Does “Color Science” even matter?

[just Tony]

Opinion piece, nothing more:

The Carnot Cycle was a product of Science. Steam locomotives were mostly the product of Engineering, and in the most beloved examples, with the collaboration of Art as well.

I have an impression that the so-called "Canon colors" and Fuji film emulations, etc. were implemented by Color Engineers who worked towards goals set by whom I would call Color Artists, or perhaps Art Directors. Color Engineers create practical applications and useful systems in hardware, firmware and software that are founded on the base knowledge that was developed by pioneering Color Scientists.

Out of those three job categories I suspect that the most prevalent of them at any given camera company are Engineers. I also think that quite a lot of this thread is about the Art of Color, and not very much about the expansion of the frontiers of fundamental knowledge per se.

I consider myself to be a Color Technician who has motivations and ambitions toward Color Art.

 

[Stan Disbrow]

Hi,

IBM sponsored Technical Electronics courses in the community colleges and the high schools in Dutchess (East Fishkill and Poughkeepsie plants) and Ulster (Kingston) counties in New York State. And probably other places but I was in the Mid Hudson Valley and so took advantage of this in High School.

That saw me leave High School with an Associate Degree as if I had been to the Community College. It also got me in the IBM door working as a Technician. Once I was there for a few months, I was introduced to classes at RPI via video conference. This, as part of my workday and so, eventually, a Bachelor's Degree in Electrical Engineering. I took six years to get there rather than four, but it was all paid for by IBM. Same happened over the next three years with a Master's Degree.

I don't think this happens with IBM and RPI any longer. It was still underway when IBM was into downsizing and I left for Ericsson in 1994. Ericsson had a different approach where folks took off from work for a semester as a time and went to school instead. At the rate of one semester per year.

What I liked about the IBM and RPI scheme was we used actual product R+D work by way of our lab work for school. And the job promotions came along automatically. So, not only on the company's dime, but more dimes in my pocket as well.

And then they show one the door and waste all that effort and money. You'd have thought they'd have kept their engineers that they had bothered to train their way....

By that time I was in North Carolina instead of New York.

In my case, I was stuffing radios into notebook PCs at that time. We used Ericsson GE for our vendor. I wandered out the IBM door on Friday and into Ericsson's door on Monday. On Tuesday, I was back in my lab at IBM, at my bench, doing what I had been doing on the previous Thursday. And it cost IBM three times what it cost them for my work when I was an employee.

Stan

--
Amateur Photographer
Professional Electronics Development Engineer
Once you start down the DSLR path, forever will it dominate your destiny! Consume
your bank account, it will! Like mine, it did!

 

[just Tony]

Hi,

Ah, yes. RIT. Which means Kodak. And we are back to my earlier comment that Kodak had it all under control color-wise long ago.

Kodak utilized RIT in the same way GE and IBM utilized RPI. So I'm a RPI guy having been an IBM guy.

And HP utilized Stanford. Fred Terman was the force behind the founding of HP, and their headquarters was on Stanford land. If you came to HP with a BSEE, you got your masters at Stanford on the company’s dime.

Many of my coworkers in Materials Science got their advanced degrees that way, primarily PhD. I worked in Building 1 within the 6 building complex just outside of the campus. The annual lease for all of the land under them was commonly said to be $1.

You already know this: The surrounding area in Palo Alto that has incubated a lot of tech is called Stanford Park.

And just downhill were my favorite Kodak processing lab (same day Kodachrome) and Keeble and Schuchat's two stores. Those were the days!

--
Wag more; bark less.

 

[JimKasson]

I would think that 'hot mirror' design may also play a role.

Their brick wall passbands are readily adjusted via the vacuum deposition processes to any desired wavelength but I don't know if brick wall behavior is always what is needed there.

The CFA pigments and dyes are chosen in concert with the hot mirror, and in consideration of the pixel spectral response. So a brick wall stop band may not be what’s desired.

Camera sensors are sensitive to IR, strong filtering near IR may affect reproduction of reds but having weak IR filtering may cause unnatural colors. Some early Leica M models had serious issues with rendering black textiles that have high IR content.

