Drew of ProPhotohome offers a new tool for white reference.

[Asher Kelman]

No matter what camera one chooses, if color images are needed a complete color aware chain is required:

Subject to Lens toSensor to translation firmware with color remapping to white point reference shot and delivered as an sRGB file

or

Subject to lens to Sensor to firmware delivering a RAW file in AdobeRGB or other wide space RGB color space to file-translation with color remapping to reference white balance shot and delivered as a TIFF,PSD or JPG file.

In each case, to view your image, the graphics card/smart monitor has to remap the file to the color space of your monitor* and then to remap the file to the color space of your printer.

Obviously, if one doesn't start off with some reference white, all the fussing afterwards will not likely give optimum color unless, like me, you are a genius! Well actually I really do have a pretty good color sense but still I have to use a white standard.

The idea of a standardization of light for photography is great in principal and when one does a formal portrait/ commercial one can often totally control the incident light on the scene. However, even then, a blue wall or a green dress can and will alter the appearance of all other colors as it presents a new source of light! After spending a fortune on studio lights that have constant temperature over a large range of Watt/second output, this is very disappointing news. However, like pits in cherries it's part of life.

So, no matter what or how you shoot, you and I, we all need WB tools! I have been using WhiBal cards (I carry one in my wallet) while others might use an Expodisc . The latter has it's advocates, but I have never used one! Sometimes I forget and have to make up for this. I hunt for some "close to neutral" target to use as a reference, like the white sclera of the eyes (can be off color), concrete (usually good), pupil of eye (not the colored iris). However, one should have that reference shot always!

Now we have a new option to consider. This is the Color Parrot . This tool has a center area of more dense whiteness and has been reviewed by our own super-critical Doug Kerr, here and discussed in OPF here and here and in Prophotohome.com here

These are not at all expensive and whatever you choose, you must take it with you. The Color Parrot has a device for stowing it with the camera. Of course the WhiBal goes in one's wallet! Either way, like the MasterCard™, (oops! Amex!) which ever you choose, "Don't leave home without it!"

Asher

*Imagine the monitor as a window to a room but it has an odd colored glass! That is what has to be accounted for in a monitor profile! Your files are never changed, just the colors are remapped so that when looking through your imperfect color monitor, the perception is that the color is perfect and there is not color shift.

 

[Doug Kerr]

Hi, Asher,

Now we have a new option to consider. This is the Color Parrot . This tool has a center area of more dense whiteness . .

And we still have no idea what that is supposed to do, although fortuitously (a) it exhibits better chromatic neutrality ("more white whiteness", perhaps) than the rest of the diffuser face and (b) the light through it predominates in the light that arrives on the sensor for measurement (over most of the frame, as a matter of fact).

Either way, like the MasterCard™, which ever you choose, "Don't leave home without it!"

In fact, that was the slogan of American Express Card.

 

[Asher Kelman]

Hi, Asher,
And we still have no idea what that is supposed to do, although fortuitously (a) it exhibits better chromatic neutrality ("more white whiteness", perhaps) than the rest of the diffuser face and (b) the light through it predominates in the light that arrives on the sensor for measurement (over most of the frame, as a matter of fact).

.......when it is used on the camera! Is that what you are saying?

You must have worked fro 20 hours on the Color Parrot!

Asher

 

[Doug Kerr]

Hi, Asher,

.......when it is used on the camera! Is that what you are saying?

Yes. The central spot is the primary contributor to the chromatic response of the Color Parrot (as used on a camera, for measuring incident light), which is good, since that the only reason the chromatic response of the Color Parrot is as near "neutral" as it is.

You must have worked fro 20 hours on the Color Parrot!

Rather more than that, I'm sure! Carla is certainly hoping that "we are about done with it".

The Color Parrot is a very workable color balance measurement tool. I have not yet learned (from my testing or otherwise) how or why it might be superior to other available white balance diffusers (such as for example the ExpoDisc) other than it's slightly lower price.

During the early days of its introduction, Drew had intimated that there was something about the Color Parrot's construction or behavior that better fitted it for making color balance measurements from the camera position (aimed at the subject) than other white balance diffusers. That notion is less prominent in his current descriptive material.

His most recentt position with regard to technique of use is that it can be (a) used at the subject (aimed perhaps at the camera, or perhaps at "the principal light source"), or (b) at the camera position for the shot (aimed at the subject), whichever gives the best results in the particular situation.

As you may know from my writings, it is my opinion that technique (b) can, on principle, only be expected to produce an "accurate" result if the incident light illumination upon the camera position for the shot is equivalent in chromaticity to the incident light illumination upon the subject. We are fortunate that this condition is evidently approximately met in many real-life photographic settings. It is another case of "better lucky than good".

There are of course many who disagree with this outlook, but I have not yet seen advanced any credible concept by which technique (b) could be in general expected to produce an accurate result other than under stipulation to the fortuitous conditions I described above.

