Fun with a model! How bad can it look?

[Asher Kelman]

LOL
You assume that I took care of the whitebalance, while all I wanted to show in that picture was that a greenish wall would give some nice contrast to her orange shirt. I really couldn't care less if her eyes are green and her hair blue... Those are just remains of a sloppy 'blend if' layermasking that I did.

If I wanted to give her a life-like appearance, the result would be something like this...

But as said, that wasn't the purpose of the first picture... And the remarks I make about color-cast and WB are reactions to the people asking questions about them.

Joachim,

At last!!

I'm so glad you did this as I personally am troubled by the overall salmon color. Putting aside the value of white balance itself, look how the picture on the right is so much more dynamic and has more life and depth. What you have done is allowed her to now emerge from the background. At last!

Salmon color, BTW, might be fine for another b.g. when a gelled light illuminates her. I never ask for the truth except where truth is expected by the purpose of the shot, decency, ethics or the law.

Asher

 

[Joachim Bolte]

By the way, I'm mixing up the ink and the light thingy again... The histogram is showing te WB-cards to have a blueish hue, and the histogram is correcting that by adding yellow to the picture.

It's also clearly visible in the picture, the bottom part of the cards is way more blue-casted than the middle and top-part. Could be the way of lighting things, using multiple lightsources... And one could think about how much damage is done by the JPG compression already. If you compress the histogram in Lab then you can clearly see where the blue patches are. Well, you can see it by eye already, but for the non-believers...

Joey

This additive and subtractive stuff will kill me one day...

 

[Kevin Stecyk]

Asher asked me to provide a quick intro to LAB, since I seem to be referring to it a lot.

My knowledge of the LAB workspace comes from Dan Margulis. If you are at all interested in LAB, I urge you to buy his book Photoshop LAB Color: The Canyon Conundrum and Other Adventures in the Most Powerful Colorspace [Paperback]. Please note that I have embedded my Amazon affiliate link in referencing Dan's book.

LAB is just another colorspace, one that I find very useful for a number of reasons. One of those reasons is ~quickly~ analyzing an image. When I have an image, I quickly scan the image in LAB looking for things that are wrong. Then I can decide if I want to fix them (often in rgb colorspace) or just accept that they are wrong. Or, I can be "artistic" and make them more wrong. The key point being that I know where I am starting from.

In LAB:

- L - Lightness 0 dark and 128 bright
- A - Magenta/Green (128 all magenta, -127 all green)
- B - Yellow/Blue (128 all yellow, -127 all blue)

Magenta and Yellow are both warm colors. And they are both positive. Many things in nature are warm. In Dan Margulis's classes, he teaches his students the following:

NEUTRALS
A and B both zero. If not exactly zero, negative numbers are cold (green/blue) and positive numbers warm (magenta/yellow).

FACES
A and B both positive. With the exception of small children and light-skinned Caucasians, the A is rarely significantly higher than the B. Asians, Hispanics, and dark-skinned Caucasians generally have B significantly higher than A.

NATURAL GREENERY
A negative, B positive. The B is almost always farther away from zero than the A, usually falling within 1.5-2.5 times the absolute value of the A.

SKIES
B is negative. A is near zero, meaning neither magenta nor green. Typically the A is slightly negative, but occasionally it is slightly positive. The B value cannot be predicted, but the A generally falls between -5 and +2.

So when looking at some of the posted images in this thread, I quickly scanned them looking at the LAB values. If I saw a negative value on the model (skin, eyes, hair), I knew something was amiss. Hair can be tricky as women often dye their hair. And then, it can be anything. However, natural hair is warm--that is, positive in the A and B channels.

If you look at my readouts in prior posts, you'll note that my right hand side readout shows LAB. It always does, even if I am in rgb colorspace. I find looking at LAB values very handy to spot errors quickly.

From my limited exposure (pun intended) to photography, I have found whites and blacks difficult. Because they are neutral, it doesn't take much to throw them off color. That is, a white top or black dress can easily appear cold (hint of green, blue, or both). Even worse, some part of the clothing might be slightly warm and other parts slightly cold. You have to decide what is acceptable.