I recall the odd photos of the "burners" on an electric stove top that were captured by my Sony RX100 mark 1. I would describe the visual experience as a "deeply saturated red", but the camera recorded them as a saturated purple.

That is almost certainly because the blue CFA filter spectrum and the hot mirror in tandem are passing long wavelengths.

With film, an analine dyes used to produce some spectacular capture metameric errors, such as turning black tuxedos pink.

 

[Stan Disbrow]

Hi,

I'm an engineer in several disciplines, including steam engines.

But I am a color technician as well. I get it all calibrated pretty well. What I'd call Good Enough. And then get on with it all, trying for the Artist part.

Stan

--
Amateur Photographer
Professional Electronics Development Engineer
Once you start down the DSLR path, forever will it dominate your destiny! Consume
your bank account, it will! Like mine, it did!

 

[JimKasson]

Hi,

Ah, yes. RIT. Which means Kodak. And we are back to my earlier comment that Kodak had it all under control color-wise long ago.

Kodak utilized RIT in the same way GE and IBM utilized RPI. So I'm a RPI guy having been an IBM guy.

And HP utilized Stanford. Fred Terman was the force behind the founding of HP, and their headquarters was on Stanford land. If you came to HP with a BSEE, you got your masters at Stanford on the company’s dime.

Many of my coworkers in Materials Science got their advanced degrees that way. I worked in Building 1 within the 6 building complex just outside of the campus. The annual lease for all of the land under them was commonly said to be $1.

You’re ex-HP? Or, as we used to always write it, hp? I worked at 395 Page Mill Road and later at AMD in Sunnyvale.

--
https://blog.kasson.com

 

[JimKasson]

Opinion piece, nothing more:

The Carnot Cycle was a product of Science. Steam locomotives were mostly the product of Engineering, and in the most beloved examples, with the collaboration of Art as well.

I have an impression that the so-called "Canon colors" and Fuji film emulations, etc. were implemented by Color Engineers who worked towards goals set by whom I would call Color Artists, or perhaps Art Directors. Color Engineers create practical applications and useful systems in hardware, firmware and software that are founded on the base knowledge that was developed by pioneering Color Scientists.

Out of those three job categories I suspect that the most prevalent of them at any given camera company are Engineers. I also think that quite a lot of this thread is about the Art of Color, and not very much about the expansion of the frontiers of fundamental knowledge per se.

I consider myself to be a Color Technician who has motivations and ambitions toward Color Art.

 

[JimKasson]

Opinion piece, nothing more:

The Carnot Cycle was a product of Science. Steam locomotives were mostly the product of Engineering, and in the most beloved examples, with the collaboration of Art as well.

I have an impression that the so-called "Canon colors" and Fuji film emulations, etc. were implemented by Color Engineers who worked towards goals set by whom I would call Color Artists, or perhaps Art Directors. Color Engineers create practical applications and useful systems in hardware, firmware and software that are founded on the base knowledge that was developed by pioneering Color Scientists.

Out of those three job categories I suspect that the most prevalent of them at any given camera company are Engineers. I also think that quite a lot of this thread is about the Art of Color, and not very much about the expansion of the frontiers of fundamental knowledge per se.

I consider myself to be a Color Technician who has motivations and ambitions toward Color Art.

Agree.

Science is a systematic endeavor that builds and organizes knowledge in the form of testable explanations and predictions about the universe.

Engineering is the use of scientific principles to design and build machines, structures, and other items,

Figuring out how humans perceive color is science. Applying that knowledge to cameras and raw developers is engineering. As you say, that engineering is informed by artistic decisions. The people designing color film were masters of that kind of engineering.

It is not uncommon for technical fields to wrap themselves in the cloak of science. Computer Science is an example. Most of computer science is mathematics or engineering.

 

[just Tony]

• You’re ex-HP? Or, as we used to always write it, hp? I worked at 395 Page Mill Road and later at AMDin Sunnyvale.

Was that the Redwood Building? K&S would have been a leisurely lunch hour walk for you.

