Home · Register · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Leica & Alternative Gear | Join Upload & Sell
1      
2
       end
   

Leica M11 Monochrome Dynamic Range
  
 
highdesertmesa
Offline
• • • •
Upload & Sell: On
p.2 #1 · p.2 #1 · Leica M11 Monochrome Dynamic Range




RustyBug wrote:
Yes, they all can be fooled ... thus the need to think your strategy to outfool the foolable.


Have you tried the Zf in highlight weighted mode? Feel like I’m listening to jazz improv or something here.
Apr 17, 2024 at 02:25 PM
Fred Miranda
Offline
Admin
Upload & Sell: On
p.2 #2 · p.2 #2 · Leica M11 Monochrome Dynamic Range


RustyBug wrote:
Highlight weighted metering with a -1 EC ... Voila, you're there.

Underexposing all your images by -1 stop significantly reduces the benefits of shooting in monochrome for improving shadow detail. When you underexpose, you're already metering for the highlights, which can result in noisier shadows. Decreasing the exposure further exacerbates this issue.

I've noticed that in the Leica M11 dpreview review, they sometimes underexpose their high-contrast night shots by 3, 4 or more stops to preserve highlight detail. Other reviewers use similar techniques to prevent blown highlights in very high contrast scenes. I think this seem extreme and impractical. In such situations, using a Bayer sensor and converting the image to black and white in post might be a better option. This method could enhance shadow detail by retaining accessible highlight information in specific color channels, reducing the need for extreme underexposure and preventing complete loss of highlight data.
Apr 17, 2024 at 04:04 PM
FrozenInTime
Offline
•
Upload & Sell: Off
p.2 #3 · p.2 #3 · Leica M11 Monochrome Dynamic Range


Fred Miranda wrote:
Underexposing all your images by -1 stop significantly reduces the benefits of shooting in monochrome for improving shadow detail. When you underexpose, you're already metering for the highlights, which can result in noisier shadows. Decreasing the exposure further exacerbates this issue.

I've noticed that in the Leica M11 dpreview review, they sometimes underexpose their high-contrast night shots by 3, 4 or more stops to preserve highlight detail. Other reviewers use similar techniques to prevent blown highlights in very high contrast scenes. I think this seem extreme and impractical. In such situations, using a Bayer sensor and converting the image to...Show more →

I've often pondered if a clear+ND overlay, similar to a bayer filter, would be a practical solution.
It could yield increased highlight dynamic range, with full resolution in the mid-tones at the minor cost of lowered resolution in the highlights and shadows.
Apr 17, 2024 at 05:30 PM
RustyBug
Offline
• • • • • • •
Upload & Sell: On
p.2 #4 · p.2 #4 · Leica M11 Monochrome Dynamic Range


Zone system ... land it where you want it.


Apr 17, 2024 at 09:30 PM
RustyBug
Offline
• • • • • • •
Upload & Sell: On
p.2 #5 · p.2 #5 · Leica M11 Monochrome Dynamic Range


highdesertmesa wrote:
Have you tried the Zf in highlight weighted mode? Feel like I’m listening to jazz improv or something here.

No Nikon's over my way.

I just understand how reflective metering works vs. incident light levels vs. direct light source.
Apr 17, 2024 at 09:47 PM
retrofocus
Offline
• • • • • •
Upload & Sell: Off
p.2 #6 · p.2 #6 · Leica M11 Monochrome Dynamic Range


Fred Miranda wrote:
Underexposing all your images by -1 stop significantly reduces the benefits of shooting in monochrome for improving shadow detail. When you underexpose, you're already metering for the highlights, which can result in noisier shadows. Decreasing the exposure further exacerbates this issue.



Glad someone is stating this - this resembles my own experience with the MM 246! When I first got this camera, I followed what most advised online - underexpose 2/3 to 1 stop to compensate for the higher sensitive sensor without Bayer array. I found the shadows too contrast-rich and underexposing being more of a debit than credit. I removed the applied setting in my camera to +/- 0 like it is on my standard M-E 240 camera, and then the photos turned out all to my pleasing. In specific situations I still underexpose with the MM 246, but it is more rare than common.


Apr 18, 2024 at 07:19 AM
RustyBug
Offline
• • • • • • •
Upload & Sell: On
p.2 #7 · p.2 #7 · Leica M11 Monochrome Dynamic Range


-1 was never meant as a means to improve shadow detail ... it is suggested as a method to accompanying reflecting metering for achieving an even more aggressive highlight protection, that had been being requested. If improving shadow detail is your mission, yeah more light. If aggressively protecting highlights, less light. Nothing new here.