Another thing I have learned from my limited exposure is that eye whites are tricky. As you've seen from my posts, I love color, lots of color. I often make aggressive moves to make my stuff overly colorful. If the eye whites are off, they become way off with my aggressive moves. So I usually make my eye whites neutral before I begin my aggressive moves. My aggressive moves move in proportion the color is away from neutral. Stated differently, neutral colors are unaffected by my aggressive moves, colorful colors are made more colorful.

In reality, eye whites are not truly neutral. According to Gry Garness (see CS4 Digital Retouching For Fashion Beauty And Portrait Photography - no affiliate link) on pages 159:

Put the cursor on the clearest whites of the eyes read the INFO panel. Are the RGB numbers near equal? For an adult, the numbers should be approximately: Red as the highest number, with Green 2-3 points lower, and Blue 2-3 points lower than Green. Teenagers and children can have almost completely neutral whites.

So my making a person's eye whites neutral might be a bit of a cheat. However, like Will, I am sometimes shooting people in the early 20s.

I hope you found this information, including references, helpful.

 

[Joachim Bolte]

Yeah, she was somewhat of a chameleon beforehand... but hey, if that's the picture the photographer wanted to take in the first place, who are we to alter that. Taste is personal, and we cannot step into his mind to see what thoughts he had...

I know which one I like best, but that's just something I know...

 

[Doug Kerr]

Hi, Joey,

By the way, I'm mixing up the ink and the light thingy again... The histogram is showing te WB-cards to have a blueish hue, and the histogram is correcting that by adding yellow to the picture.

The histogram adds yellow?

I assume you really mean the result of using the gray eyedropper on the Curves tool (or such) to do white balance color correction.

Best regards,

Doug

 

[Joachim Bolte]

- L - Lightness 0 dark and 100 bright
- A - Magenta/Green (100 all magenta, -100 all green)
- B - Yellow/Blue (100 all yellow, -100 all blue)

I have the book also, It's a great piece of work and very very insightfull... I think it's a book that should be on the shelf of everybody being serious about image-processing.

A little correction though: In the book, Margulis is solely working with the diagrams set tot 'ink coverage' mode. Then we are talking about a* and b* scales reaching from -100 to 100. When you let the diagrams be in the default 'Light' mode, the values are fitted within a binary range, and will reach from -127 to +128. Those values also apply when you do a Lab readout in the RGB info-pallette.

Does not change the concept though.

 

[Doug Kerr]

Hi, Joey,

If you compress the histogram in Lab then you can clearly see where the blue patches are.

How do I do that?

Thanks.

Best regards,

Doug

 

[Joachim Bolte]

@Doug,

"... in the histogram we can see that, by using the gray dropper, PS is correcting that by altering the blue curve so that the B values will decrease within the mid-range of colors..."

better?

 

[Joachim Bolte]

How do I do that?

1. you are in RGB. I mostly convert the picture to 16 bit before converting to Lab, but we can do without that now... In the menu, select Image - Mode - Lab

2. Add a curves adjustment layer to your layerstack

3. open it up, and set the grid to 10% increments

4. in the a as well as the b channel pull in the corner points about 40%. The line should be still passing through the middle of the histogram. If you want to be really precise, use the input/output boxes to traget your values.

5. When done, your color will be emphasized. Your lines are still through the Lab midpoint, so grays will be grays. You will clearly see that there are big blue and red patches in the image, especially on the bottom part. The way the colors form 'blocks' makes me suspect JPG-compression has something to do with it. The circular form makes me suspect it's a matter of flash vs. sunlight.

 

[Kevin Stecyk]

1. you are in RGB. I mostly convert the picture to 16 bit before converting to Lab, but we can do without that now... In the menu, select Image - Mode - Lab

2. Add a curves adjustment layer to your layerstack

3. open it up, and set the grid to 10% increments

4. in the a as well as the b channel pull in the corner points about 40%. The line should be still passing through the middle of the histogram. If you want to be really precise, use the input/output boxes to traget your values.

5. When done, your color will be emphasized. Your lines are still through the Lab midpoint, so grays will be grays. You will clearly see that there are big blue and red patches in the image, especially on the bottom part. The way the colors form 'blocks' makes me suspect JPG-compression has something to do with it. The circular form makes me suspect it's a matter of flash vs. sunlight.

Given that ~extreme~ move, I doubt you'd find much that was completely neutral throughout, especially in an outdoor setting. Moreover, because the target is likely a fabric, one that flexes, I would be careful about reading too much into patterns. You might find that the circular shapes reflect (pun intended) the shape of flex as much as it reflects differences in light sources.