I was doing polymer synthesis at 1501 Page Mill, making negative resists for direct exposure by electron beam and X-Ray. I adjusted the sensitivity and contrast by selecting ranges of molecular weight (the chain length).

The initial project for HP inket printers (in an unrelated research group of course) used a few bottles of solvents that came out of "my" cabinets.

--
Wag more; bark less.

 

[JimKasson]

• You’re ex-HP? Or, as we used to always write it, hp? I worked at 395 Page Mill Road and later at AMDin Sunnyvale.

Was that the Redwood Building?

Was that the Redwood Building?

The Redwood Building was in the complex, but I worked in the main building, which I think was the first hp building with roof ripples. The division I worked for was initially called Dymec, and old timers filed their hp catalogs upside down so the logo read like the the Dymec logo, which was dy inside a circle,

K&S would have been a leisurely lunch hour walk for you.

K&S?

I was doing polymer synthesis at 1501 Page Mill, making negative resists for direct exposure by electron beam and X-Ray. I adjusted the sensitivity and contrast by selecting ranges of molecular weight (the chain length).

In the research labs? When?

The initial project for HP inket printers (in an unrelated research group of course) used a few bottles of solvents that came out of "my" cabinets.

Great. That sort of answers the when question. When I was doing color work for IBM at Almaden, I worked with some of the hp color folks. That was in the early 90s, so quite a bit later.

 

[SrMi]

• You’re ex-HP? Or, as we used to always write it, hp? I worked at 395 Page Mill Road and later at AMDin Sunnyvale.

Was that the Redwood Building?

Was that the Redwood Building?

The Redwood Building was in the complex, but I worked in the main building, which I think was the first hp building with roof ripples. The division I worked for was initially called Dymec, and old timers filed their hp catalogs upside down so the logo read like the the Dymec logo, which was dy inside a circle,

K&S would have been a leisurely lunch hour walk for you.

K&S?

Keeble & Shuchat was the largest photo store in Bay Area.

I was doing polymer synthesis at 1501 Page Mill, making negative resists for direct exposure by electron beam and X-Ray. I adjusted the sensitivity and contrast by selecting ranges of molecular weight (the chain length).

In the research labs? When?

The initial project for HP inket printers (in an unrelated research group of course) used a few bottles of solvents that came out of "my" cabinets.

Great. That sort of answers the when question. When I was doing color work for IBM at Almaden, I worked with some of the hp color folks. That was in the early 90s, so quite a bit later.

 

[JimKasson]

• You’re ex-HP? Or, as we used to always write it, hp? I worked at 395 Page Mill Road and later at AMDin Sunnyvale.

Was that the Redwood Building?

Was that the Redwood Building?

The Redwood Building was in the complex, but I worked in the main building, which I think was the first hp building with roof ripples. The division I worked for was initially

called Dymec, and old timers filed their hp catalogs upside down so the logo read like the the Dymec logo, which was dy inside a circle,

K&S would have been a leisurely lunch hour walk for you.

K&S?

Keeble & Shuchat was the largest photo store in Bay Area.

Oh, I thought it was an HP division I didn’t know about. I once stopped by with my son and afterwards, he said, “Dad, everybody in the store knows you!”. Terry was a good guy, and the pro store on the south side of California Avenue was a candy shop.

I was doing polymer synthesis at 1501 Page Mill, making negative resists for direct exposure by electron beam and X-Ray. I adjusted the sensitivity and contrast by selecting ranges of molecular weight (the chain length).

In the research labs? When?

The initial project for HP inket printers (in an unrelated research group of course) used a few bottles of solvents that came out of "my" cabinets.

Great. That sort of answers the when question. When I was doing color work for IBM at Almaden, I worked with some of the hp color folks. That was in the early 90s, so quite a bit later.

 

[just Tony]

K&S would have been a leisurely lunch hour walk for you.

K&S?

Sorry too much shorthand; Keeble and Schuchat.

In the research labs? When?

'78 to '84. I likened my polymer chain lengths to silver halide grain size. In my mental model diversity of size leads to diversity of particle capture chances leading to lower contrast.