Drew's assertions about the beneficial "greater targeted-ness" of the Color Parrot is not accompanied by any such conceptual explanation (nor even a description of what technical property that refers to, although we might conjecture it refers to the acceptance directivity pattern of the Color Parrot).

Of course in actual practice this matter is complicated by the possibility that we actually strive for a color correction that is different from what I have characterized as the "theoretical ideal". That color correction is one that would result in the chromaticity of an object in the (corrected) image being the same as its reflective chromaticity (the chromaticity we think of as characterizing the "color" of a familiar object). In other words, that is the color correction that would result in a "neutral white" object having, in delivered the SRGB image, R=G=B.

It is this color correction that is executed when, in processing a raw file, we apply the "color balance eyedropper" to the image of an object we know to be reflectively neutral (saying, "make this white").

Of course, we may wish to take account of the "actual viewer" of the scene having a chromatic adaptation not in accordance with the general incident light at the battle zone but rather in accordance with the reflected light from the area surrounding the principal subject. (Noel Carboni, in discussing this, speaks of a man reading a book with white pages, illuminated by daylight through a window, seated in a corner immediately surrounded by red walls illuminated by tungsten lamps, to which I add, "but in a room where the remaining walls are off-white".)

Now, the question is, "is the color correction we desire automatically attained by making a measurement of the light reflected from those red walls?" (based on the notion that this is what determined the visual chromatic accommodation of a "live viewer" of the actual scene). If our answer is "yes", then we might in fact want to make a measurement of chromaticity from the camera position for the shot, aimed at the "man with book", using a measurement instrument (i.e., our camera equipped with a particular diffuser) having an "acceptance pattern" narrower than that of the classical white balance diffuser.

Now, if we are satisfied by our intellectual achievement there, we might want to contemplate this: If there were saloons on Mars, would they have to close on Christmas?

At the end of the day, there is little question but that (and Drew does not disagree with this) the best way to accurately determine the effective chromaticity of the incident light upon the subject (using our camera as the actual "guts" of the instrument, so it can automatically use the findings to control white balance correction) is to have the camera regard a reflectively neutral target ("gray card") at the subject (assuming of course that the target was "very close" to true neutrality).

A diffuser (such as an ExpoDisc or Color Parrot) on the camera, used at the subject, is probably the next best from an "accuracy" standpoint (depending on the chromatic accuracy of the device, of course, just as with the gray card). The use of a diffuser-equipped camera at the shooting position is an often convenient but useful compromise.

I was recently asked by a colleague, "For what type of subject would you feel that measurement at the camera, rather than at the subject, was best?"

"A Kodiak bear", was my reply.

Best regards,

Doug

 

[Bart_van_der_Wolf]

.......when it is used on the camera! Is that what you are saying?

I have not physically handled it like Doug has, but I looks like the central area of the device offers a more diffuse contribution than the surrounding area. This is also shown when one examines the actual 'image' it produces (which is to be evaluated by the camera's CWB algorithm). This also causes CWB differences due to the actual size of the front lens element.

I carry a square piece of opaline glass in my camera bag. Opaline glass is a nice diffuser of transmitted light, and the only non-neutral color it introduces is that of 3mm sheet glass. I've measured it with a spectrophotometer to introduce a color-temperature shift of -455K for a single layer, and -620K for a double layer. This indicates that the diffusion alone added more of a shift (-455K), due to a different weighting of sky and ground contribution, than a single additional layer of glass itself (-165K).

It could be interesting to compare the resulting CWB that this cheap piece of non-yellowing glass produces, with that of the much more expensive 'color parrot', or the even more expensive Expodisk.

Bart

 

[Doug Kerr]

Hi, Bart,

I have not physically handled it like Doug has, but I looks like the central area of the device offers a more diffuse contribution than the surrounding area.

It seems as if the "base" layer of the Color Parrot does not come at all close to being an "ideal" (Lambertian) transmissive diffuser.

One criterion is doesn't meet is that the distribution of observed luminance ("downstream") with angle of observation should not vary with the angle of incidence of the "upstream" illumination. (It should always be symmetrical, even if not uniform.)

Said in a more homey way, for the Color Parrot base layer, much of the light that strikes its face "goes straight through" rather than being scattered (which is the mechanism of diffusion).

It is this aspect of the transmission of a "filter" that is measured in the "regular transmission" mode of a transmissive densitometer/colorimeter.

However, it is likely that the "overlay" layer of the Color Parrot, in addition to making a worthwhile (but incomplete) correction of the substantial non-neutrality of the base layer, also augments the Lambertian properties of the device, making it more of a "real diffuser" with respect to light striking that spot.

Of course. it is difficult to model in one's head the effects of all these properties on the various aspects of the operation of the device in actual photographic use.