I just use the LAB readouts to get a quick impression.

 

[Joachim Bolte]

nothing is completely neutral, I agree, but I find it somewhat strange that there are clearly defined areas within the picture that have a (very slight) colorcast of two different colors... how would you correct for that?

Even more disturbing I find the fact that in between those color-ranges, there are patches that ARE virtually grey.

Those factors make color-correction using this image as a base a lucky guess, at best. Therefore I did my correction using nothing but the dropper and eyeballing. I think that if you want to do correction, you should let your model hold up a card on a scene-to-scene base. The card should be the same distance from the camera as your subject is. Putting a graycard on the floor, and shooting it from 1 yard with a ringflash, will not give you very reliable results.

 

[Doug Kerr]

Hi, Kevin,

In LAB:

- L - Lightness 0 dark and 100 bright
- A - Magenta/Green (100 all magenta, -100 all green)
- B - Yellow/Blue (100 all yellow, -100 all blue)

I think that is a little misleading.

For one thing, the coordinates a* and b* are not limited (by definition, or by the working of the defining equations) to a range of ±100. Even within the gamut of the sRGB color space, the value of b* can be outside that range (not by a gigantic amount, of course - I think that within the sRGB gamut, the largest absolute value is -112, for b* - but the limits are broader for the gamut of Adobe RGB).

But let's for a moment stipulate to ±100 as "essentially the limits". So, for example, b*=-100 is as blue as this color space can legitimately represent (subject to those adopted limits) for the current value of L* But to describe that as "all blue" doesn't mean much. Maybe "all the way blue" is what you mean.

It does not (even when it appears in connection with a*=0) imply "fully saturated blue" in any of several credible definitions of that, other than perhaps the one implied by "C" in the LCh(ab) color space).

I don't believe that L*=50, a*=0, and b*=-100 is even an actual color (what some would call a "visible color", but of course "color" only applies to visible radiation).

(In RGB, it would be RGB=-93, 126, 291.)

Moving to the a* axis, we note that the Adobe RGB color RGB=0,255,0 has a theoretical a* value of -129 (although that can't be represented in the familiar 8-bit form). And that color is not fully-saturated green in the general colorimetric sense.

Best regards,

Doug

 

[Doug Kerr]

Hi, Joey,

"... in the histogram we can see that, by using the gray dropper, PS is correcting that by altering the blue curve so that the B values will decrease within the mid-range of colors..."

better?

Much. Thank you.

Best regards,

Doug

 

[Kevin Stecyk]

Hi, Kevin,

I think that is a little misleading.

...

I agree with all your points raised. Also, I stand corrected on the +/- 100. As pointed out by Joachim, it should be -127 and +128.

You're correct, LAB is able to create imaginary colors. That however can be helpful when sharpening or doing color corrections and coming back into RGB afterward.

 

[Doug Kerr]

Hi, Kevin,

I agree with all your points raised. Also, I stand corrected on the +/- 100. As pointed out by Joachim, it should be -127 and +128.

You're correct, LAB is able to create imaginary colors. That however can be helpful when sharpening or doing color corrections and coming back into RGB afterward.

I think we are together on all this. It gets tricky to try and explain things in a way that can be easily grasped and yet still be "correct".

Your discussion was very valuable.

Best regards,

Doug

 

[Kevin Stecyk]

nothing is completely neutral, I agree, but I find it somewhat strange that there are clearly defined areas within the picture that have a (very slight) colorcast of two different colors... how would you correct for that?

Even more disturbing I find the fact that in between those color-ranges, there are patches that ARE virtually grey.

Those factors make color-correction using this image as a base a lucky guess, at best. Therefore I did my correction using nothing but the dropper and eyeballing. I think that if you want to do correction, you should let your model hold up a card on a scene-to-scene base. The card should be the same distance from the camera as your subject is. Putting a graycard on the floor, and shooting it from 1 yard with a ringflash, will not give you very reliable results.

I corrected the two color casts as shown below the ONE and TWO colordropper/colorpicker. I used blendifs to restrict where the curves acted. This correction applies to near the colordropper/colorpicker. If you move away from those locations, you again see colors.