That's a different matter from how to get an exposure that optimizes shadow detail. There's no such thing as a free lunch, and it cuts both ways. The person has to know what it is they're trying to achieve, and how / where they want to land it.

Sometimes, we have to make a (conscientious and controlled) choice ... in that we can't always have it both ways.

It's kinda hard to get a suntan when you're wearing a raincoat. And, if you go out in the rain with a swimming suit on, you're gonna get wet.

But, if someone wants to wear a raincoat all the time ... that's their prerogative. Just know that you'll never get much of a tan.
Apr 18, 2024 at 08:00 AM
Fred Miranda
Offline
Admin
Upload & Sell: On
p.2 #8 · p.2 #8 · Leica M11 Monochrome Dynamic Range


RustyBug wrote:
-1 was never meant as a means to improve shadow detail ... it is suggested as a method to accompanying reflecting metering for achieving an even more aggressive highlight protection, that had been being requested. If improving shadow detail is your mission, yeah more light. If aggressively protecting highlights, less light. Nothing new here.

That's a different matter from how to get an exposure that optimizes shadow detail. There's no such thing as a free lunch, and it cuts both ways. The person has to know what it is they're trying to achieve, and how / where they want to...Show more →

Exposure compensation isn't something you'd normally use all the time. It's intended to compensate exposure when your camera's metering misjudges very bright or dark subjects within the scene, as the metering is calibrated to expose to neutral gray.

Lowering exposure by -1EV across all your photos to prevent overexposure can indeed help preserve highlight detail. However, this approach may also reduce the DR and increase noise in the midtones and shadows of your images. Therefore, it's best used sparingly and not as a general practice that limits the camera's capabilities. Basically, when shooting with monochrome camera without Bayer filtering, more light can reach the darker areas of your photo. But if you reduce exposure compensation too much, you lose this advantage.

With Bayer filtering sensors, the camera captures highlight data across different color channels, particularly red and green, which allows for partial recovery of details in these areas. In contrast, monochrome sensors lack color data -- if it's not there, it's gone!. Perhaps due to this limitation, depending on scene contrast, b&w imagess from a monochrome sensor may have a distinctive look that many photographers appreciate. However, others still prefer shooting b&w film for its classic aesthetic, including grain, and its flexibility in preserving highlight detail even when overexposing a scene to capture more detailed shadows.
Apr 18, 2024 at 10:03 AM
rscheffler
Offline
• • • • • •
Upload & Sell: Off
p.2 #9 · p.2 #9 · Leica M11 Monochrome Dynamic Range


highdesertmesa wrote:
Random thought of the day: the unedited DNGs from the monochrome sensors remind me of Ilford XP2 printed at a 1-hour lab back in the day. XP2 was/is a C41 process color film that you could have developed at any color lab to produce b&w negatives. I always liked the smooth tones and lower contrast of that film and how it looked when printed to color paper from the lab. It was also great for darkroom printing to high contrast gloss b&w paper.

Fred Miranda wrote:
Perhaps because the XP2 is known for its wide dynamic range and smooth tonality. It's convenient that the XP2 can be developed using the C41 process.

One of the distinctions between using XP2 film vs a monochrome digital camera is the level of latitude in the highlight area.

It's interesting that you mentioned this, as I read an article a while back that discusses this topic:

"Scanned XP2, especially when dust reduction or grain suppression software is applied, can look somewhat 'digital', i.e. very clean, almost 'waxy'. The behaviour of the grain mentioned above plays a part here too. If you dig...Show more →
highdesertmesa wrote:
Good point. I remember it being nearly impossible to ruin an XP2 exposure even with the cheap point-and-shoot film cameras of the time.



Remember a rule of thumb for shooting B&W negative films: expose for the shadows and develop for the highlights. XP2's C41 process of course limits development flexibility, but the replacement of the silver grains by dye clouds, which when overexposed blend together more, and which the latitude of XP2 and generally all C41 films tolerate, pushes it towards that grainless pseudo-digital look...


highdesertmesa wrote:
I wish Leica would offer a more aggressive and more accurate version of Highlight Weighted metering. For example, Nikon's highlight weighted meter is extremely aggressive on the Zf, but it also is not easily fooled into underexposing by too much. And in combination with that, Leica needs to add a robust dynamic range tool like Nikon, Fujifilm and others have. Fujifilm calls it D Range Priority with settings of Auto, Strong, Weak, or Off. That would go a long way to making the Leica monochrome exposures looking better on the LCD and for the JPEGs. As it stands now, many...Show more →
RustyBug wrote:
Highlight weighted metering with a -1 EC ... Voila, you're there.

