• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

ColorRight developer flip-flops - film at 11

Doug Kerr

Well-known member
Drew Strickland has often made fun of reflective white balance targets, and of adjusting color balance during raw development based on a reflective neutral target in the scene.

But now he has found a way to sell the "opaque" real estate surrounding the "orifice" of the ColorRight white balance tool for something that in fact is to be used in just that technique.

In his new tool, the ColorRight Skin, he places around that opaque annulus several reflective target sectors, some of them of them nominally neutral, and the others slightly offset from neutral in a couple of chromatic directions by various amounts.

Then, one places the instrument in the scene (as a target) and takes the shot, taking the raw output (or the JPEG output). Then, during raw development (or in postprocessing color correction of the JPEG image), one can "use the eyedropper" on one of those sectors, which one depending on the direction and degree of color correction, other than "theoretically ideal", one wants to employ to get the skin tone that is desired in the developed image.

Well, that's pretty clever.

I'm not sure I can find on the ColorRight site any more the information about why using reflective targets and color correction during raw development are silly. Maybe that just came to me in a dream.

**********

It's interesting to reflect on why that opaque annular real estate is there in the first place. Drew has said that it is responsible for the ColorRight tool being "more targeted" than other white balance measurement diffusers. What does that mean? Well we think it means, "has a narrower acceptance pattern". And it does have such.

So, is that desirable?

No.

So, does masking out part of the face of the diffuser narrow its acceptance pattern?

No. In fact, with the annular mask removed, the acceptance pattern is almost the same.

So although the designer seeks, and promotes, a not-necessarily desirable result, fortunately the implementation doesn't provide it!

But in fact, the ColorRight nevertheless does have a narrower acceptance pattern than traditional white balance measurement diffusers. How does that happen?

Well, it is inherent in any reasonable simple diffuser design, especially one that needs to have a relatively high transmissivity (said to be another beneficial feature of the ColorRight unit).

In fact, traditional measurement diffusers (such as the ExpoDisk diffuser) require the use of such components as lenticular plates in front of the basic diffuser plate to widen the acceptance pattern. Without those, the pattern would inherently be relatively narrow.
 

Michael Tapes

OPF Administrator/Moderator
Hi Doug,

I guess Drew was against reflective WB reference before he was for it. Anyway there is no better reflective reference than WhiBal so the more he talks the "proper" talk the better. And in an interesting coincidence, the WhiBal G7, under development, has similar "warmer/cooler" patches to Drew's new device. Of course many have already used this idea, including QP and WarmCards, and ColorEyes Gelcard. So we are both following others in this regard. But the G7 has some other very useful features that I believe will set it apart. And of course we will continue to certify (measure) each and every card to assure absolute neutrality.

Was that "10 WB devices compared" test every published?

Thanks for the heads up.
 

Doug Kerr

Well-known member
Hi, Michael,

And in an interesting coincidence, the WhiBal G7, under development, has similar "warmer/cooler" patches to Drew's new device. Of course many have already used this idea, including QP and WarmCards, and ColorEyes Gelcard.
Yes, and I think it is a clever idea (even as deployed in the ColorRight)

Was that "10 WB devices compared" test every published?
I don't remember who was going to do that test, or if it was published.

Thanks for the heads up.
Le crayon rouge ne dort jamais.

By the way, although I didn't mention it in my message at the head of this thread, the ColorRight Skin is also said to have a "a new lighter skin balancing center". I don't know what that would mean (fancy that, for something said in a ColorRight ad!), but the subsequent text suggests that the basic quasi-diffuser in the ColorRight Skin may not be nominally neutral but rather provides some chromaticity offset to give an "an average pleasing skin tone".

Sounds great. For me, I really like Halle Berry's skin tone (although she's hardly average).

Best regards,

Doug
 

Doug Kerr

Well-known member
Update - the target-with-a-hole formerly known as

We are nonplussed to announce that the device formerly known as the ColorRight Skin is now known as the ColorRight Max (as of 2008.09.24 2340 GMT).

The manufacturer has apparently now totally moved to promoting the color correction in post processing with a reflective target paradigm (you know, the one they used to decry as a waste of time).

The main page warns, "Please note that basic computer editing is required to fully utilize all capabilities of the COLORRIGHT MAX line of tools. "

Their pseudo-diffuser only product, the basic ColorRight, is now relegated to a page reached through a link designated, "Please follow this link if you are looking for our groundbreaking tools that do not require a computer."

Science marches on!

Best regards,

Doug
 
Their pseudo-diffuser only product, the basic ColorRight, is now relegated to a page reached through a link designated, "Please follow this link if you are looking for our groundbreaking tools that do not require a computer."

Snake oil apparently helps to slither fast, doesn't it?

Science marches on!

While I'm cautious about trouncing (how about covering all bases here?) a product that I don't fully grasp (ouch?), I do tend to adhere to boring science when marketoid speak prevails.

Go Doug, go (meaning I do agree with your observations, and then some)...
However, let's keep it civil (and nasty if deserved).

