Ko.Fe.
Lenses 35/21 Gears 46/20
I'm not sure why Monochrom is nessesesary. I've had and have some cameras with so-so bw SOOC and some with great SOOC BW, but none of them are Monochrom. I was looking at Fuji 16MP BW and it was good, IMO, without Monochrom version.
rogue_designer
Reciprocity Failure
A better question is why Leica thought they should. :/
Dogman
Mentor
This......................^^^^^^^^^
ferider
Mentor
And I would believe them. The Leica Monochromes only make sense for sensors with low max. ISO. Fuji seems to be engineering, not marketing driven.
While I don't think what I said was nonsense, you do bring up a good point.
jarski
Mentor
They could introduce fixed lens model with different coatings for mono-sensor. Like Voigtländer SC vs MC lenses.
Am not so worried abt niche argument. Different niches are where camera makers make money these days. Pro market like it used to be have almost disappeared, and mass market gone phone cameras. All that's left is enthusiast market with appetite for niche of their preference.
Am not so worried abt niche argument. Different niches are where camera makers make money these days. Pro market like it used to be have almost disappeared, and mass market gone phone cameras. All that's left is enthusiast market with appetite for niche of their preference.
airfrogusmc
Mentor
if they said it it isn't true.
http://www.red.com/learn/red-101/color-monochrome-camera-sensors
http://www.azooptics.com/Article.aspx?ArticleID=1122
Keith
The best camera is one that still works!
And you're expecting to get a free Fuji Monochrome if they make such a thing?
On a more serious note Joe any Fuji stuff I see in the digital black and white thread always impresses me but like you I would like to see it happen ... a dedicated camera. I lusted after a Leica MM for two years and turned myself inside out over the price but never did it thankfully and once the Nik software became a free download from google I tried Silver Effex and really liked it. That effectively killed off the remaining Monochrom lust.
A monochrom XPRO2 would interest me depending on price.
willie_901
Mentor
In my view the Monochrome 262 is the very best platform for B&W work. But I am unable to see how maximum ISO differences are relevant to B&W rendering.
Maximum ISO for the Monochrome is 256,000 which is only one stop lower than the X-Pro 2's maximum of 512,000.
ISO increases are achieved using in-camera electronic and, or digital multiplication.
When one does what the meter suggests, increasing the ISO parameter always decreases sensor exposure when the shutter is open compared to the maximum possible exposure (which would be at base ISO). As ISO increases the decrease in signal (due to under exposure) is more significant than the increase in noise.
At maximum ISO all contemporary cameras resort to digital multiplication of the of the raw file numbers (data numbers) to increase brightness. At the highest ISO values all brands automatically apply some type of high-pass filtering in-camera to average noise (even for raw files).
In my opinion the Monochrome 246 has the best B&W image rendering because it is engineered for monochrome rendering. When one considers Leica's optical engineering expertise, manufacturing standards and use of high quality materials in to the mix, the Monochrome 246 would be difficult to beat.
back alley
IMAGES
And you're expecting to get a free Fuji Monochrome if they make such a thing?
On a more serious note Joe any Fuji stuff I see in the digital black and white thread always impresses me but like you I would like to see it happen ... a dedicated camera. I lusted after a Leica MM for two years and turned myself inside out over the price but never did it thankfully and once the Nik software became a free download from google I tried Silver Effex and really liked it. That effectively killed off the remaining Monochrom lust.
A monochrom XPRO2 would interest me depending on price.
not expecting anything for free
...but buying a new camera for some software upgrade (acros) is hard to reason for me...I have done the same thing with my X100S and X-E1, Tony, but still shoot only in RAW. I only print and display mono images so the VF gives me an idea of what is possible. I would love to at least try a Monochrom but even renting is more than I want to pay. Maybe I will win the lottery some day....
JonB
Member
Re: UV/IR model. Making that change is relatively straightforward; companies like Kolari Vision offer it as an aftermarket service. In contrast, the color filter array seems to be more tightly integrated into the sensor itself. So, a monochrome camera requires spec'ing a whole new sensor rather than just slightly modifying the manufacturing process.