Way simpler than a bunch of menu changes / mode changes / program adjustments for diff scenarios.

That's how I've got mine set. Depending on scene contrast, I may adjust my EC if it is inherently low DR lighting / scene.
Set EC to taste for how much you want to offset things.

highdesertmesa wrote:
"Yeah, but"...

The Leica implementation is easily fooled, so I have to futz with the EC regardless of where I set it by default. Not much of a problem with an EVF camera, but it's a pain when using the rangefinder.

Nikon's highlight weighted metering is no more complicated than Leica's. If you're referring the DR function as a complication, it's something separate completely. I would find such a feature useful simply for giving a better preview on the LCD. It's not needed to achieve the correct exposure.

RustyBug wrote:
Yes, they all can be fooled ... thus the need to think your strategy to outfool the foolable.


It's funny with decades of advancements that metering still often sucks.

I feel like the only way to solve this problem is something like timing how long it takes each pixel to reach near saturation for a given image/exposure and use that to reverse engineer/reconstruct the desired image with a tone curve applied to the time differences for saturation. Fastest to near saturation would be near-white values along the curve and longest to near saturation would be near-black. Perhaps this would give a higher signal-noise ratio for mid/shadow values kind of like the old days of exposing film for the shadows and developing for the highlights. Or ETTR in the digital era but with the ability to selectively pull pixel values back from the right based on time to saturation.
Apr 18, 2024 at 02:09 PM
 


Search in Used Dept. 

RustyBug
Offline
• • • • • • •
Upload & Sell: On
p.2 #10 · p.2 #10 · Leica M11 Monochrome Dynamic Range


rscheffler wrote:
It's funny with decades of advancements that metering still often sucks.


Reflective metering is still reflective metering ... and all the foibles associated with the assumptive and presumptive algorithm's.

Incident metering is actually measuring the amount of light falling on / illuminating the subject / area. Incident metering is a measurement.

Reflective metering is an assumptive / presumptive "guess" that your scene best fits a variety of different models ... as it pertains to the calibration of the camera meter. So, depending on which reflective metering mode you choose, you may still have to "out think" the reflective metering algorithm when it doesn't fit the model of the mode.

Routine stuff that fits the model / mode ... golden gravy. Extreme situations or "mismatches" to your model / mode ... you gotta use your noggin'.



Apr 19, 2024 at 09:34 PM
highdesertmesa
Offline
• • • •
Upload & Sell: On
p.2 #11 · p.2 #11 · Leica M11 Monochrome Dynamic Range


RustyBug wrote:
No Nikon's over my way.

I just understand how reflective metering works vs. incident light levels vs. direct light source.

Which has nothing to do with the point I was making about Nikon's highlight weighted metering. It literally just keeps the highlights from clipping anywhere in the image even when bright highlights only occupy a very small portion of the image (but not so small as to be confused by specular highlights). Conversely, Leica's version of highlight weighted metering mode is more like matrix/evaluative metering with merely a bias toward highlights. It's not a metering mode that preserves the highlights at any cost.

So what I'm saying is a bit opposite of what you might think. I actually found Nikon's highlight weighted metering to be not very useful with a color sensor because often I only want to protect one or two color channels so the shadows have better integrity. But it made me think about how great Nikon's implementation of highlight weighted metering would be for a monochrome sensor because if the highlights clip at all on monochrome, they're gone forever.
Apr 20, 2024 at 12:16 AM
RustyBug
Offline
• • • • • • •
Upload & Sell: On
p.2 #12 · p.2 #12 · Leica M11 Monochrome Dynamic Range


highdesertmesa wrote:
Which has nothing to do with the point I was making about Nikon's highlight weighted metering. It literally just keeps the highlights from clipping anywhere in the image even when bright highlights only occupy a very small portion of the image (but not so small as to be confused by specular highlights). Conversely, Leica's version of highlight weighted metering mode is more like matrix/evaluative metering with merely a bias toward highlights. It's not a metering mode that preserves the highlights at any cost.

So what I'm saying is a bit opposite of what you might think. I actually found Nikon's highlight...Show more →

Gotcha.

That's kinda why I prefer regular old average metering that takes the high / low and splits the diff. While that might not land it quite where you want it (vs. say spot metering), it is easily "in range" of most pp capability. The combination of that +/- EC to your preference makes it an easy "safeguard" to split the diff, then an easy "adjust" to your knowledge of the scene ... and / or you knowledge of your desired preference for highlight retention vs. shadow detail.

