Learning the fp(m)

(Ed. note: Before we get going here, I know that very few people are directly interested in my explorations of the fp(m). It's a quirky, offbeat camera, not a mainstream choice, and few people will ever see one much less own one. But I'm recounting some of my experiences with it to convey a sense of what I go through when I familiarize myself with a new camera and get set up to use it. You don't have to be familiar with every camera to be a good photographer...you just have to be thoroughly familiar with yours.)

So here's one thing I worked out last night: I've been griping about the lack of an ISO dial on the fp(m). I can assign ISO to the rear wheel—you know, the four-way pad made into a dial? But I hate fiddling with those. I'm too clumsy and my fingers are too big, and I'm forever activating one of the buttons when all I want to do is spin the dial. But I figured out a different way. If I leave the Quick Menu set on ISO, it will stay there. I did not know that. And then the top dial, which is comfortable and unambiguous, will change ISO. So I have the fast ISO change dial I wanted; I just didn't know it. All I have to do for my (relatively infrequent) ISO changes is hit the QM button (which Sigma labels "QS," for "quick set") and use the top dial. Viola, as the guy in the string quartet used to say. There's nothing else in the QS menu I'll be wanting to reset with any regularity.

ISO tests
Speaking of ISO, here's another thing about the stripped sensor that my tests last night made clear. Since I had the tripod set up and a test scene (a.k.a. my kitchen cooktop) ready, I did an ISO series, from 100 to 6400. Take a look at these 100% details (they'll be at 100% after you click on them, and you can pull both up at the same time to compare them side-by-side):

ISO 200 (100% after you click on it)

ISO 3200 (100% after you click on it)

Since a lot of digital camera noise at high ISOs is chroma noise, the monochrome sensor is relatively insensitive to changes in ISO. ISO 6400 is perfectly usable, and I can honestly hardly tell the difference between ISO 200 and ISO 800. Not sure I could pick out which is which in a blind test. I plan to do one more series outdoors with full-spectrum midday sunlight today to confirm.

Oh, and I discovered that colored filters on the lens make virtually no difference. They should, if my obviously incomplete understanding of the physics had been correct—but they don't. So, despite my historical fondness for yellow filters with B&W, I'm going to stop using them. At least that will restore the native sensitivity of the monochrome sensor. Depending on the light, the mono sensor is up to two stops more sensitive than a sensor with a color filter array (CFA).

Mike

Featured Comments:

Jason Melancon: "Well I'm glad I'm not the only one. First thing I did when I was initially testing the fp(m) was put yellow and red filters on the lens and then be puzzled for a while. (And I did well in my physics courses way back in yesteryear, so I didn't have an excuse.) But I quickly became so enamored with the fp(m) mono images that my chagrin was mollified."

MikeR: "I may never buy one, but I'm still interested in what you're doing with that unique camera."

Ken Tanaka: "Your higher-ISO observation with this fp(m) seems perfectly consistent with Leica's M10 Monochrome, as reported by David Farkas of Leica Store Miami in 2020."

Ken James: "I am loving your explorations of the Sigma even though I have zero chance of ever owning one. Keep it up. Thanks."

MHMG: "Mike wrote: 'Oh, and I discovered that colored filters on the lens make virtually no difference. They should, if my obviously incomplete understanding of the physics had been correct—but they don't. So, despite my historical fondness for yellow filters with B&W, I'm going to stop using them.'

"That's a rather sobering conclusion in one's quest for the 'ideal' monochromatic camera sensor!

"Perhaps Ctein or some other qualified expert can weigh in on this, so correct me if I'm wrong because it's not my area of expertise, but I recall that the native spectral sensitivity of silver halide emulsions is in the the UV/blue spectrum. It took decades of research to push that sensitivity further (using chemical sensitizing dyes) into the green for 'orthochromatic' spectral response and then finally into the red for 'panchromatic' spectral response, whereas the native spectral response of silicon sensors is almost the inverse, i.e., high native sensitivity in the red/infrared, lower in the green, and lowest in the UV/blue portion of the spectrum. This spectral imbalance gets corrected when the sensor is designed to deliver accurate/pleasing color response from, for example, a Bayer patterned array, i.e., amplifying the red, green, and blue filtered pixels to different levels. This native sensor response would also account for why the blue channel tends to show the most noise in modern digital cameras because it needs the greatest sensitivity/gain correction. Lastly, it may explain why traditional red and yellow filters used so successfully with B&W films have so much less effect (hard to believe it's zero, but still logically much less) when a modern silicon sensor designed for color imaging gets stripped of its bayer color pattern array and reverted to a 'naked' silicon sensor.