Re: benefits of monochrome sensor. To me, it comes down to the character of the noise, which is always an issue at high ISO but can also be an issue at low ISO with significant processing (such as is almost always the case with high dynamic range scenes, aka bright daylight). Let's simplify things and assume that "noise" means that an arbitrary sensel incorrectly reports a value of 255 rather than the tone of the actual scene. When the image is derived from interpolated values from neighboring sensels, this means that at least three other pixels are affected in a Bayer filter, with more than that likely in an X-Trans filter, but no other pixels are affected in a monochrome camera. At this level, then, a 24 megapixel Bayer camera could be meaningfully compared to a 6 megapixel monochrome camera. For some applications, this is not a meaningful advantage; for others, it makes all the difference.
In brief: monochrom sensors yield images that have finer details and smoother fields than color sensors of comparable technology and at the same image resolution and pixel pitch.
Why hasn't Fuji offered a monochrome camera? Well, I'd guess it is because the trade-offs are difficult to understand, and forcing consumers to choose between products without clear differentiators is a recipe for marketing failure. Offering a monochrome camera is tantamount to admitting that the color camera's B&W images are inferior, which they are even if they remain excellent, but which never sounds good to a prospective customer. Moreover, a monochrome camera sacrifices the possibility of correcting lateral chromatic aberrations in software, not to mention emulating the color responses of different filmstocks. So there are engineering, finance, marketing, and customer satisfaction concerns.
Leica and Phase can get away with it because they are already far enough up the ladder of diminishing returns that customers are expected to understand the hardships they'll endure to achieve the other benefits of those systems. Fuji is far more mass-market and therefore needs to be more pragmatic in its designs -- although both the dwindling/upmarket-trending dedicated camera market and the GFX system might open doors here.
At least, that's my $0.02.
Cheers,
Jon
Re: benefits of monochrome sensor. To me, it comes down to the character of the noise, which is always an issue at high ISO but can also be an issue at low ISO with significant processing (such as is almost always the case with high dynamic range scenes, aka bright daylight). Let's simplify things and assume that "noise" means that an arbitrary sensel incorrectly reports a value of 255 rather than the tone of the actual scene. When the image is derived from interpolated values from neighboring sensels, this means that at least three other pixels are affected in a Bayer filter, with more than that likely in an X-Trans filter, but no other pixels are affected in a monochrome camera. At this level, then, a 24 megapixel Bayer camera could be meaningfully compared to a 6 megapixel monochrome camera. For some applications, this is not a meaningful advantage; for others, it makes all the difference.
In brief: monochrom sensors yield images that have finer details and smoother fields than color sensors of comparable technology and at the same image resolution and pixel pitch.
Why hasn't Fuji offered a monochrome camera? Well, I'd guess it is because the trade-offs are difficult to understand, and forcing consumers to choose between products without clear differentiators is a recipe for marketing failure. Offering a monochrome camera is tantamount to admitting that the color camera's B&W images are inferior, which they are even if they remain excellent, but which never sounds good to a prospective customer. Moreover, a monochrome camera sacrifices the possibility of correcting lateral chromatic aberrations in software, not to mention emulating the color responses of different filmstocks. So there are engineering, finance, marketing, and customer satisfaction concerns.
Leica and Phase can get away with it because they are already far enough up the ladder of diminishing returns that customers are expected to understand the hardships they'll endure to achieve the other benefits of those systems. Fuji is far more mass-market and therefore needs to be more pragmatic in its designs -- although both the dwindling/upmarket-trending dedicated camera market and the GFX system might open doors here.
At least, that's my $0.02.
Cheers,
Jon
shawn
Mentor
not expecting anything for free
It isn't just software though, ACROS needs the faster processor in the newer cameras to work. The way it handles noise is totally different and apparently unable to run on the earlier processors. That is also why ACROS in post isn't the same thing as ACROS in camera. It is more than just a tone curve.
More about ACROS:
http://fujifilm-x.com/x-stories/the-newest-film-simulation-acros/
https://jonasraskphotography.com/2016/02/22/acros-again/
Shawn
BlackXList
Well-known
Leica weren't using an XTrans sensor though.