As to the Nikon's sensitivity to where it "draws the line" ... I guess I'm of the perspective that every camera (i.e. different mfr / model / iterations) employ varying (i.e. improving it you will, yet still different) metering algorithm nuances. That's the one thing I like about the M's using the shutter blades ... consistency of expectation. But, either way ... once you study / learn / test / know your camera's reflective metering response ... from that baseline, your +/- EC can let you dial it in to your preference(s).

I guess I'm of the opinion that there's never a need to complain about a camera's reflective metering ... they all have their foibles. It is inherent to the variables in play with the reflective metering process. If we choose to use reflective metering, then we should also choose to learn it to that specific camera. I have ZERO expectation that all mfr's approach this the same for all models. Some models have built in "headroom", others let you have it all without a safety net.

My Kodak DCS SLR/C was one pro camera that did not have a safety net vs. the contemporary cameras of the day that did. Imo, the monochrome camera is such a specific tool, that it isn't a product that the user is of the ilk to want a big safety net ... rather, they want the most exacting control. Kinda like folks going from an automatic transmission to a manual transmission ... they operate differently, and there is a learning curve to those differences.

Imo, the monochrome camera is different for a reason ... reasons to be understood / embraced ... rather than picked apart because it isn't like the BFA cameras. Learn the diff's, dial it in and harness it to your creative / personal desires.

I get that folks want everything to be "as they expect it should be", yet ... that's just not the reality of different tools made for different approaches. The mono ... it's a unique tool, and yes it is set up a bit differently wrt to its "no safety net" or headroom as other models may have.

Race cars get a different setup than consumer cars. Monochrome camera's get a different setup than BFA camera's. I accept that, and then learn how to harness the rebel / thoroughbred / mustang that it is.

Learning to ride a thoroughbred or mustang is a different matter than learning to ride a Tennessee Walker or Quarter Horse. You adapt to the nature of the horse, not try to make the horse into a different breed that it isn't.

Others will dissent.




Apr 20, 2024 at 09:08 AM
highdesertmesa
Offline
• • • •
Upload & Sell: On
p.2 #13 · p.2 #13 · Leica M11 Monochrome Dynamic Range


RustyBug wrote:
Gotcha.

That's kinda why I prefer regular old average metering that takes the high / low and splits the diff. While that might not land it quite where you want it (vs. say spot metering), it is easily "in range" of most pp capability. The combination of that +/- EC to your preference makes it an easy "safeguard" to split the diff, then an easy "adjust" to your knowledge of the scene ... and / or you knowledge of your desired preference for highlight retention vs. shadow detail.

As to the Nikon's sensitivity to where it "draws the line" ... I guess...Show more →

Center-weighted plus EC is how I normally shoot when not using highlight-weighted. I've never liked the multi-segment/matrix/evaluative metering modes. They are much too schizophrenic and sometimes require more EC than is available.
Apr 20, 2024 at 10:54 AM
RustyBug
Offline
• • • • • • •
Upload & Sell: On
p.2 #14 · p.2 #14 · Leica M11 Monochrome Dynamic Range


highdesertmesa wrote:
I've never liked the multi-segment/matrix/evaluative metering modes. They are much too schizophrenic and sometimes require more EC than is available.

Yeah, we're in agreement on that.
Center weighted average (tailored with EC) ... I'll take that when average isn't available (diff models).


Simple, reflective metering I can "out think".

Trying to out think something that I can't be sure what it was thinking is a thought that I don't wanna think about.

Sunny 16 (EV 15) works well for me, too.

Knowing that shade is -3 EV from there ... then it's just a matter of deciding if I want a given area to be approached as key, fill, rim, background or kicker exposure levels.

Other variants, but the principle of understanding the diff between key vs. fill luminance / arrangement in the lighting is something that I've found the (automated) reflective metering CANNOT decide as to knowing which it is I want, for how I want to render a scene. Today, I may want the shadows to be shadows / fill. Tomorrow, I may want the shadows to be key. For that, I have to be the one in control ... thus, the +/- from my average baseline to switch between key vs. fill.
Apr 20, 2024 at 11:16 AM
zhangyue
Offline
• • • • •
Upload & Sell: On
p.2 #15 · p.2 #15 · Leica M11 Monochrome Dynamic Range


thrice wrote:
ISO is a representation of apparent brightness, it is not a setting of the sensor.


I think this is questionable definition for digital camera. ISO is gain applied to electric signal before converting to digital domain.

thrice wrote:
As you can see the sixth gain setting has a DR jump on both cameras as the sensor switches to its other gain mode.


Dual gain setting is a way to reduce noise introduced by amplifier. It is purely in electrical domain.

thrice wrote:
There is no such thing as increased signal due to gain setting.