And, I agree, you should use a card on a scene by scene, time by time basis. Depending upon the time of day and latitude, 15 minutes alone can make a substantial difference. And if there are any clouds of any sort, they too can interfere with your color balance. Put differently, when shooting outdoors where you can't control the environment, you are likely to encounter some variability.

The card should be at your subject.

Another product is Colorchecker Passport. Again, no affiliate link. For those wanting even tighter color control, they might want to use this product, as--I understand--you can calibrate your camera's colors against 24 patch set of colors. I ~believe~ there's software included with this product that matches the red in camera with the red in the color patch. So now, all your colors are correct.

Some believe that certain brands of cameras have color biases. This product will supposedly remove that bias to give a closer match when using Adobe PS.

Different raw converters convert differently. This product will eliminate/reduce camera or software bias to help get you closer to the actual colors.

 

[Joachim Bolte]

@Kevin,

It's a pity the values for your current cursur position are not given.... because if you take the sample point at the bottom of the image, you will still read a -4 on the b* scale. Still a blue cast.

So now you (think you) have this corrected, could you copy paste the picture of the young lady in there to see how your adjustment curves affect the picture? That should be nice and balanced now...

Joey

ps: quite confusing that you edited a PS screenshot of an RGB histogram in PS Lab...

 

[Deleted member 55]

This is the Color Checker card I used for this shot!

 

[Kevin Stecyk]

@Kevin,

It's a pity the values for your current cursur position are not given.... because if you take the sample point at the bottom of the image, you will still read a -4 on the b* scale. Still a blue cast.

So now you (think you) have this corrected, could you copy paste the picture of the young lady in there to see how your adjustment curves affect the picture? That should be nice and balanced now...

Joey

ps: quite confusing that you edited a PS screenshot of an RGB histogram in PS Lab...

Joachim, it took literally less than three minutes to do this color correction. It's something you can easily replicate. Just place a colorsample point in two regions. Adjust accordingly, using your blendifs to restrict the colors.

I doubt this change will fix/correct Will's photo. If you note in Will's photo, the eye white are quite yellow. This minor move/correction will not restore the near neutral color.

Will's photo has a strong warm color cast. As you see from the gray card, the color casts are small, minor. The small tweaks I gave will do nothing for restoring "true" color. Beside, I have a strong suspicion that Will wanted the warm tones.

As discussed throughout this long series of posts: 1) If we say that the gray card white balance rules, then there should be no color cast. While the gray card is not absolutely neutral, it isn't far off. And 2) If we say that reflections are causing the strong color casts (or strong sun, whatever), then the gray card white balance was irrelevant.

So I am not sure what to make of Will's colors. Overall, they don't look too bad to me except that there is some overexposure and that a small tweak in her eye whites are required to make the warm in both channels. I am okay with his warm color cast. I think the picture would be less impressive if the true colors were shown.

Honestly, my small tweak to arrive at neutral colors was easily accomplished. You can reproduce it in a heartbeat.

With regard to the confusion, I just edited the last graphic that I used that included the rgb histogram. Ignore it if you like.

 

[Joachim Bolte]

... my small tweak to arrive at neutral colors

But as I just stated they are not neutral all over... the blue is still there, just not on the specific node you chose to measure.

If you want, make 6 measuring points. three in the 'black' part, top, bottom and middle, and tree in the 'gray' part, same locations. Now try your 3-minute tweak and get all the points neutral.

 

[Asher Kelman]

1. you are in RGB. I mostly convert the picture to 16 bit before converting to Lab, but we can do without that now... In the menu, select Image - Mode - Lab

2. Add a curves adjustment layer to your layerstack

3. open it up, and set the grid to 10% increments

4. in the a as well as the b channel pull in the corner points about 40%. The line should be still passing through the middle of the histogram. If you want to be really precise, use the input/output boxes to traget your values.

5. When done, your color will be emphasized. Your lines are still through the Lab midpoint, so grays will be grays. You will clearly see that there are big blue and red patches in the image, especially on the bottom part. The way the colors form 'blocks' makes me suspect JPG-compression has something to do with it. The circular form makes me suspect it's a matter of flash vs. sunlight.

Kevin,

This is a most logical and helpful addition to the discussion. In fact the gray reference strip is indeed a pop out fabric by Wescott but now unobtainable. The light was from a ring flash and the sun. So this appears to be the basis for the shifts in color we observe. That Will like's this is part of chance that he gets the same effect at different locations. The common set up of that fabric and flexible reflector and the use of ring flash added to sunlight gives him look he "likes".

