ColorChecker Passport vs SpyderCheckr

[brntoki]

Yea, I got that. But is there anything wrong with the ExpoDisk as a WB target? Other than the additional cost?

The ExpoDisk is supposed to be aimed at the light source, so if there is more than one light source the ExpoDisk will probably not be as accurate as a white balance card, assuming you can position the card in a way that it is reflecting those multiple sources (which may be really tricky often times). However, you can probably find some other way to do what the ExpoDisk is doing. For example, I found a cap to a bottle of salad dressing that was very similar to what you might find on a light meter for incident readings. I placed that over my lens and set the camera white balance using it. It was very, very good. I learned at that time, like has been pointed out, "perfect" white balance can kill the atmosphere. "Perfect" white balance is supposed to show you true color, while the colors we are seeing in everyday life are only "true" in a certain, ever-changing, context. Sometimes we need the "hard" truth, and sometimes the contextualized truth.

[Bart_van_der_Wolf]

The ExpoDisk is supposed to be aimed at the light source, so if there is more than one light source the ExpoDisk will probably not be as accurate as a white balance card, assuming you can position the card in a way that it is reflecting those multiple sources (which may be really tricky often times).

Aiming an Exposdic at the lightsource will miss whatever local ambient light reflection is also falling on the main subject surface, unless one walks up to the subject, turns one back to it and averages the incoming light. A graycard is much easier, and in case of the Whibal the best option if one also wants to be able and use it in rainy conditions or even under water or submerged in an aquarium tank. Also very easy to use with subjects that are covered by glass, if the card can be temporarily be put behind/under the glass.

However, you can probably find some other way to do what the ExpoDisk is doing. For example, I found a cap to a bottle of salad dressing that was very similar to what you might find on a light meter for incident readings. I placed that over my lens and set the camera white balance using it. It was very, very good.

As long as it doesn't yellow with age, and assuming it is spectrally neutral to begin with, that may work. I occasionally use a square piece Opaline glass that I also use for creating an LCC for flatfielding and color cast correction. The opal layer diffuses the transmitted light with microscopic bubbles, so it is pretty neutral and diffuse.

I learned at that time, like has been pointed out, "perfect" white balance can kill the atmosphere. "Perfect" white balance is supposed to show you true color, while the colors we are seeing in everyday life are only "true" in a certain, ever-changing, context. Sometimes we need the "hard" truth, and sometimes the contextualized truth.

Measuring white balance indeed is usually just a starting point, followed by a shift in color temperature, unless one does reproductions (which would require color balancing along the entire tonescale, or a specific profile).

Cheers,
Bart

[rubencarmona]

Actually, if white balance is the only thing you need, a SpyderCube can be enough. It will last longer than a grey card, given that it's not out of carton but of pigmented synthetics. So even if you scratch it, the grey, black and white will stay neutral.

So the Spydercube is to correct the color temperature...

Tools like color checker and spydercheckr will calibrate the colors your camera interprets. So it will correct the color input of your camera to match the colors in real life. Thanks to the color patches as reference. Here both products do the same good work. It's just different methods. One is creating dng profiles, the other one is creating a preset to correct the colors in the raw metadata.

[digitaldog]

Actually, if white balance is the only thing you need, a SpyderCube can be enough.

He needs a DNG camera profile!

So it will correct the color input of your camera to match the colors in real life.

Ah, no, not really.

[Robert Boire]

The goal of a White balance tool is to get an impression of ... ambient reflections, that will determine how the subject color reflections will look....

...Also remember that the scene illumination mostly comes from illumination sources that are behind the camera and to it's sides. Measuring (averaging color reflection) of what's in front of the camera only makes sense if it is of a subject with known spectrally neutral reflection (of that illumination behind the camera), or the same color as the illumination.

