority of sunny days at almost any hour tom morning to late afternoon.
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: brntoki on September 13, 2014, 08:39:08 am
Hi,

DNG Profile Editor makes that automatically, but you can adjust them manually. I would guess that the automated solution optimises the conversion matrices while manual editing makes more local adjustments.

In my tests, using an IT8 target for evaluation DNG Profile Editor did a very good job.

Best regards
Erik


I'm not sure you know what I meant. I mean to display a Macbeth target on screen and shoot that to obtain a profile. It would only be a base profile and one would need to have a gray balance card to set the white point from any particular shoot, but it seems it should give correct colors, no?
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: AFairley on September 13, 2014, 10:14:47 am
Hi,

DNG Profile Editor makes that automatically, but you can adjust them manually. I would guess that the automated solution optimises the conversion matrices while manual editing makes more local adjustments.

In my tests, using an IT8 target for evaluation DNG Profile Editor did a very good job.

Best regards
Erik


Erik, to clarify, you used the 24-patch target to make the profile and then used the IT8 target as a reference to check the resulting profile, right?  I don't see a way in the editor to use the IT8 target in the editor itself.  Thx.
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: deejjjaaaa on September 13, 2014, 10:16:22 am
and that you can't create dual illuminant profile
who prevents you from creating it manually from 2 single illuminant profiles ?
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: digitaldog on September 13, 2014, 11:53:37 am
But I've heard the value of the classic (big) 24 patch colorchecker is different than the passport.
The 24 patches are within the Passport target on one 'page' (there are only 24) and are the same as the classic Macbeth. And there is the small 24 patch Macbeth that can also be used with the Passport or Adobe software and looks to me the same (I haven't measured each).

AFAIK, you can't build a DNG profile with an IT8 (that's news to me), that target is for creating scanner profiles. Maybe the 24 patches are in there somewhere and can be extracted?
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: deejjjaaaa on September 13, 2014, 12:21:04 pm
The 24 patches are within the Passport target on one 'page' (there are only 24) and are the same as the classic Macbeth.

no, he was referring to http://www.rmimaging.com/information/ColorChecker_Passport_Technical_Report.pdf -> page 3

I guess the precision of making dcp profiles by an average joe shall not suffer from those differences.

Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: Rhossydd on September 13, 2014, 02:36:17 pm
And there is the small 24 patch Macbeth that can also be used with the Passport or Adobe software and looks to me the same (I haven't measured each).
They do vary. Have a look at http://www.babelcolor.com/main_level/ColorChecker.htm which gives some hard data. I also did measure all mine to see how they varied, but haven't the full data set immediately available, but a quick look suggests variations from 0.6 - 4.0 dE between patches on different CC24s
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: digitaldog on September 13, 2014, 04:50:27 pm
They do vary. Have a look at http://www.babelcolor.com/main_level/ColorChecker.htm which gives some hard data.
Here's the report from that software comparing my Passport to a ColorChecker Mini:

AVERAGE
All samples:    0.77
Best 90%:    0.70
Worst 10%:    1.56

STANDARD-DEVIATION
All samples:    0.40
Best 90%:    0.34
Worst 10%:    0.03

MAXIMUM-ERROR
10th percentile:        0.32
Median (50th perc.):        0.60
90th percentile:        1.47
95th percentile:        1.52
Of all samples:        1.59
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: ErikKaffehr on September 13, 2014, 05:04:53 pm
Yes Sir, you are absolutely right!

The reason I did this evaluation was to find out about the accuracy of the profiles. The ColorChecker target has just 24 fields and the profiles generated by Color Checker Passport or Adobe DNG Profile Editor are automatically adjusted to reproduce these 24 fields optimally. It is quite possible that other colours would be reproduced diffeently.

Now, the color checker colours are chosen very carefully, the skin tone fields for instance have spectral reflectance very similar to human skin. So the ColorChecker is a good target.

The test I was to use a well known reference target with far more colours. The chart I used was one from Wolfgang Faust and it included Lab values for each field.

So I made a carefully exposed shot of the IT8 target and a ColorChecker. Exposure was adjusted so the ColourChecker's light grey and black fields were rendered properly and the files were exported as 16 bit tiffs in Profoto RGB. These TIFFs were than read using Argyll CMS and the lab values were saved as a text files (CGATS? I don't recall). The LAB values were than compared with the reference values coming with the IT8 target.

What I have found?

The Adobe Standard Profile was pretty exact on my Sony Alpha 99, but using the Adobe DNG Profile improved it very slightly. The difference was so small that I stayed with the Adobe Standard Profile.

With my P45+ I also compared with Capture One. On the P45+ the Adobe DNG Editor generated profile was slightly more accurate than the Adobe Standard profile. Capture One, using linear curve, gave more accurate colour than Lighroom with either profile. The default "film curve" gave worse results than Lightroom.  These data were evaluated using CIEDE 2000 DeltaE* values. These data include luminance and luminance is certainly affected by choice of gradation curve.