"Anyway, it would be interesting to learn what the true spectral response of your converted-to-monochrome digital camera is really like, but from what you say, it sounds like it cannot be expected to match output tonal rendering similar to traditional B&W film stocks. Seems a deal-breaker to me if that was the goal. However, Mike's recent monochrome imaging results are impressive on their own merits, so perhaps I should just 'get over it.'  :-) "

Jon Maxim: "Looking at the side-by-side 100% images I believe I can see the difference quite clearly in the midtones, like a kind of 'grain.' Look at the salt shaker and I think you will see what I mean—even in the shaded part of the white background. However, it is not unpleasant—kind of like comparing Tri-X to Plus-X."

Christer Almqvist: "Film photography was a great learning experience, but through it I also acquired a bad habit; thinking that shooting at low ISO was a good thing in itself. Perhaps it was—then. But now I shoot digital. It has been really difficult to get rid of this low-ISO-is-good habit. Now I shoot at all sorts of ISO, sometimes at incredibly high numbers for a film user. Working to get rid of that old habit has paid off. It broadened my scope."

Comments

[GKFroehlich]

Were you shooting in incandescent light? If so, the limited spectrum was mostly yellow and red anyway, so blocking blue and green light (which wasn’t present) would have had little to no effect.

[Sroyon]

Did you try other filters too, like red for example? I've never used a monochrome sensor and nor do I plan to, but this is very intriguing...

[brian]

It's great that you are having a good time with your new camera. The old grey cells, and in my case they are old, certainly get stimulated by new gear.

Some people could easily say it's GAS; I suspect it's much more than that in your case. from what I have seen on my large screen, the images appear to have a quality that is hard to describe (one knows it when one sees it) about them.

I too gave a new camera. I sold my Sony RX100vii with which I never really bonded—I suspect because of the haptics of using it—for a nearly as small Ricoh GR III.

Took it out yesterday after a trying time setting it up for street work. It performs like a charm.

I need to perfect my technique, though. Funnily enough, I think I can see a different but nicer IQ from this camera. The larger sensor and more megapixels may have something to do with it.

Now I have to set it up for general use.

[Soeren Engelbrecht]

Very interesting/surprising that the colour filters don't have a significant effect on the fp(m). At least Leica recommends them for the Monochrom: https://leica-camera.com/en-MY/photography/accessories/filters/e39-filter-yellow-black. And the effect is visible to the naked eye, too. Have you tried a red or orange one ?

[Jeff]

Colored lens filters do make a difference on the Leica Monochrom(s), at least on the M9/M10 platforms.

[Grant]

When we view photographs we all tend to see things differently. I agree that while the grain is noticeable in the ISO 3200 shot, it is indeed a very usable photograph.

Now I also notice that the man who preaches healthy, plant based eating, uses salt in his kitchen. Have you not read of the horrors of salt?

Like I say, we all see things differently. :)

[The larger shaker is Morton Lite Salt, which is 50% salt and 50% potassium chloride. Dr. Greger recommends it as most people get too much salt and too little potassium, so Lite Salt helps with both. There are just a few things that don't taste right with the slightly metallic taste of potassium chloride, which is why I keep the little shaker of salt around. I don't use it much though. --Mike]

[robert e]

Congratulations on your new "m" camera, Mike!

Interesting how the ISO 3200 image seems to show more detail, especially at reduced size. I assume that's actually acutance, enhanced by the little bit of noise texture in the higher ISO image, in effect acting like a sharpening filter.

Curious why you expected the scraped sensor to have the same panchromatic sensitivity as BW film. That would strike me as an odd coincidence.

I've been using a new-to-me old camera (X-E2) for a few weeks and realizing that I'm going to have to put in the time and work to get familiar with it enough to have any real fun with it. That part of the process used to be fun, too. I must be getting lazier.

[Tristan]

Apologies if I have missed this in a previous post, but assuming the monochrome conversion is a hardware conversion, removing the bayer filter and leaving the firmware unchanged. The camera would still have white balance settings, which if left in auto would lead the cameras software to "correct" the effect of the filter, effectively cancelling it out, because the camera still thinks it is taking a colour image.

[Joseph L Kashi]

I'm really glad that you are exploring the BW Sigma conversion. This is one of the more refreshing ser9es of photo articles that I've seen on the 'Net within recent memory.

In fact, you are greatly whetting my appetite for a Sigma camera, whether color or monochrome.

Thanks!

[Jay Burleson]

"Oh, and I discovered that colored filters on the lens make virtually no difference."

That would be a deal changer. Like you, I fail to understand why that sensor wouldn't react to filtration.
Every one of the Leica Monochrome cameras I have (M9M, M10M & Q2M) responds more or less* like B&W film when using filters.

*The amount of tonal change with a digital sensor compared to film seems to be slightly different depending on the color temperature of the illumination, in my limited testing.

[Peter, in Boulder]

The insesitivity to color filters is a surprise Mike. I would be interested in seeing some comparison shots of a scene with some contrasting primary colors. I’ll bet the braintrust here has a lot of theories on why this is the case.

[Speed]

A saying I heard many years ago ... This camera has 100 features. Nobody uses more than three, but everybody uses a different three.