Well, I guess that depends on the nature of the scene and the illumination. As you and others have pointed out, putting a grey card in the scene may not always be convenient. I would think this is especially true when there is no definable object. Or to put it another way, when the whole scene is the object (ie no tomato to use your example), such as in a landscape.  The scene may have light sources in front and behind the camera and from the sides, for example street lights at night, which is my particular case. Pointing the ExpoDisk (or any of the other similar plethora of products) at the source would be a challenge. But then again, I would think it would be a challenge for a grey card as well. Or the scene may not have illumination coming from a particular direction, but may be generally amorphous such as on an overcast day.

I would think that if the goal is to get an `approximate` calibration of white balance for later post production that pointing the ExpoDisk at the scene would give reasonable, average results. The disk will be lit up from the reflected light from the various objects in the scene, much as a grey card may be lit up by reflections from the many objects.

Hey, either way I guess I will give it a shot. Its not that expensive. Or maybe I`ll just look for a suitable bottle of salad dressing

Cheers and thanks

Robert

[Bart_van_der_Wolf]

I would think that if the goal is to get an `approximate` calibration of white balance for later post production that pointing the ExpoDisk at the scene would give reasonable, average results. The disk will be lit up from the reflected light from the various objects in the scene, much as a grey card may be lit up by reflections from the many objects.

Hi Robert,

Just using a predefined 'Daylight' white balance setting will produce a more predictable result, because it disregards any prominently colored objects in the scene that might skew the true WB determination. A gray card is different from a diffused scene measurement in that it is an object itself, with known 'color' and also intrinsically unaltered by surrounding mixed illumination, it just reflects the surrounding light and thus allows to separate the illumination from object colors.

If you want to base White balancing on an image of a diffused scene (which is all an Expodisc has to offer), you might as well use Auto WB on a defocused image of your scene and save some money ...

Cheers,
Bart

[rubencarmona]

He needs a DNG camera profile!

Why?

Ah, no, not really.

What else?

[Tim Lookingbill]

Why?

What else?

Study the subject of color constancy as it relates to "pleasing" vs "accurate to real life" color imagery. Then you'll be able to use more words in your responses here.

Also find out why landscape/portrait painters going as far back as the Renaissance used a set number of specifically named paint colors to render photo realistic images that would allow them to maintain visually balanced color for long hours at the easel whether working under candlelight or window daylight.

[rubencarmona]

I create camera-dedicated presets myself with the SpyderCheckr. But as he was most of all talking about white balance, I wanted to suggest him starting with a correct white balance. Probably that's enough for him, if not, of course he could go further...

[Isaac]

And if you are the lucky owner of a Sony a7(r), it's even easier and better: This camera allows you to set the WB in real time in live view - no need to make test shots; and you can adjust BOTH the yellow-blue AND the magenta-green balance!

As does the SLT-A35 (although the range is not large enough to set UniWB); and as-you-say adjusting K G/M until the EVF is a visual match with the scene, can work as a kind-of what I saw when I was there hint. Of course, what you saw when you were there depends on whether you were looking at the sunlit scene for a couple of minutes and then looked into the shade ;-)

I'm more likely to set camera WB with a Digital Gray Card.

[Hening Bettermann]

> Of course, what you saw when you were there depends on whether you were looking at the sunlit scene for a couple of minutes and then looked into the shade ;-)

...at which time you can establish a second WB for the shadows...:-)

[Robert Boire]

I would think that if the goal is to get an `approximate` calibration of white balance for later post production that pointing the ExpoDisk at the scene would give reasonable, average results. The disk will be lit up from the reflected light from the various objects in the scene, much as a grey card may be lit up by reflections from the many objects.

Hey, either way I guess I will give it a shot. Its not that expensive. Or maybe I`ll just look for a suitable bottle of salad dressing

Well, so much for that idea. Tried it and it really sucks.  Brought it back. What I saw on the camera display screen  (yes I know its only an approximation) using AWB or Tungsten preset (they were virtually the same) was much closer to the scene than what I got with a preset based on the ExpoDisk, which gave completely garish results. For that matter I got decent results in LR using a strategically placed styrofoam cup as a reference.