 

[JimKasson]

K&S would have been a leisurely lunch hour walk for you.

K&S?

Sorry too much shorthand; Keeble and Schuchat.

In the research labs? When?

'78 to '84. I likened my polymer chain lengths to silver halide grain size. In my mental model diversity of size leads to diversity of particle capture chances leading to lower contrast.

Interesting. I knew and liked John Doyle. Tom Whitney was gone by then, I think.

 

[Stan Disbrow]

Hi,

Yep. They do go hand in hand. Science and Engineering.

The term Engineer comes from the steam engine. A manually controlled power plant that loved to blow sky high if the guy running the engine wasn't paying close attention.

Science figured out that if one heated water in a closed vessel, one got steam that increases in power with increased temperature and pressure. And, scientists were also likely the first ones to figure out that the vessel blew sky high if they weren't careful.

The early engineers were the ones who figured out how to build that vessel so it didn't blow up. Much. Mostly. And then how to take that steam energy and make useful power out of it. And each new engine was designed to be better than the one that came before it. All the way from simple pistons to compound ones to turbines.

NASA is often referred to as Rocket Science. Nope. Engineering. Some Science, for sure, but lots of Engineering. Mostly Engineering.

And so, too, color. Kodak all the way for a long time. Film and then Digital. They had all sorts of different sensor silicon and CFA types and dyes. And post processing software and even printers. And not just color, but also monochrome for both film and digital. All engineered to do different jobs. It was all pretty comprehensive, actually. IR light, Visible light, UV light and mono and color for all those. Even different filters to go with them. Gone now. No one does all that any longer.

But what we do have now is built on top of what they did. Fuji in particular paid pretty good attention to what Kodak was doing. The first Fuji, the DS-5xx series, was up against the Kodak DCS 500 and 600 series in the marketplace. I noticed at the time that the second pass from Fuji was a whole lot different.

And so it always goes. Engineers pay attention to what other engineers do. I recall a Far Side comic we had around many places. It showed three guys around a fire holding meat and grimacing. The fourth guy is alone at another fire and has his meat on a stick. The caption is: Hey! Look what Zog do!

We had a guy in my area at IBM, known as Z (Zahorski). There's one in every R+D group. The guy with all those patents. Z was that guy. And so we stuck copies of that comic in all the labs and crossed out the OG in ZOG and left just the Z. Got the point across better than the old Think signs.

And, in the scheme of color, I am most decidedly a Technician. Not an Engineer, and probably will never get there. I certainly will never get to Scientist.

Stan

 

[JimKasson]

Hi,

Yep. They do go hand in hand. Science and Engineering.

The term Engineer comes from the steam engine. A manually controlled power plant that loved to blow sky high if the guy running the engine wasn't paying close attention.

Science figured out that if one heated water in a closed vessel, one got steam that increases in power with increased temperature and pressure. And, scientists were also likely the first ones to figure out that the vessel blew sky high if they weren't careful.

The early engineers were the ones who figured out how to build that vessel so it didn't blow up. Much. Mostly. And then how to take that steam energy and make useful power out of it. And each new engine was designed to be better than the one that came before it. All the way from simple pistons to compound ones to turbines.

NASA is often referred to as Rocket Science. Nope. Engineering. Some Science, for sure, but lots of Engineering. Mostly Engineering.

And so, too, color. Kodak all the way for a long time. Film and then Digital. They had all sorts of different sensor silicon and CFA types and dyes. And post processing software and even printers. And not just color, but also monochrome for both film and digital. All engineered to do different jobs. It was all pretty comprehensive, actually. IR light, Visible light, UV light and mono and color for all those. Even different filters to go with them. Gone now. No one does all that any longer.

But what we do have now is built on top of what they did. Fuji in particular paid pretty good attention to what Kodak was doing. The first Fuji, the DS-5xx series, was up against the Kodak DCS 500 and 600 series in the marketplace. I noticed at the time that the second pass from Fuji was a whole lot different.