In this overall connection, you might be interested in this recent supplementary report on Color Parrot behavior:

http://www.prophotohome.com/forum/w...ot/76180-color-parrot-use-smaller-lenses.html

Best regards,

Doug

 

[Drew Strickland]

Just thought I would chime in here to provide a little more information on the Color Parrot.

Doug has been very diligent in his analysis of the Color Parrot. I really appreciate all of his work. However, he is primarily interested in the product as an intellectual endeavor and is still not convinced the center-weightedness provides a notable advantage. I hope to get some more info up about this soon for Doug.

From a practical standpoint here are the primary advantages of this tool over most of the competition.

1. More Convenient (usually no need to adjust color in post)
2. Center Weighted
3. Accurate (one of the few truly spectrally neutral white balance tools)
4. Well Constructed
5. More Cost efficient (one size fits all lenses up to 82mm/ 5 different filters in one (the pro version includes 3 different warm settings and one cool setting))
6. Works better in Low Light

We have already sold around 200 of these tools. We have received no complaints about the core product, yet. A few folks had some issues with shipping which we are remedying. We have also gotten back the results from a survey from working pros who are using this tool in the field. Here is some of the feedback we are getting from working pros.

“My husband and I are very satisfied with the Color Parrot. It is the easiest white balance tool we have found. We started using it the day we received it, and we don’t take our cameras anywhere without it; which means it is always with us.”
“We are so happy with this tool. It is one of the best purchases I have ever made for my husband and myself. I’m always trying new things to try and make things easier, and this is one of the first things I have purchased that my husband really likes. Thank you for such a good item.”
Norine Piet



“The Color Parrot works very well. It saves lots of time and colors are very good. It performed very well in a number of gyms that had various light sources. Extremely pleased. This really cut down on my post processing. This is very simple to operate. I shoot with a Canon 5D.”
Kent Holder
Portrait and Indoor Sporting Event Photographer
Before the Color Parrot he set white balance in post-production



“The Color Parrot performed surprisingly well over the Expodisk. The Color Parrot gave me a good white balance.”
Tom McKean
Studio and Outdoors/Nature Photographer
Owns Expodisk as well


“Great Experience”
Nicholas Ong
Fashion/Studio Photographer


Compared to the WhiBal “it is great. Getting the color is easier, as it is real time, versus having to go fix the white balance in post-production.”
David Schow
Social Events Photographer
Owns WhiBal as well


“Easy to use.”
Jerry Morris
Avaiation Photographer


I “was blown away when I saw the results and ease of use.”
Donald Keith
Portrait/Wedding Photographer,


The Color Parrot is “great” compared to the grey card. “I like the parrot.”
Gene Carl
Sporting Events Photographer
Also uses grey card


I “am really enjoying using it, that’s for sure.”
“Recent scenario: In a wedding chapel. Party is near candles. There is flourescent lighting in the chapel. Recessed spot lighting above the bride and groom and massive amounts of window light peering through stained glass windows. The color of the light? Pft. This is where the CP came into action. My subjects skin tones looked nice and so did the surrounding areas. Time in Photoshop = NONE!
Talk about saving on workflow.”
Kix Pix Photography


There is a lot more information about the product(s) here.

 

[Colleen Vermillion]

firmware delivering a RAW file in AdobeRGB or other wide space RGB color space

This is probably just nit picking, but it is a common misconception so I'll pick it anyhow. RAW files are not RGB, so they can't be in any RGB space. That is one of the primary reasons for shooting RAW - because the data hasn't been rendered into an image yet, you can change your rendering intent without losing too much information. The profile that your raw converter uses to render RAW files from a particular camera model into an RGB image is a critical part of the color processing chain and the underlying cause of many of the debates about which converter is better than the other. Raw images are more like latent images than negatives, except the latent image isn't destroyed when we "develop" them.

-Colleen

 

[Doug Kerr]

Hi, Drew,

Just thought I would chime in here to provide a little more information on the Color Parrot.

Thanks for coming by, Drew.

[From here on, for conciseness, I will use color to indicate who is writing. I will be OPF color, Drew will be blue.]

Doug has been very diligent in his analysis of the Color Parrot. I really appreciate all of his work. However, he is primarily interested in the product as an intellectual endeavor[AArgh! the mummy's curse!] and is still not convinced the center-weightedness provides a notable advantage.[I don't even know what it is.] I hope to get some more info up about this soon for Doug.

I'm anxious to learn about that (as an intellectual endeavor, of course).

From a practical standpoint here are the primary advantages of this tool over most of the competition.

1. More Convenient (usually no need to adjust color in post)

More convenient than what? It sounds as if you are speaking of the greater convenience of using in-camera color correction per se, not of a particular tool for use in that.

2. Center Weighted

Are we speaking here of a nonuniform angular directivity pattern, or of some greater "sensitivity" of the center of the disk, or what?