Guess I get that grey card, though it would be cooler to do an in-camera correction.

R

[Bart_van_der_Wolf]

Well, so much for that idea. Tried it and it really sucks.  Brought it back. What I saw on the camera display screen  (yes I know its only an approximation) using AWB or Tungsten preset (they were virtually the same) was much closer to the scene than what I got with a preset based on the ExpoDisk, which gave completely garish results.

Hi Robert,

Thanks for the feedback, it can help others. The outcome isn't all that surprising, but it never hurts to verify in practice.

AWB algorithms are usually much more clever than just averaging the reflected scene. They look for recognizable features/colors and use those to adjust the estimate. But nothing beats actually measuring the illumination itself (or the reflection of a known spectrally neutral subject), with the specific camera itself.

Guess I get that grey card, though it would be cooler to do an in-camera correction.

Do note that not all Gray/Grey cards are spectrally neutral (equal reflectance for all relevant wavelengths). They could also change over time, with use. That's why I like the WhiBal, easy to clean, weatherproof, very neutral, brighter than an exposure card (lower noise).

Cheers,
Bart

[deejjjaaaa]

brighter than an exposure card (lower noise).

if you make a specific shot w/ WB target you can just expose it properly (spot meter and adjust), now if you want it to be in your regular shot along with the object it is another story - but then if it is somewhere along the border you might have other issues with colorshift for example.

[Bart_van_der_Wolf]

if you make a specific shot w/ WB target you can just expose it properly (spot meter and adjust), now if you want it to be in your regular shot along with the object it is another story ...

And the latter is the normal situation, which is what I was referring to. For example with studio flash, the color temperature can change when the output power is changed. In addition, Adobe DNG profiles can exhibit a hue shift with altered exposure levels. That's when it helps to use a brighter WB target, because it will have lower noise with actual exposure levels, which allows more accurate WB sampling.

- but then if it is somewhere along the border you might have other issues with colorshift for example.

Call me an optimist, but a competent photographer is implicitly assumed ...

Cheers,
Bart

[Robert Boire]

Guess I get that grey card, though it would be cooler to do an in-camera correction.

Well that certainly worked out better, at least I am in the right ballpark. One thing I like about the WhiBal is that it is reasonably portable (fits in my back pocket) and looks more or less indestructible.

However even though the results are in the ball park..they are not quite right. Which probably leads to all sorts of questions about things I don't understand in WB in general.

In the test shots I did, the subject was sitting in artificial, indoor light (ie a living room) with bright blue clothing. The subject was holding the WhiBal. Uncorrected colors had a very yellow-orange cast as can be expected. After correction in LR using the WhiBal, the colors were slightly too cool (ie blue), particularly for the WhiBal  If anything I would have thought that if there was an overcompenstation it would have tended toward the warmer...ie remove the blue reflection on the WhiBal.  Incidentally I tried to correct using a white (but not clipped) surface in the image and got closer to where I should be (ie the correction was slightly warmer)

Or have I got all this reversed?

One thing I do not understand is LR recommends correction on a neutral gray object rather than a white object.

Thanks

[deejjjaaaa]

One thing I do not understand is LR recommends correction on a neutral gray object rather than a white object.

if they key word is neutral then it is obvious, if the key word is is gray may be to avoid the situation when you either clip "white object" /assuming that you might expose WB target along with something else and... might happen/ or (allegedly, with some cameras) expose it to the area near sensel(s) saturation where the situation with the sensel charge might get non linear (hence even a neutral object will be non neutral).

[Tim Lookingbill]

I believe as I understand from what I've read from Rags Gardner's explanation of sensor linearity that mid gray is more linear (accurate) over white which can be highly nonlinear as RGB channels go. Mid gray is just a safe zone for sensor non-linearity.

Very good explanation on sensor linearity from this site: http://rags-int-inc.com/

Click on "Technology" button on the left>click on "ETTR" button.