Bart
 

Michael Tapes

OPF Administrator/Moderator
Quote from the manufacturer's own unbiased test results, regarding post production methods (which they now embrace)....

Both of these approaches have a few pluses, but they have even more minuses.

Looks like they have taken their sights off of ExpoDisc and put them on WhiBal. Or maybe it is the Kodak Gray card they are after, or maybe there is more profit this way (not that there is anything wrong with that). Let's hope the thing is really as neutral as they say so at least people will get a good product for their $129. Even if they do meet their new spec, WhiBal is still more neutral, and far more cost effective. The warmer and cooler is a good feature but is far from original (not that there is anything wrong with that).

OT: many of you might be interested in my new Instant JPEG from Raw utility as I desribed yesterday on Scott Kelby's blog

Of course the profit for me is not so high. I am giving it away for free (not that there is anything wrong with that). Just trying to give back to a community that has treated me well. The good folks at Imagenomic (Noiseware) coded it for me. It was a cooperative project.
 

Nill Toulme

New member
Very cool, Michael, thanks for this.

Just by way of historical context, BreezeBrowser and BreezeBrowser Pro have long included the ability to extract those RAW-embedded jpg's. And it's long been a mystery to me why the camera manufacturers' own software does not.

Those programs also, by the way, make it much more practical and convenient to shoot RAW + jpg, in that they treat the two files (and now their accompanying xmp sidecars) as one file for many purposes, including display, copying, moving, deleting, renaming, etc. etc.)

When may we expect to see Raw Converter? I've been holding my breath for it, and I'm starting to turn a little blue-ish around the edges... ;-)

Nill
 

Doug Kerr

Well-known member
Hi, Michael,

Looks like they have taken their sights off of ExpoDisc and put them on WhiBal. Or maybe it is the Kodak Gray card they are after, or maybe there is more profit this way (not that there is anything wrong with that).

One interesting wrinkle is that Drew's literature about the new order of things seems to imply that doing white balance color correction in post processing is just as attractive for JPG files as for raw. (He does mention that some editors might not have the feature of doing color correction on JPG files.)

My understanding has always been that doing WB color correction starting with a JPG file is a very poor second compared to doing it during development of the raw file. Do you subscribe to that outlook?

Best regards,

Doug
 

Michael Tapes

OPF Administrator/Moderator
Hi Doug,

As you well know, adjusting the WB of the JPEG after the fact is trying to fix something that is broken. Adjusting the raw WB is building something from scratch. Like trying to fix a car that has been in an accident, the JPEG will never quite be right. And like the car, the more damage to the JPEG (deviation from desired WB) the harder it is to fix and the more wrong it will be, as opposed to the raw.

Of course the key to adjusting the JPEG WB in post is getting it as close to "correct" as possible so that the adjustment is small (again like the car).

Of course the fact that JPEG is only 8-bit is only one of the factors that cause this problem, the other is that one will have to construct data, whereas with the raw the data is only manipulated as opposed to constructed.

So, yes, Doug we are in agreement.
 

Doug Kerr

Well-known member
Hi Doug,

As you well know, adjusting the WB of the JPEG after the fact is trying to fix something that is broken. Adjusting the raw WB is building something from scratch. Like trying to fix a car that has been in an accident, the JPEG will never quite be right. And like the car, the more damage to the JPEG (deviation from desired WB) the harder it is to fix and the more wrong it will be, as opposed to the raw.

Of course the key to adjusting the JPEG WB in post is getting it as close to "correct" as possible so that the adjustment is small (again like the car).

Of course the fact that JPEG is only 8-bit is only one of the factors that cause this problem, the other is that one will have to construct data, whereas with the raw the data is only manipulated as opposed to constructed.
That all makes sense to me, and I enjoy your succinct characterization of the matter.

Best regards,

Doug
 

Doug Kerr

Well-known member
Well, silly me

When I first reported on the new COLORRIGHT SKIN tool (now known as the COLORRIGHT MAX - wait, let me check - yes it still is), I said that evidently the material originally on the Web site for the product, sarcastically reflecting negatively on doing white balance color correction in post-processing (the approach used with the COLORRIGHT MAX), was no longer there.

I guess that was just another example of my impatience in finding stuff (Carla could certainly tell you about that!). That material is still available, here:

http://www.colorright.com/pro-white-balance-review

(I think I missed it because it was under the "Pro Review" tab, not in the actual product area.)

It makes fun of the post-processing color correction concept, and of the neutral targets used in it, and illustrates the point with a photo of a cute, but sullen looking, model holding those silly white and gray cards.

Incidentally, Drew, congratulations on having attained the milestone of the sale of over 1000 units of the Color Parrot and COLORRIGHT products.

Best regards,

Doug
 

Doug Kerr

Well-known member
Silly me - part deux

As I further read the descriptive material on the new COLORRIGHT MAX tool (a "color wheel" reflective target), I see that I had slightly misconstrued its objective.

