What does Q2 Monochrom clipping warning actually represent?

[blakley]

Advertisement (gone after registration)

I recently shot a new (to me) Q2 Monochrom in very contrasty light outdoors at night.  I used Highlight-weighted metering, but still got significant clipping indications in subjects' faces - even though there was lots of auto-ISO and shutter speed headroom when shooting in aperture-priority mode.  So this led me to wonder - what does the Q2 Monochrom's clipping indication actually mean?  Does it show areas which are true 255-255-255 white, or is it more like a video camera's 80% zebra - an area that's brighter than some displays can show, but not yet actually clipped?

Does anyone know this?  Is it in any Leica documentation?

Thanks, and apologies if I'm asking a question which has already been answered but evaded my search of the forums so far.

[hedleyw]

It’s not 255-255-255 as there’s no RGB information. I suspect it is 100%. With the Monochrom models highlight protection is more important, even with highlight weighted metering I have used up to one stop of negative compensation. Highlight weighted metering isn’t in itself a cure for blown highlights.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[jimtong]

I find using multi field metering better at preserving highlight compared to the new highlight weighted metering.

[DenverSteve]

First, you didn't state whether you were clipping highlights or shadows.  Similarly, no mention of the primary subject matter. If it's the faces, I would use center weighted or spot metering for the faces.  If you're clipping highlights, or shadows, you are over or underexposing that part of the scene.  A simply fill flash would likely balance the scene.  A camera can only record a limited dynamic range.  I'm sure you know this but you asked about the clipping.  

[hdmesa]

On 11/29/2022 at 7:06 PM, blakley said:

I recently shot a new (to me) Q2 Monochrom in very contrasty light outdoors at night.  I used Highlight-weighted metering, but still got significant clipping indications in subjects' faces - even though there was lots of auto-ISO and shutter speed headroom when shooting in aperture-priority mode.  So this led me to wonder - what does the Q2 Monochrom's clipping indication actually mean?  Does it show areas which are true 255-255-255 white, or is it more like a video camera's 80% zebra - an area that's brighter than some displays can show, but not yet actually clipped?

Does anyone know this?  Is it in any Leica documentation?

Thanks, and apologies if I'm asking a question which has already been answered but evaded my search of the forums so far.

Clipping means clipping. It’s not a guesstimate on monochrome.

Highlight-weighted is just that — weighted. It’s not “highlight perfect”. You still have to compensate for bright areas sometimes by reducing the exposure a little. With the monochrome cameras, this can often make the image look so dark, you can’t see the subject very well. Sometimes it’s helpful to turn off exposure preview and just trust the clipping warnings. Note that you will always clip some of the shadows in contrasts scenes if you want to preserve all highlights.

[blakley]

21 hours ago, DenverSteve said:

First, you didn't state whether you were clipping highlights or shadows.  Similarly, no mention of the primary subject matter. If it's the faces, I would use center weighted or spot metering for the faces.  If you're clipping highlights, or shadows, you are over or underexposing that part of the scene.  A simply fill flash would likely balance the scene.  A camera can only record a limited dynamic range.  I'm sure you know this but you asked about the clipping.  

Sorry for the ambiguity; what was clipped was the highlights. 

Center-weighted metering would not have produced my desired result as the scene was extremely contrasty and the subject's face took up a small portion of the frame. 

I wanted to keep the scene contrasty, so fill flash would also not have accomplished my intention. 

Spot metering off the subject's face would have given me an image underexposed by at least a stop, as my subject's face is something like zone VI 1/2, so that would have required compensation. 

What I really want is exactly what Highlight-weighted metering advertises, but apparently does not deliver - an image in which the brightest area of the frame is zone IX and still retains detail, so I can balance the rest of the tones in post, assuming that I don't care about clipping blacks (which I don't, in this case).

I've attached an image from the session which I exposed in manual mode after noting that highlight-weighted mode did not give me what I was hoping to get.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[blakley]

Advertisement (gone after registration)

9 minutes ago, hdmesa said:

Clipping means clipping. It’s not a guesstimate on monochrome.

Highlight-weighted is just that — weighted. It’s not “highlight perfect”. You still have to compensate for bright areas sometimes by reducing the exposure a little. With the monochrome cameras, this can often make the image look so dark, you can’t see the subject very well. Sometimes it’s helpful to turn off exposure preview and just trust the clipping warnings. Note that you will always clip some of the shadows in contrasts scenes if you want to preserve all highlights.

I did in fact compensate by setting exposure compensation to -1, and I *still* got clipped highlights.  As you can see from the sample photo I posted a second ago in my other reply, the image did indeed look dark - which I intended - but the subject was still perfectly visible because of the high contrast of the subject matter.