Regarding the directivity pattern, essentially all diffusers are "center weighted" in that they do not have a uniform pattern (many approximate the "cosine" pattern, which is of course not uniform, but "center weighted"). The Color Parrot's pattern is indeed, however, narrower than that of most other diffusers.

3. Accurate (one of the few truly spectrally neutral white balance tools)

Would you care to identify those that are not "truly spectrally neutral"? ExpoDisc diffuser? WhiBal gray card? Picture Perfect gray card? Phoxle SpectraSnap? Two layers of unbleached coffee filters between two Canon UV filters?

What degree of spectral uniformity, or what closeness to chromatic neutrality, do you claim that the Color Parrot achieves?

4. Well Constructed

It's very nice.

5. More Cost efficient (one size fits all lenses up to 82mm/ 5 different filters in one (the pro version includes 3 different warm settings and one cool setting))

More cost efficient than what? My ExpoDisk works on all sizes up to 77mm (I didn't buy a larger one because I didn't need it.)

6. Works better in Low Light

Better than what? How does it do that?

[Testimonials:]

Compared to the WhiBal “it is great. Getting the color is easier, as it is real time, versus having to go fix the white balance in post-production.”
David Schow
Social Events Photographer
Owns WhiBal as well

Curious comment. I use my WhiBal to control in-camera color correction.

Best regards,

Doug

 

[Drew Strickland]

Hi Doug,

You know I love you and Carla. Or were you just kidding with the St. Valentine's day thing. But, sometimes people just want/ need the overview.

Here's a little more conciseness.


1. More Convenient (usually no need to adjust color in post)

Just what it says. Usually no need to adjust color in post. Many of the tools you mention require adjusting in post. This includes the WhiBal. It can be used to set a custom in-camera wb on the fly, but is less suited for this purpose. It primarily functions as a gray card replacement. As David said, the Color Parrot is easier.

If you wish to use the Color Parrot in a similar fashion as a gray card you would simply walk up to the subject and take the reference image from that position. According to your own test results, the results are very accurate in this condition. Also, your study, while very interesting, has a fatal flaw. It assumes the WhiBal is truly spectrally neutral. It is not. Of course, you do mention this in the report. I do not know how far off the WhiBal is, or in what direction, because I do not have yours for measurement. I will probably pick one or two up to get an idea of the variance between units.

2. Center Weighted

I've already expressed my theory to you regarding this along with illustrations. When I get a chance I will try and get a more convincing illustration up for you, as I said.

3. Accurate

In the white balance tool teardown, the ExpoDisc is not neutral, the expocap is not neutral. The gray card illustrated, is not neutral. The other diffuser you mention is less neutral than the color parrot.

4. Well Constructed

Our per unit cost is likely much higher than any of these other products you mention due to the quality of the construction.

5. More Cost Efficient

Actually this is understated. I know of no other diffuser that allows you to attain 3 different shades of warm, in addition to a neutral setting. Just to get 1 warm setting and 1 neutral setting from expodisc products it would cost $124 for a Neutral ExpoDisc 82mm + $124 for a Portrait ExpoDisc 82mm. You can get 3 warm settings, a neutral setting, and a cool setting for $99 with the Color Parrot Pro.

6. Works Better in Low Light

Another under marketed point. The ExpoDisc is hampered by its legacy of being an exposure tool. It only allows 18% light transmission. The Color Parrot allows a much higher transmission of light. Give it a try. Hold your 20D up to a lower lit scene and look at the suggested exposure numbers with both products. Being able to get a better white balance in low light is a big deal, if you are in low light. Just as some people may not care much about high iso, others can't live without it.

 

[Bart_van_der_Wolf]

1. More Convenient (usually no need to adjust color in post)

Just what it says. Usually no need to adjust color in post.

You are describing a generic benefit of a Custom White Balance. You don't explain how the Color Parot could even differ from other tools in that respect. You also seem to suggest that the resulting CWB is more accurate when your tool is used as compared to other tools used to point in the direction of the subject, yet you agree that using it from the subject position pointing to the camera provides better results.

Also, your study, while very interesting, has a fatal flaw. It assumes the WhiBal is truly spectrally neutral. It is not.

Actually, it is pretty good. Measured with a spectrophotometer, it has an almost neutral response between 430 - 730 nm (there's a very gradual change over that full range without peaks). Below 410 nm it rapidly loses its reflectance (it absorbs near-UV). I've measured 2 versions (Pocket and Studio) of the WhiBal and they are (besides a slightly different, 71.1 vs 72.7%, overall reflectance) very similar along all relevant wavelengths.

Of course, you do mention this in the report. I do not know how far off the WhiBal is, or in what direction, because I do not have yours for measurement. I will probably pick one or two up to get an idea of the variance between units.

If you want a better tool/subject to substantiate claims of spectral neutrality, I suggest the BabelColor White Target. It exhibits a very good 'Color Consistency' and it has a very neutral, diffuse, and extremely high reflectance.