Your LAB histogram narrowed in the way you have does point out the bias the set up gives. I prefer to overwhelm the sunlight or use a reflector to add back more of it, but mixing light is a complex setup I prefer to avoid!

Asher

 

[Doug Kerr]

Hi, Will,

This is the Color Checker card I used for this shot!

Cute! Is this something new from Drew Strickland? Or from Melissa?

Best regards,

Doug

 

[Kevin Stecyk]

But as I just stated they are not neutral all over... the blue is still there, just not on the specific node you chose to measure.

If you want, make 6 measuring points. three in the 'black' part, top, bottom and middle, and tree in the 'gray' part, same locations. Now try your 3-minute tweak and get all the points neutral.

Joachim, it would be trivial to do. Just keeping adding curves with blendifs until the cows come home. And the cows would come home in less than 10 minutes total, if they were walking slowly. But my resultant set of curves would be useless in applying to Will's image.

Refocus the discussion: what part of the gray card was used in the white balance. When I use my gray card, I focus in quite closely so that I see only the gray stripe and set my white balance. On my gray card, there is a bulls-eyes type grid so that the camera is able to focus. Once my white balance is set, I zoom out and ensure that I have three spikes.

So in Will's gray balance, did he focus on the entire gray card, a portion of it, the center of it, the wonky bottom black left hand corner? Does he even remember?

In my view, from my casual perusal of it, his card is close enough. Yes, the camera left hand corner is wonky. He likely didn't focus there. I say that because I doubt he used "black" as his white balance.

I don't think this gray card issue explains the ~problem~.

 

[Kevin Stecyk]

This is a most logical and helpful addition to the discussion. In fact the gray reference strip is indeed a pop out fabric by Wescott but now unobtainable. The light was from a ring flash and the sun. So this appears to be the basis for the shifts in color we observe. That Will like's this is part of chance that he gets the same effect at different locations. The common set up of that fabric and flexible reflector and the use of ring flash added to sunlight gives him look he "likes".

Your LAB histogram narrowed in the way you have does point out the bias the set up gives. I prefer to overwhelm the sunlight or use a reflector to add back more of it, but mixing light is a complex setup I prefer to avoid!

Nope, this ain't it. I fully expect that the gray card was close enough. The values are not way off. Yes, there is some variability. That will always be the case. But it ain't enough to cause a dramatic shift in color. If that subtle shift in color was responsible for the strong amber look in Will's photograph, then photographers would have one heck of a time in using gray cards outdoors. Yet many do, including me from time to time.

Here's a test for you. Have a person with a white top on. Photograph him or her outdoors in various positions. Move your cursor around person's top. I expect you'll find that it is not completely even throughout.

Look at the image with the LAB readouts. (bottom rhs should be 100,0,0) The variability ain't that great. And I doubt that Will focused on the black portion of his gray card to obtain his white balance.

Now, perhaps Doug can chime in, what would happen if he focused on the white portion that was blown out? Could the camera even obtain a reliable reading for white balance? My guess says no. My guess says that the white needs to be well within the rgb colorspace. Again, only guesses on my part.

If he focused on the gray portion, he'd be close enough.

In summary, I don't think the gray card is your problem.

 

[Asher Kelman]

Kevin,

TTBOMK, Will uses the entire grey card, and includes in his frame the three re, gions, black, a grey and white(in this case, not a card but a pop-out material from Westcott, no longer sold, that stretches out to be a flat surface), as the neutral WB standard. He takes the picture in the light he'll use, as near as possible, (see # 87, below for further details) and selects it for the in camera custom WB setting.

Asher

 

[Kevin Stecyk]

Kevin,

TTBOMK, Will uses the entire grey card as the neutral WB standard.

Asher

Thus, I assume the gray card filled the frame or nearly the frame? Okay, so the camera picked the portion in the middle where it relatively "calm" gray. So no big deal.

The camera will preference the "bulls-eye" portion of the image.

From my 50D manual:

The plain, white object should fill the spot metering circle.

Instead of a white object, an 18% gray card (commercially available) can produce a more accurate white balance.

The LAB values of 83, 0, 0 looks pretty darn good to me. It ain't the gray card.