I had thought that the object of it having a number of "patches" of differing chromatically was to allow to the photographer to intentionally "bias" the color correction to attain the desired artistic rendering of the model's skin in the delivered image. This would be particularly attractive to photographers with a "germ phobia", who would not want to touch a slider in their postprocessing software someone else might have touched.

But now, reading more patiently, I see that the intent of the different chromaticity patches is to allow accommodation of models with differing actual skin tones - "darker" and "lighter". I leave the technical, artistic, or social interpretation of that up to the reader.

Somehow, I am reminded of what I have called the "dirty little secret" of the YIQ color space (the one used in the NTSC color television "coding" system, developed in the late 1940s).

The modulated signals for the two chrominance axes, I and Q, were given different amounts of bandwidth (in the interest of overall bandwidth conservation), and the orientation of the axes was chosen so that the best chromaticity resolution (along the "I" axis) was between small variations in skin color (for which read, "Caucasian skin color").

Best regards,

Doug
 

Michael Tapes

OPF Administrator/Moderator
Bought one for test evaluation as I do with most WB products. Shipping is $10 for slow boat! Good profit in that s/h.<g> For me it is a business expense and I do not really care, but for the average customer, +$10 to ship something light...it could go Priority Mail flat rate for $4.75..... Oh well the buyer decides.

Always interesting to evaluate products.

See ya all.
 

Doug Kerr

Well-known member
Hi, Michael,

Bought one for test evaluation as I do with most WB products.

Always interesting to evaluate products.
I'll be anxious to learn of your observations.

We have taken a vacation from evaluating white balance tools. The economic and emotional load is just not justified as a pro bono activity.

Best regards,

Doug
 

Doug Kerr

Well-known member
Lambert, be still

When we place a transmissive diffuser on a camera lens to collect light for white balance measurement, we are not in the main drawing upon the "familiar" purpose of a diffuser: to spread out the light emitted from its far side over a range of angles.

Rather, we are drawing upon the converse (which we have owing to reciprocity): the device "collects" light arriving over a range of angles.

In fact these two complementary concepts are both implied by Lambert's law, seen as defining a "perfect diffuse reflective surface", as we adapt it to a diffuse transmissive object.

As I was seeking to grasp the manufacturer's concept of the operation of the Color Parrot white balance tool (a diffuser of a sort), to see how the concept I mentioned above might fit it, I ended with, "So, Drew, what is it that we expect the diffuser to do in this process, anyway?"

The reply was simple and straightforward. "Why Doug, it diffuses."

And I used to think there was no science involved in the concept of that product.

Best regards,

Doug
 

Doug Kerr

Well-known member
The problem with many diffusers

While waiting for some epoxy to set, I happened upon the Pro Reviews page of the COLORRIGHT Web site (at the ProPhoto Home forum), which I still had open in a browser instance from some previous scientific research.

As you may recall from my earlier observations, this is a page that principally serves to denounce various white balance color correction techniques, including doing the color correction in post processing (you know, the technique for which the new COLORRIGHT MAX is intended).

My eye lit on this passage, part of the manufacturer's criticism of other "measurement diffuser" tools (described as "white balance caps"):

"These caps act as an 'integrator of broad light acceptance angle:'

What does that mean? One of the main challenges with most existing white balance diffusers, is that they treat all light entering the lens the same. In other words, light at the edge of the frame is given a similar weight to light at the center of the frame.
"

Well, this passage I'm afraid reflects this manufacturer's paucity of grasp of the principles of photometry and the like. Let me comment on two phrases:

"...they treat all light entering the lens the same"

Hardly. The "classical" diffuser gives different weight to light striking the diffuser from different angles of arrival, in the most classical case weighting by the cosine of the angle of arrival.

"In other words, light at the edge of the frame is given a similar weight to light at the center of the frame."

What does that mean, indeed? Is it a reference to light emerging from parts of the scene that will be on the edge, versus in the center, of the object frame when the diffuser is removed to actually take the shot? The diffusers being criticized have no idea what the field of view of the lens proper is, so they couldn't do that even if their designers for some reason aspired to such a behavior.

Or does it somehow refer to light that emerges from the rear of the diffuser and is "imaged" by the lens onto the edge, versus the center, of the sensor frame during measurement? It doesn't sound like it, since the reference to "being given the same weight" suggests a treatment of arriving, not "delivered", light. So it is hard to imagine what that phrase might actually mean.

I'm afraid that the entire cited passage is some type of pseudo-scientific (or anti-scientific) mumbo-jumbo.

They well ask, at the introduction of the passage proper, "what does that mean".

But, one can see Russia from Georgia.

Best regards,

Doug
 

Doug Kerr

Well-known member
Hi, Michael,

And in an interesting coincidence, the WhiBal G7, under development, has similar "warmer/cooler" patches to Drew's new device.

Well, the ColorRight MAX doesn't have "coolering" patches - just "warmering". (Well, it's always possible that the "neutral" patches are a little coolering - I haven't seen any data on that.)

So if the G7 has both, that could be a handy advantage.

I'm of course not "defending" it - I don't even know what it is!

Best regards,

Doug
 
Top