6. Works Better in Low Light

Another under marketed point. The ExpoDisc is hampered by its legacy of being an exposure tool. It only allows 18% light transmission. The Color Parrot allows a much higher transmission of light.

Which is totally unimportant for acquiring a CWB. The CWB target 'image' is diffused, and the camera's metering will produce an identical auto-exposure level for those reference images, as it should if one wants to determine a CWB as suggested by the camera manufacturer.

Bart

 

[Doug Kerr]

Hi, Drew,

Hi Doug,

You know I love you and Carla. . . .

I've never questioned that, Drew.

Or were you just kidding with the St. Valentine's day thing.

I'm not sure what you are referring to, Drew. Maybe I attached something I didn't mean to. (Wow, I hope it wasn't the picture of Carla with . . . Well, never mind.)

But, sometimes people just want/ need the overview.

Here's a little more conciseness.

1. More Convenient (usually no need to adjust color in post)

just what it says. Usually no need to adjust color in post. Many of the tools you mention require adjusting in post. This includes the WhiBal. It can be used to set a custom in-camera wb on the fly, but is less suited for this purpose.

Well, which is it: Does the WhiBal diffuser "require adjusting in post", or "can [it] be used to set a custom in-camera wb on the fly"?

Why is it "less suited for this purpose"? (And do you mean, "less suited to that technique than to using it as a reference in post", or "less suited for custom WB setting than than the Color Parrot")?

It primarily functions as a gray card replacement

It is a gray card.

If you wish to use the Color Parrot in a similar fashion as a gray card you would simply walk up to the subject and take the reference image from that position.

Quite so. That may be a bit easier than asking the model to hold the gray card and then walking up and taking the reference shot. That's why a diffuser can be a useful product.

Also, your study, while very interesting, has a fatal flaw. It assumes the WhiBal is truly spectrally neutral. It is not. Of course, you do mention this in the report. I do not know how far off the WhiBal is, or in what direction, because I do not have yours for measurement. I will probably pick one or two up to get an idea of the variance between units.

That would be good. Indeed, in my report, I stipulated to my WhiBal's chromatic neutrality being within the range of the manufacturer's test requirements (±0.5 unit of a* and b* for an L* of approximately 70). I made no presumption as to its spectral uniformity.

Subsequent testing suggests that the WhiBal I used in my report has a chromatic neutrality (in the CIE L*a*b* model) of within ±0.025 units on the a* and b* axes. That corresponds to an uncertainty of less than ±0.0004 unit on the du' and dv' axes of my charts.

2. Center Weighted

I've already expressed my theory to you regarding this along with illustrations. When I get a chance I will try and get a more convincing illustration up for you, as I said.

That will be good. I'm sorry but I wasn't really able to follow the illustrations you earlier gave.

As a first step, could you please confirm exactly what you mean by "center weighted". Are you speaking of the relative transmission of the diffuser versus angle of arrival of the incident light (that seems a reasonable assumption), and by "center weighted" do you mean that the response is greater for light arriving on axis than from off axis (again, a reasonable use of "center weighted" if it is indeed the response pattern that is being spoken of), and then, is the distinction of the Color Parrot from other diffusers that it has a "narrower pattern"?

That would help me right now. As to what aspect of the functional performance of the Color Parrot that improves, and how, I'll be glad to wait for your further discussion of that.

3. Accurate

In the white balance tool teardown, the ExpoDisc is not neutral, the expocap is not neutral. The gray card illustrated, is not neutral. . . .

Nothing is neutral. Our concern is with "how close to neutral".

. . . The other diffuser you mention is less neutral than the color parrot.

Which other diffuser? Or do you mean, "the other diffusers mention are less [nearly] neutral than the Color Parrot"?

I just don't believe that. I think you are being deluded by your belief that one can make a meaningful test of the spectral uniformity, or the of the transmissive chromatic neutrality, of a transmissive diffuser with a reflective spectrophotometer, colorimeter, or densitometer.

Although tests with a transmissive densitometer/colorimeter/spectrophotometer are not fully illustrative of the performance of a diffuser in the setting in which we use them, they are at least a good basic indication of the property in which we were interested. But I don't have one here (and no budget for commissioning a study by a competent laboratory).

[/quote]4. Well Constructed

Our per unit cost is likely much higher than any of these other products you mention due to the quality of the construction.[/quote]

The Color Parrot is indeed nicely made. Even the center overlay in my Color Parrot is almost circular. But I can't imagine how you judge it better made than the "other products mention". (I know you don't have a sample of my coffee filter diffuser there to examine. It's lovely.)