 

[Asher Kelman]

It appears that will does the following. He over-exposes, or better, exposes hot "somewhat" and then uses Canon's Highlight Tone Priority (HTP) function in his camera settings.

Why this would shift the color I have no idea. His ring flash is at 3ft set to auto for the shot of the Wescott black-gray-whte WB popout cloth. This shot is selected for the custom WB. Then the model shots are taken at a larger distance, again with the Canon ring flash on Auto. So in each case, sunlight and controlled flash illuminate the subject in a pretty similar fashon.

 

[Kevin Stecyk]

It appears that will does the following. He over-exposes, or better, exposes hot "somewhat" and then uses Canon's Highlight Tone Priority (HTP) function in his camera settings.

Why this would shift the color I have no idea. His ring flash is at 3ft set to auto for the shot of the Wescott black-gray-whte WB popout cloth. This shot is selected for the custom WB. Then the model shots are taken at a larger distance, again with the Canon ring flash on Auto. So in each case, sunlight and controlled flash illuminate the subject in a pretty similar fashon.

Let me see if I got this straight:

1) 3 feet, shoots with ringflash and sets his white balance.

2) Moves back with camera and ringflash and blasts away again.

Because he has moved his lights from where he set his white balance to where he shoots, he has affected his white balance.

What you should do is as follows: 1) get your lighting set up; 2) set your white balance; and 3) without changing your lighting, create your photographs.

I read in your comment that the lighting was set to auto in both instances. Obviously, as it camera and flash are moved further away, the flash has to work harder. I suspect that Canon flash maintains reasonably constant color temperature throughout its power range.

I doubt that this error caused a dramatic shift in white balance. Nonetheless, for purists, you shouldn't move your lights between white balance setting and your actual shots.

Where does Will's Colorchecker come into play?

 

[Asher Kelman]

Let me see if I got this straight:

1) 3 feet, shoots with ringflash and sets his white balance.

2) Moves back with camera and ringflash and blasts away again.

Because he has moved his lights from where he set his white balance to where he shoots, he has affected his white balance.

Could be!

I read in your comment that the lighting was set to auto in both instances. Obviously, as it camera and flash are moved further away, the flash has to work harder. I suspect that Canon flash maintains reasonably constant color temperature throughout its power range.

Could it be that at 3ft, reflectons of Will's clothes might color the light a tad?

Where does Will's Colorchecker come into play?

The pink one, LOL!!! That's from the house of the mouse, Mickey Mouse! That Disneyland humor!

Asher

 

[Doug Kerr]

Hi, Kevin,

Now, perhaps Doug can chime in, what would happen if he focused on the white portion that was blown out?

I'm not sure what scenario is being asked about, so I'll identify several and discuss each.

Scenario 1: We are speaking of obtaining a custom white balance setting for the camera to use with in-camera white balance color correction, from a "calibration" shot taken by the camera of a neutral target exposed to the scene illumination.

In a Canon camera, when we do that, the camera, when led to the "calibration" shot, regards the center of the frame. There is no opportunity for the user to "focus on" one portion or another. If we want the camera to regard a certain portion of the target at that time, we must frame the calibration shot to put that portion in the critical center area.

If the part of the calibration shot frame regarded by the camera is blown out (however that came to be), then the process is useless for white balance.

Scenario 2: We are speaking of making color correction during raw development. Then the question becomes, "what if the photographer uses the eyedropper on a portion of the image that came from a neutral part of the target, but which is blown out" (at the raw stage)?

The answer is, "this is worthless for white balance color correction".

Scenario 3: We are speaking of making color correction on an existing JPG file. Again, the question becomes, "what if the photographer uses the eyedropper on a portion of the image that came from a neutral part of the target, but which is blown out" (at the JPG stage)?

Again, the answer is, "this is worthless for white balance color correction".

The other part of the issue is the exposure used for the calibration shot. If (to make the story simple) we have only a big uniform target card with a reflectance of 0.98, or one with a reflectance of 0.31, or one with a reflectance of 0.18, and shoot metered (with the meter regarding only the card), we would get essentially the same result in each case (for any of the three scenarios) insofar as white balance is concerned. The metering system would produce the same photometric exposure on the sensor in each case, one that is far from saturation. (I ignore here the matter of whether the chromaticity of the flash burst varies with flash output).

Best regards,

Doug