[blakley]

On 12/30/2022 at 3:58 AM, jimtong said:

I find using multi field metering better at preserving highlight compared to the new highlight weighted metering.

I'll give that a shot; thank you.

[hdmesa]

1 minute ago, blakley said:

I did in fact compensate by setting exposure compensation to -1, and I *still* got clipped highlights.  As you can see from the sample photo I posted a second ago in my other reply, the image did indeed look dark - which I intended - but the subject was still perfectly visible because of the high contrast of the subject matter.

It may take much more than -1. With a monochrome sensor, clipped highlights are not recoverable, they are simply not recorded. That’s why I said the image preview may end up super dark (looks as if it’s grossly underexposed) in order to save the highlights sometimes.

Also don’t go chasing specular highlights with any camera (sun reflecting off chrome, etc.), they are supposed to clip.

[blakley]

2 minutes ago, hdmesa said:

It may take much more than -1. With a monochrome sensor, clipped highlights are not recoverable, they are simply not recorded. That’s why I said the image preview may end up super dark (looks as if it’s grossly underexposed) in order to save the highlights sometimes.

Also don’t go chasing specular highlights with any camera (sun reflecting off chrome, etc.), they are supposed to clip.

I don't mind a dark preview; I'm very familiar and comfortable with recovering shadows in Lightroom.

I'm also fine with specular highlights being 255/255/255.  Faces, however, are not specular highlights.

[hdmesa]

3 minutes ago, blakley said:

I don't mind a dark preview; I'm very familiar and comfortable with recovering shadows in Lightroom.

I'm also fine with specular highlights being 255/255/255.  Faces, however, are not specular highlights.

Lol, yes faces are not specular highlights, and I never said that — unless your faces are made of chrome, mirrors, glass, or glossy paint.

Still stands that expecting -1 and highlight weighted to be perfect with a mono sensor is not reasonable. It’s up to you to make sure there is no clipping by applying much more than -1 when needed.

[blakley]

20 minutes ago, hdmesa said:

Lol, yes faces are not specular highlights, and I never said that — unless your faces are made of chrome, mirrors, glass, or glossy paint.

Still stands that expecting -1 and highlight weighted to be perfect with a mono sensor is not reasonable. It’s up to you to make sure there is no clipping by applying much more than -1 when needed.

If it is not reasonable to expect highlight-weighted metering mode to protect highlights against clipping, what exactly is the purpose of having highlight-weighted metering mode at all?  All the predecessor cameras already had manual exposure, spot metering, and exposure compensation.

[hdmesa]

15 minutes ago, blakley said:

If it is not reasonable to expect highlight-weighted metering mode to protect highlights against clipping, what exactly is the purpose of having highlight-weighted metering mode at all?  All the predecessor cameras already had manual exposure, spot metering, and exposure compensation.

It’s supposed to be more weighted toward highlights than other automatic metering modes. It is not highlight preservation mode. Even so, like all automatic metering, it’s subject to being fooled.

That said, I do wish we could have the option between highlight-weighted metering and a mode that preserves all highlights at any cost.

Edited December 31, 2022 by hdmesa

[blakley]

On 11/30/2022 at 3:12 AM, hedleyw said:

It’s not 255-255-255 as there’s no RGB information.

This is wrong BTW.  The Q2M DNG file contains a 14-bit (0 - 16,383) brightness value for each pixel, and that's exactly the same on the regular Q2.  The regular Q2 also has a mosaic map which guides RAW developers in decoding these brightness values into RGB colors for each pixel using information about which color filters cover which pixels on the sensor - and exactly the same thing happens on the Q2 Monochrom (except that there's no demosaicing because there are no color filters).  In Lightroom you end up with exactly the same result - a file in the RGB color space of your choice (mine's set by default to Adobe RGB), which *usually* happens to have the property that the R, G, and B values are all the same.  But the Q2M DNG files do not *always* have equal R, G, and B values, because the Q2M supports toning, and if you've got toning turned on, the R, G, and B values are decoded in a way that achieves the menu-specified color toning effect.  So Q2M DNG files are always *decoded* (but not demosaiced!) into color files for processing, even though the only colors that actually come out of the decoder (unless toning has been applied) are shades of gray.

The Q2M JPEGs are also in RGB color space, not monochrome, for the obvious reason that toning effects would not be visible if an RGB space were not used.

Edited December 31, 2022 by blakley
corrected "encoded" to "decoded" in two places

[jimtong]

Try to use multi field metering, it will have less clipping. The highlight weighted has higher clipping areas. Clipping does not mean it will blow up but try to reduce clipping by compensate lower.