6. Works Better in Low Light

Another under marketed point. The ExpoDisc is hampered by its legacy of being an exposure tool. It only allows 18% light transmission. The Color Parrot allows a much higher transmission of light

This matters not at all. I assume one would use the camera's exposure metering system to attain whatever exposure result one thinks is good for a CWB reference frame. With the higher density of the ExpoDisc, this will indeed likely result in a longer shutter speed (compared to a similarly-metered exposure with the ColorParrot). I can usually spare the extra 1/50 sec or so.

In any event, once the desired photometric exposure is on the sensor (unless we're talking about such a long exposure that we get a differential reciprocity failure by color channel), it is all the same.

Perhaps you are confusing this matter with the argument that, when a gray card is to be included in an actual pilot shot of the subject (for determining color correction in raw postprocessing), it is advantageous for it to have a "fairly high" reflectance. (That point has been made by Michael Tapes, father of the WhiBal gray card. Good morning Michael. Are you finding this all entertaining?)

Thanks again for clarifying your position on these matters.

Finally, since you mentioned Carla and Valentine's Day, I thought I would close with one of my favorite examples of mediocre results from Canon's Average White Balance (here on a 300D, on Valentine's day a couple of years ago):

Best regards,

Doug

 

[Andrew Rodney]

Doug, here's one more product to investigate. The English translation leaves something to be desired.

http://www.cbllens.com/

Oh, I should mention I scored one at PMA but haven't done a thing with it (yet).

 

[Andrew Rodney]

Obviously, if one doesn't start off with some reference white, all the fussing afterwards will not likely give optimum color unless, like me, you are a genius!

My first question and comment (and I'm shooting Raw), what about images that don't have a white reference and what says shooting one in another frame and using it would produce something we want to express photographically? Case in point, images #4, #30 and #31 here:

http://digitaldog.net/ARsAmazonPicks/

#4 especially is an image that when provided to others as a Raw ends up with a huge difference in color rendering, its all very subjective. Putting a white card in the previous frame and balancing isn't at all useful in this kind of photography.

The idea of a standardization of light for photography is great in principal and when one does a formal portrait/ commercial one can often totally control the incident light on the scene.

I'd agree. How often do we shoot with standardized lighting? Well often I did (still have a few Balcar's left). But more often its not like the examples above. And with Raw, I don't have any difficultly getting the color balance rendering I desire with a WB tool or just moving tint and temp sliders.

So, no matter what or how you shoot, you and I, we all need WB tools!

I'm not so sure. But then some teach and would suggest you should always be using an external, incident light meter when shooting or one that combines Spot. Yet how many do this? Is their work necessarily suffering? (Yes if under exposed <g>).

I wouldn't at all be opposed to having a tiny built in Spectrophotometer that read the ambient light and uploaded some EXIF data somehow. I might still totally ignore it.

 

[Drew Strickland]

Glad to see we've moved the party over here. I was missing you guys.

We'll try not to tear up the place, Asher.

I am absolutely slammed today, but I will try to get back with you in more detail. I guess I need to make my assertions more clear. Big surprise.

Just one quick note that is easy to illustrate.

It matters very much that the diffuser transmits more light, when you are in a low light setting. The "old directions" go into great detail about how the expodisc is dark and you may need to manually adjust your exposure up a few stops to get a useable reference frame. I'll try to get a shot of that up for the non-believers.

This is a phenomenon that you will only notice at lower light. If you use Canon, you may have a harder time discerning this difference because of the manner in which Canon sets CWB. If you are not getting enough light through the lens Nikon will reject the cwb by saying "no good" in the display.

There are many instances where I can easily take a valid wb setting with a Nikon where the expodisc fails. Canon does not reject anything for being out of bounds. It will dutifully obey and use your last reference frame, even if it is way too dark to be useful. Nikon does some assessment on the reference image in real time and provides feedback to the photographer. Quite frankly, Nikon is a superior camera system for frequent cwb use. This is just one of the reasons.

 

[Doug Kerr]

Hi, Drew,

This is a phenomenon that you will only notice at lower light. If you use Canon, you may have a harder time discerning this difference because of the manner in which Canon sets CWB. If you are not getting enough light through the lens Nikon will reject the cwb by saying "no good" in the display.

There are many instances where I can easily take a valid wb setting with a Nikon where the expodisc fails. Canon does not reject anything for being out of bounds. It will dutifully obey and use your last reference frame, even if it is way too dark to be useful.

Indeed. We need to pay attention to the exposure for the refernce frame (and maybe later for the real shot as well). The Canon cameras have a useful feature, called "exposure metering", that can help in this regard. I generally use it when taking CWB reference frames.

Best regards,

Doug

 

[Doug Kerr]

Hi, Andrew,

I wouldn't at all be opposed to having a tiny built in Spectrophotometer that read the ambient light and uploaded some EXIF data somehow. I might still totally ignore it.

Then people would still argue about where to place the instrument when we used it and in which direction to point it!

Of course, that's what we do when we put a real diffuser on our Canon EOS camera.