Applies to cameras, phones, software and now automobiles.

Related: A friend and I were returning from a morning shooting a car show. "Lets stop at Starbucks for coffee," he said. "Where would that be?" I replied.

He tapped "FIND, STARBUCKS" into the car's map. And there it was -- 1.78 miles away.

I don't know which it more amazing, the car's map or that Starbucks is never more than a couple of miles away wherever you are.

[They say you're never more than ten feet from a spider, even on a jetliner. Maybe something similar could be said of Starbucks? Although a Starbucks is yet another thing my little town doesn't have. I've also heard that the average Starbucks order contains more sugar than the average dessert, so it would be more accurate to think of it as a candy shop! --Mike]

[Luke]

We are interested in your explorations of the fp(m) BECAUSE it's a quirky, offbeat camera.

And, yeah, the salt shaker surprised me, too.

[Jeff1000]

There’s a product called “No Salt,” which is 100% potassium chloride. Most people get plenty of salt without consciously adding it. The ratio of potassium to salt should be around 4 to 1.

[The trouble is, as Dr. Greger points out, that PC tastes better on (or in) some things than others. 50% PC seemed like a good compromise. I don't use much salt anyway, and always check the ingredients list and stay away from packaged foods with too much sodium. --Mike]

[Dave]

Mike,

The comment by Tristan about the auto white balance would seem to make sense. I would guess you had manually set the white balance for the environment, but just in case . . .

The Quick Set Menu sounds like the Pentax Smart Function dial on my K1 II. A handy feature and I found that the Pentax also saves an ISO setting, even when you move the Smart Function dial to the "dot" setting (which prevents changes if the setting dial is moved accidentally -- although the stiffness of the dial would make that unlikely.)

After you play around with the new camera, you'll have to read the manual to find all the hidden gems that camera includes.

[Bob G.]

Mike,
Always interesting. Keep it going. Without the experimenters there wouldn’t be any experimenting! And oblique thinking, insightful alternatives, adventurous escapes, etc…

[William Barnett-Lewis]

Count yourself lucky to live away from a Staryucks and their burned coffee.

A nice cuppa tea is a much better option 😉

[Alex Mercado]

Photo filters are designed to affect panchromatic film and sensors. Your fp(m) is no longer panchromatic, as the color filter array has been removed.

To put it simply, the sensor is color blind. Colored filters will make virtually no difference.

[Jeff]

Alex Mercado wrote: “To put it simply, the sensor is color blind. Colored filters will make virtually no difference.”

This is clearly not the case with the Leica Monochrom models, which have no color array. The tonal response is nevertheless altered. Mike can verify this with his loaner Monochrom.

FYI, one could also make a color picture from a Monochrom camera using red/green/blue lens filters, in series, or via neural filters in Photoshop. Not perfectly, but it can be done, as various articles/videos demonstrate.

[Henning]

Alex M. - The fp(m) is still panchromatic. Panchromatic refers to spectral sensitivity, not the ability to produce colours.

On the other hand, I too find the Leica Monochrom cameras respond to filters; not the same way as any specific film, but definitely responding. From my experience the the Leica Monochrom sensors, and possibly the converted Sigma sensor, are actually closer in luminance sensitivity to our eyes. We need to use a yellow filter with most B&W films to restore their sensitivity to something that approaches our visual sensitivity, whereas the digital sensors don't seem to need that as much. However, when using different or stronger filters, such as orange, red, green or blue, a definite effect is visible on the Monochrom files.

[Dave]

Mike,

Maybe you should borrow a Harris shutter and do some experiments!

Did you use a yellow filter (and other colors) on the Leica and notice a stronger effect? Maybe some examples would help us to see in exact detail what is going on.

[Rob Campbell]

Salt in the diet.

About twenty years ago, at the time of my first heart attack, I mentioned to my cardio that I supposed I’d now have to eschew the use of salt. He asked me why, seeing that I did not suffer from hypertension.

Perhaps we make too much of this salt thing?

[Alex Mercado]

“The fp(m) is still panchromatic. Panchromatic refers to spectral sensitivity, not the ability to produce colours.”—Henning

That is not completely correct. Panchromatic means it is sensitive to all wavelengths of visible light. Without the CFA, the individual photosensors are only detecting light intensity. The CFA, in tandem with the system’s demosaicing algorithm, is what establishes wavelength specificity and the ability to separate color information. The fp(m) is sensitive to light, but it has no means of determining what wavelength the light is; all it can do it determine which photosensor is absorbing more light relative to the other photosensors.

In contrast, Leica stresses their monochrome cameras have no filter array at all. This implies the photosensors are likely tuned specifically for visible light natively; which makes their sensors truly panchromatic, and explains why they react to ordinary photo filters, whereas the fp(m) does not.

[Brian]

maybe when you have time, this may be an interesting experiment

https://petapixel.com/2015/11/23/how-to-make-color-photos-using-only-bw-shots/