The Sony Alpha in LR was more accurate than the P45+ in either C1 or LR. C1 was less accurate than LR on the Sony Alpha SLT.

The result? I use Adobe Standard profile on the Sony. With the P45+ I use a DNG profile tuned automatically with the DNG Profile Editor.


Best regards
Erik

Erik, to clarify, you used the 24-patch target to make the profile and then used the IT8 target as a reference to check the resulting profile, right?  I don't see a way in the editor to use the IT8 target in the editor itself.  Thx.
Title: DNG Profile Editor vs. Color Checker Passport software
Post by: ErikKaffehr on September 13, 2014, 05:41:16 pm
Hi,

I have used both solutions and also QPCard's software. Of the three, Xrite Passport is the easiest to use, DNG Profile Editor is most flexible. QPCard is sort of a different animal.

My understanding is that both Adobe DNG Profile Editor and Xrite Passport start from a base profile, like Adobe Standard that probably has been developed by using a spectral sensivity data for the sensor and tweaked by some means. So, my guess is that both softwares make corrections on top of the Adobe Standard profile, or whatever base profile is used.

Both softwares can do this automatically, Xrite Passport locates the ColorChecker fields automatically, while DNG Profile Editor needs some assistance.

The main difference between the Adobe DNG Profile Editor is that the DNG Profile Editor can edit/tweak profiles, while with the Color Checker Passport it is more like you get what you got. Personally, I feel DNG Profile Editor is more accurate.

Best regards
Erik
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: Stefan Ohlsson on September 14, 2014, 04:37:28 am
I tried both the old pre-digital ColorChecker and the more recent QpCard.
Despite the fact that the manufacturer of the latter seems to be dead (you now have to convert do ACR 5.4 compatible DNG to properly calibrate), I found the QpCard to provide the most pleasing results, especially with nasty artificial lightning conditions.


You don't have to convert a file to DNG before importing it to QPCard, you can use the native raw file. But I do agree that QPcard makes profiles that are pleasing. And the possibility to choose contrast curve and gamut mapping makes it easy to customize how I want my profile. In general, I make profiles with a little less saturated colours when I make a profile with QPcard, compared with Passport. It is then easy to adjust the colour saturation in Lightroom or ACR.
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: Ligament on September 14, 2014, 03:40:13 pm
Is qpcard still in business? I've emailed them multiple times for tech support (qpcard delaminated, color chart fell off cardboard book), no replies...
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: mac_paolo on September 14, 2014, 04:13:16 pm
You don't have to convert a file to DNG before importing it to QPCard, you can use the native raw file. But I do agree that QPcard makes profiles that are pleasing. And the possibility to choose contrast curve and gamut mapping makes it easy to customize how I want my profile. In general, I make profiles with a little less saturated colours when I make a profile with QPcard, compared with Passport. It is then easy to adjust the colour saturation in Lightroom or ACR.
Actually you may need to convert to DNG beforehand in order to make the profiles appear under ACR/Lightroom, depending on the camera, especially when it's 1-2 years or newer. I speak by experience.
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: mac_paolo on September 14, 2014, 04:15:10 pm
Is qpcard still in business? I've emailed them multiple times for tech support (qpcard delaminated, color chart fell off cardboard book), no replies...
Same here. They're probably dead, but the software still works under Mavericks. Fingers crossed.
I wouldn't want to switch back to the dated ColorChecker once again.
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: Stefan Ohlsson on September 14, 2014, 05:00:11 pm
They're probably dead, but the software still works under Mavericks. Fingers crossed.
I wouldn't want to switch back to the dated ColorChecker once again.
I know both the developer and the owner of the company. I'm not in any way involved in the company, but as far as I know they are still in business.
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: ErikKaffehr on September 14, 2014, 05:36:30 pm
Hi,

Just to say, I bought QPCard, sometimes last fall. It was delivered promptly, and I liked the results. On the other hand, I also found that Adobe DNG Profile Editor worked best for me.

I would guess that QPCard may work better with skin tones as it has more fields with subtle colours.

Best regards
Erik

I know both the developer and the owner of the company. I'm not in any way involved in the company, but as far as I know they are still in business.
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: Rhossydd on September 14, 2014, 05:40:03 pm
I know both the developer and the owner of the company. I'm not in any way involved in the company, but as far as I know they are still in business.
It might be helpful to give him/them a nudge to get on top of replying to emails then. Over the last year there have been several mentions of poor(non-existent) service from QP on Lula.
Which is a shame as just about everyone that has bought one thinks they offer the most pleasing camera profiles, me included.
Title: Re: X-rite Colorchecker Passport vs. Datacolor Spyder ColorCHECKR vs. QPcard
Post by: Stefan Ohlsson on September 15, 2014, 02:19:12 am
It might be helpful to give him/them a nudge to get on top of replying to emails then. Over the last year there have been several mentions of poor(non-existent) service from QP on Lula.
Which is a shame as just about everyone that has bought one thinks they offer the most pleasing camera profiles, me included.
I have, but will do that again.

Best/Stefan