We can read its findings by using that frame as a CWB reference frame, setting WB to CWB, shooting anything we want (with a raw output), opening the CR2 file in CSPS/ACR, and reading the "as shot" color correction (which will be in terms of CCT and 3000 times the Planckian offset - I think it was you that clued me to the latter).

This is interpreted from the color correction data, expressed as "channel gains", that is in the Exif metadata (the makernote portion) in both the JPG and CR2 files. It's just that PSCS/ACR won't interpret it except in the case of a CR2 file.

Its admittedly not nearly as handy as what you suggest, but . . .

Best regards,

Doug

 

[Bart_van_der_Wolf]

Canon does not reject anything for being out of bounds. It will dutifully obey and use your last reference frame, even if it is way too dark to be useful.

That's correct, but then the reference frame was not taken as instructed by Canon, and they warn that such behavior can result in a different CWB.

To quote from the 1Ds Mark III manual: "Shoot the white object so that a standard exposure (gray) is obtained. If it is underexposed or overexposed, a correct white balance setting might not be obtained."

Bart

 

[Doug Kerr]

Hi, Drew,

With regard to the difference in performance of the Color Parrot and (say) the ExpoDisc in low light:

The photometric measure we are intersted in here isn't (necessarily) precisely the same as the "percentage of light transmitted", but that simplification will serve our purposes for the moment.

For my ExpoDisc, the manufacturer's testing indicates that this value is 18.2%.

Yopu have apparently intentionally designed the Color Parrot to have a higher value of that property, for the reason you have just discussed.

What value should we expect a typical Color Parrot to exhibit for this property?

Thanks.

Best regards,

Doug

 

[Asher Kelman]

My first question and comment (and I'm shooting Raw), what about images that don't have a white reference and what says shooting one in another frame and using it would produce something we want to express photographically? Case in point, images #4, #30 and #31 here:

http://digitaldog.net/ARsAmazonPicks/

Andrew,

I loved the visit to your website. First I thought you had swiped them from Amazon.com LOL! Then I realized they are far to good, although I don't want your head to grow too much! I am so impressed and will bookmark and study the images. Of course, with these images, I personally would use my memory which is still pretty damn good! One does not want to neutralize special light. However when one is taking pictures under a tree, one may hate the green light altering the color of the brides dress.

Again, I realy am drawn to the pictures. Kudos and no disagreement on your take on need to correct.

Now back to the issue of Color Parrots with thicker diffusion circles inside them and why they might be better! I personally am open to any simple device that helps us in a bind. In a reception where light is so different with mixed lighting changing all over the place, I take WhiBal readings in every part of the space and correct pictures in batches. With a transmissive device, I'd have to keep taking new refernce shots and that would break the workflow. However, for many events it could be fine.

Asher

 

[Asher Kelman]

Glad to see we've moved the party over here. I was missing you guys.

We'll try not to tear up the place, Asher.

Drew,

You are more than welcome! We love to examine the technical basis for differences in performance. This discussion is delightful and will make someone a good thesis! They better aknowledge the brains here!

Apart from the technical examination and challenging practical questions, we still, in the end, need tools that can be relied on. As professionals and serious enthusiasts, we all have to know where they excel and where they are not so worthwhile and as Andrew perhaps implies, where they should not be used at all.

Asher

 

[Doug Kerr]

This is regarding the assertion that it is more advantageous to use a white balance diffuser with a higher diffuse transmission, since it produces a "hotter" reference frame that would be "more acceptable" to a Canon EOS camera for CWB setting (particularly under "low light" conditions.

Of course, it will not produce a hotter reference frame if we utilize any form of automatic exposure control.

Here are two reference frames taken in my closet (!), with an incident light on the face of the diffuser measured at 1.0 lux (which I'm sure, Drew, you would consider to be "low light"). One frame was taken with my ExpoDisc diffuser, the other with my Color Parrot diffuser,

Both were taken with the same camera and lens settings: ISO 800, f=24mm, focus at infinity, f/4, Av metering mode, partial metering, EC=+1.0.

Color balance for both reference shots (which only affects the visible appearance of the shot, not its implication as a CWB reference) was K mode at 6500K.

Here is the frame taken with the ExpoDisc:

Here is the frame taken with the Color Parrot:

Do we really believe that the latter would be "better accepted" by the Canon CWB system than the former?

Best regards,

Doug

 

[Doug Kerr]

Color parrot passes drop test with flying (whatever)

We have inadvertently conducted a drop test of our Color Parrot, and we are pleased to report that its robust construction allowed it to perform magnificently.

We were making some tests on the device (you've all hear of that), and,. since I needed to make repetitive shots with varying camera settings, I taped the Color Parrot to the front of the lens with a strip of masking tape (the 3M blue kind, the "delicate surfaces" subspecies). It wasn't a really secure affixation - just intended to keep it place for about a two minutes while I made 24 test shots.