I assume if they've said that they've investigated it, then they've probably given it a try
xenohip
Established
Regarding highlight recovery: how about a monochrome camera with a checkerboard pattern of ND (full spectrum) filters, let's say ND2 on 1/2 the pixels? The RAW processing would be easy, almost trivial. The resolution, noise characteristics, and most importantly the dynamic range would be dramatically better than any color sensor converted to monochrome.
The lack of imagination, ingenuity, innovation is astonishing. The manufacturers keep building 1000's of essentially the same camera.
The technology would be more likely to appear in a cell phone camera than an enthusiast camera the way the market has been heading (and probably explains the sales and usage figures).
From what I can tell Fuji didn't even bother to make a dedicated monochrome film when they released "monochrome" Instax-- just left out the color from the re-hashed regular film.
The lack of imagination, ingenuity, innovation is astonishing. The manufacturers keep building 1000's of essentially the same camera.
The technology would be more likely to appear in a cell phone camera than an enthusiast camera the way the market has been heading (and probably explains the sales and usage figures).
From what I can tell Fuji didn't even bother to make a dedicated monochrome film when they released "monochrome" Instax-- just left out the color from the re-hashed regular film.
shawn
Mentor
I think you can so something similar with the Sigma Merrill sensor. Top layer if full spectrum, and converts to very high quality monochrome in full resolution. Next two layers are at a lower exposure (about a stop each) and can be used as part of a monochrome conversion. I believe this is essentially what Iridient Developer does with the Foveon Monochrome XDR raw process option.
The other option for this would be a monochrome version of Fuji's older SuperCCD or EXR sensors. With the last versions of the EXR (with same sized pixels) you could choose between double the resolution or greater DR.
Shawn
willie_901
Mentor
This is an excellent definition of noise: the difference between true, but unknown, value for a parameter estimate. In this case the parameter estimate is a number in the raw file representing the total charge accumulated by a sensel.
ferider
Mentor
Technically/Theoretically you are right, sure.
Compared to the same sensor with thin de-mosaic filter, a monochrome sensor theoretically gives you (1) about 20% more resolution and (2) ~2 stops more iso. I remember some early 262 tests that systematically compared noise and resolution to the 240, and showed that in practice, the 262 makes you gain a single stop to get to comparable noise and resolution at high ISO (>= 1600 ASA).
Both (1) and (2) were more useful with the "Ur Monochrome" due to the sensor limitations of the M9. Now, with >= 24 MPixel, and > 100k ISO, yes there will be improvement, and the owner will get bragging rights on places like RFF, but how useful is it really in practice compared to what you loose ? I'm not talking dollars, but the ability with a color sensor to change filtering in post or generate filtered in-camera B+W JPG files (as I do both with my 240 and X100s). Of course making that trade-off is an individual decision, but I wouldn't. The Fuji marketing department seems to agree with me.
Roland.
willie_901
Mentor
Statistical analyses of unrendered raw file data shows the 262's signal-to-noise ratio is about a stop better at ISO 800 and above.[1] It also shows below ISO 800 the 262's SNR is about a stop lower than the Type 240 .
Here's similar data comparing total noise levels of the two sensors before ISO amplification or digitization.
If the absence of a color-filter array was solely responsible for signal-to-noise differences, the M262 would be better at base ISO as well.
A camera's maximum possible ISO parameter setting means very little. When ISO is set that high, brands mathematically filter raw data in-camera to average noise. In my view maximum ISO is a basically a marketing tool.
What counts are the analog dynamic range as base ISO and the relative shadow region image quality at ISOs of 800 to 1600.
The raw file signal-to-noise ratio determines the maximum dynamic range.
At ISO 1600, a shadow region with a three stops under exposure compared to the brightest highlight region(s) in the frame would require an ISO equivalent to 128,000 to be rendered with the same brightness.
The signal-to-noise ratios for the M240 and MM262 are excellent and similar.
[1] Dynamic range is directly proportional to the datas' signal-to-noise ratio.
Dogman
Mentor
You guys lost me at the technical stuff. I'm more interested in what the picture looks like than how the camera works. You can get really bogged down in this minutiae. Most people, most of the time are really fine with the mono images they get from their regular sensors.