And so it always goes. Engineers pay attention to what other engineers do. I recall a Far Side comic we had around many places. It showed three guys around a fire holding meat and grimacing. The fourth guy is alone at another fire and has his meat on a stick. The caption is: Hey! Look what Zog do!

We had a guy in my area at IBM, known as Z (Zahorski). There's one in every R+D group. The guy with all those patents. Z was that guy. And so we stuck copies of that comic in all the labs and crossed out the OG in ZOG and left just the Z. Got the point across better than the old Think signs.

And, in the scheme of color, I am most decidedly a Technician. Not an Engineer, and probably will never get there. I certainly will never get to Scientist.

Stan

A lot of the actual science in what we loosely refer to as color science came from Kodak.

Stiles. MacAdam. Hunt. Wyscecki. Etc.

We engineers built on that.

OBTW, IBM had some great psychologists-- real scientists -- on staff at Watson Research Center. I worked with Bernice Rogowitz, and she was very helpful, in addition to being a really nice lady.

 

[TechTalk2]

I’m curious how many universities offer degree programs in color science vs universities offering degrees in signal processing (EE), optical engineering, Ui etc. I’m guessing 10% or maybe even 1%. I dont know the exact number but I’m thinking that itself should tell us how important color science is in the practical scheme of things.

Whether someone has a degree in color science or a degree in another field which intersects with color science knowledge and application (and many fields do); color science is incorporated into an array of devices and technologies — which are useful as well as practical in any scheme of things. Color scientists are making very practical contributions to both scientific knowledge and useful devices — like cameras, displays, projectors, printers, software, etc.

As one example, Ján Morovic received his PhD in Color Science from the University of Derby - Colour & Imaging Institute in the U.K. in 1998 and joined HP Barcelona in 2003 putting his scientific knowledge and research efforts into the development of large format printers. He served for a time as Director of the CIE Image Technology Division. He has well over 100 patents. He is also active in publishing books and articles. His book on color gamut mapping is included in the Society for Imaging Sciences and Technology collection of books on color science and he is a contributor to the book Colorimetry: Understanding the CIE Systems (link is to Internet Archive Library version if you want to save $100 by reading it online) which is found in the previously mentioned and linked book collection and also in the selected books recommended in the wikipedia article on color science. He also still finds time to contribute interesting articles, along with a group of scientists from other varied fields, to a blog called Wonderverse.

That's an example of one color scientist advancing the scientific knowledge base and practical application of color science to products. There are definitely others that come immediately to mind as well — like Karl Lang who, among other things, designed the last and best CRT monitor I used (Sony Artisan) and currently works with the International Committee for Display Metrology (ICDM) in developing standards for displays — but reviewing any more of their contributions may make me feel even lazier than I do already.

 

[yogi4fitness]

Again, I am not saying color science doesn't matter. All I am asking is where it fits in the practical scheme of things. For example, would you buy a glitchy camera that has accurate colors? Would you buy a lens thats gives you distorted images but accurate colors?



As I said maybe this calls for a new poll

 

[SrMi]

Again, I am not saying color science doesn't matter. All I am asking is where it fits in the practical scheme of things. For example, would you buy a glitchy camera that has accurate colors? Would you buy a lens thats gives you distorted images but accurate colors?

As I said maybe this calls for a new poll

I believe that most photographers, and especially artists, do not care about accurate colors.

 

[JimKasson]

Again, I am not saying color science doesn't matter. All I am asking is where it fits in the practical scheme of things. For example, would you buy a glitchy camera that has accurate colors? Would you buy a lens thats gives you distorted images but accurate colors?

As I said maybe this calls for a new poll

Very few users want accurate color. Most users want pleasing color. There are applications where accuracy is an important goal, and the most effective way to deal with that seems to be to add more spectra to the captured channels, restrict the number and variety of the illuminants, and do the same with the subject reflectance spectra. It is theoretically possible to build a three channel Luther-Ives camera, but there are drawbacks to trying to do that. I know of no commercial camera that claims to meet that criterion.

The design of a LI camera is an engineering problem. The science has been worked out for decades.

--
https://blog.kasson.com