When I finished the series, I rushed off to the darkroom to take a quick look-see. Then the phone rang, and well, you know how it is for we old guys. I forgot all about my test setup.

Later, Carla was startled by a loud noise in the kitchen, which she soon discovered was from the Color Parrot going to the floor as the tape gave way. She carefully recovered it.

It had fallen about 50 inches to a ceramic tile floor.

I'm pleased to note that it shows no visible damage - nothing cracked, there were no dents or dings, nothing.

Its robust construction served it well in this inadvertence.

Still, Drew, do you recommend I send it back to the factory for recalibration?

Best regards,

Doug

 

[Doug Kerr]

Hi, Drew,

With regard to the difference in performance of the Color Parrot and (say) the ExpoDisc in low light:

The photometric measure we are intersted in here isn't (necessarily) precisely the same as the "percentage of light transmitted", but that simplification will serve our purposes for the moment.

For my ExpoDisc, the manufacturer's testing indicates that this value is 18.2%.

Yopu have apparently intentionally designed the Color Parrot to have a higher value of that property, for the reason you have just discussed.

What value should we expect a typical Color Parrot to exhibit for this property?

Not being a manufacturer of filters or diffusers or anything like that, we have no densitometers here, but, while,waiting for the official answer, we sharpened up the two nails and oiled up the potato, and made some tests of the transmission of stuff around the house. (Just a metaphor, folks - we really do it with our 20D's metering system.)

Just as a matter of interest, my Color Parrot seems to have an in situ diffuse transmission about 0.5-0.75 stops higher than my ExpoDisc. (I have to give a range, since this property depends on the distribution of the incident test light. Those are the results I got from two different "convenient' distributions. Your mileage may vary.)

In terms of percentage, my ExpoDisc is said by its manufacturer to have a diffuse transmission of 18.2% (I have to guess what measurement conditions that is taken under, but I won't bore you with talk of integrating spheres and all that).

But just starting with that number, then we might estimate that my Color Parrot has a diffuse transmission in the range of between 27% and 31%. Does that jibe with your actual measurements of typical Color Parrot production samples, Drew?

Best regards,

Doug

 

[Drew Strickland]

Actually, it is pretty good. Measured with a spectrophotometer, it has an almost neutral response between 430 - 730 nm (there's a very gradual change over that full range without peaks). Below 410 nm it rapidly loses its reflectance (it absorbs near-UV). I've measured 2 versions (Pocket and Studio) of the WhiBal and they are (besides a slightly different, 71.1 vs 72.7%, overall reflectance) very similar along all relevant wavelengths.

Hi Bart,

Would you mind uploading that spectral curve and density readout. I would love to see it.

 

[Drew Strickland]

But just starting with that number, then we might estimate that my Color Parrot has a diffuse transmission in the range of between 27% and 31%. Does that jibe with your actual measurements of typical Color Parrot production samples, Drew?

Hi Doug,

Here are the density measurements from the expodisc and the color parrot (v2.0).

ExpoDisc Density Readings in CMYK
Shows around the 18% stated in the cyan and the black (subtract to get transmission numbers)

Color Parrot Density Readings for center target area in CMYK
Shows around 60% light transmission

The Color Parrot version 2.0 transmits around 3 times the amount of light that the ExpoDisc does.

Version 1.0 (the one you have) shows light transmission of about 40% through the center. This would be about twice the light transmission of the ExpoDisc.

 

[Drew Strickland]

Still, Drew, do you recommend I send it back to the factory for recalibration?

Sure. Let me get with my repair technician to get you a quote.

 

[Drew Strickland]

This discussion is delightful and will make someone a good thesis!

Hi Asher,

Thanks.

I think Doug's already got that covered.

Andrew, you must have more than enough fodder for another article by now. Do you need a Color Parrot to add to the review?

 

[Drew Strickland]

If you want a better tool/subject to substantiate claims of spectral neutrality, I suggest the BabelColor White Target. It exhibits a very good 'Color Consistency' and it has a very neutral, diffuse, and extremely high reflectance.

Hi Bart,

Would you also mind uploading this curve. Perhaps I'll pick one up. Maybe I'll pick one up for Doug too since he has done such a good job with the WhiBal. Please don't hurt me, Carla.

My wife is about way over posing for me for this stuff.

 

[Drew Strickland]

I take WhiBal readings in every part of the space and correct pictures in batches. With a transmissive device, I'd have to keep taking new refernce shots and that would break the workflow. However, for many events it could be fine.

Hi Asher,

You can do the exact same thing with the Color Parrot, if you like to do it in post that way. Just walk to each corner of the room, or however you deem appropriate given the lighting situation, and take a quick reference shot to be used in post. The advantage is you have a much bigger reference area to click.

You can also use it exactly like the WhiBal, just hold it out and take a reflective reading from the surface, or have the Bride hold it for you while she talks to all of her guests and does the first dance.