Again, this need to be careful. As gain definitely increase signal, however it also boost noise.

thrice wrote:
If a sensor records more photons at base iso (compared to another at base iso) it has a higher quantum efficiency and thus higher signal to noise ratio if the noise produced by both sensors is identical.
Agree

thrice wrote:
If the noise introduced by equivalent gain is less then the snr is likewise higher. ISO 125 on the M11M is the exact same gain setting on the IMX455 as ISO 64 on the M11.


Agree, RGB filter is 0.5X gain so ISO64 and 1/2 reduce of light make M11 equivalent to M11M in peak DR.

thrice wrote:
At ISO3200 (for example) both cameras will produce an image of similar brightness, but the M11M is effectively using a lower gain mode due to higher quantum efficiency from removal of the CFA.
I don't think QE is different as that is only related to pixel diode itself. However, for the same electric headroom of sensor design, and same ISO gain applied for both, M11M allow almost 1 stop more light hit on the sensor. So almost 1 stop DR advantage.

thrice wrote:
The reason the exact same gain setting on the M11M produces less DR than on the M11 is because of minor benefits to noise introduced by Bayer interpolation.

If you check P2P plot, you will find there is not a single case at the same ISO, M11M has lower DR than M11. as explained in above.

thrice wrote:
The reason the M11 is so close to the M11M at ISO200-320 is because of the coincidental overlap of the best SNR of second gain mode on the M11 with the worst of the first gain mode on the M11M. This is pretty clear on the graph.

Agree
Apr 26, 2024 at 02:24 AM
RustyBug
Offline
• • • • • • •
Upload & Sell: On
p.2 #16 · p.2 #16 · Leica M11 Monochrome Dynamic Range


thrice wrote:
There is no such thing as increased signal due to gain setting.

zhangyue wrote:
Again, this need to be careful. As gain definitely increase signal, however it also boost noise.



Semantics ... clarification.

Gain = another term for amplification.
Thus, gain does not actually increase the signal, it amplifies its value. In order to increase the signal, I can open the aperture or increase the EV intensity of the light. The amount of signal (once received / captured) remains constant. And, since gain is amplifying the signal to increase its value ... the amplification of the noise (of the S/N) occurs in concert with the amplification of the signal.

It would be nice, if we could amplify the signal, while leaving the noise unamplified, but I'm not aware of that being an option.

My perspective would be that Thrice's comment that there is "no such thing" as increased signal (due to gain setting) is correctly stated. The amount of the signal ... is what it is ... what we do it it after that, can be pushed / pulled / etc. to alter its value. But, the signal remains the fixed amount that it was captured / received. Gain just tosses a multiplier at that value ... and noise comes along for the ride as part of S/N continuum relationship.

My (unqualified, layman, aiui) .02

Feel free to convert my "untechnical" to correct engineering terminology, as needed.
Apr 26, 2024 at 07:18 AM
zhangyue
Offline
• • • • •
Upload & Sell: On
p.2 #17 · p.2 #17 · Leica M11 Monochrome Dynamic Range


Engineering term is rigid so I don’t want to agree with you.

There are many perspective to look at this.
Use an example: signal is 2 noise is 1. Gain 4. Assume gain is wide band cover all signal and noise freq. Signal is 8, noise is 4. You have the same DR, but signal (minus noise floor) increased.
Thinking about Listening aid device, noise is comparable to signal at low level and that device can boost volume(of dialog freq), you can hear the dialog better.

Another point is amplifier can be bandpass/lowpass/high pass that only gain up at certain freq which signal sit. Total system SNR improves. There are also cases with multi stage gain system, signal can be boosted and noise can be cancel out and boosted signal can convert to higher bit precision for different application.

Even use Sony camera as example. For the same input signal, shooting at ISO at 320 vs ISO400, you can see ISO 400 has better SNR. Just illustrate that is not just input signal matters but how you applied gain stage can changed system SNR.


RustyBug wrote:
Semantics ... clarification.

Gain = another term for amplification.
Thus, gain does not actually increase the signal, it amplifies its value. In order to increase the signal, I can open the aperture or increase the EV intensity of the light. The amount of signal (once received / captured) remains constant. And, since gain is amplifying the signal to increase its value ... the amplification of the noise (of the S/N) occurs in concert with the amplification of the signal.

It would be nice, if we could amplify the signal, while leaving the noise unamplified, but I'm not aware of that being an...Show more →
Apr 26, 2024 at 08:12 PM
1      
2
       end
FM Forums | Leica & Alternative Gear | Join Upload & Sell

1      
2
       end
    
 

You are not logged in. Login or Register

Username       Or Reset password



This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.