Optimize In Camera Jpeg White Balance RGB Histogram For Better ETTR.

[kirkt]

All of these issues have been thoughtfully described, with the pros and cons inherent in the process, here:

https://www.fastrawviewer.com/blog/in-camera-histogram-doesn%27t-represent-exposure

https://www.rawdigger.com/howtouse/exposure-for-raw-or-for-jpegs

https://www.rawdigger.com/howtouse/beware-histogram

https://www.rawdigger.com/howtouse/lightmeter-calibration

Putting aside for a moment the idea that the histogram is influenced by the JPEG produced by the user's cameras settings, there are some assumptions that one is making when attempting to establish the camera's histogram as a useful tool for evaluating raw exposure - the most important being the white balance itself, as this will dramatically change the per-channel "exposure" (brightness) indicated by the histogram relative to the actual exposure of the sensor.  In daylight, the green channel is a good proxy for evaluating sensor saturation as red and blue will channels will receive less exposure than green in the raw data.  But consider non-daylit scenes - the prevailing lighting will affect the usefulness of the histogram as well and may completely skew the histogram away from the underlying raw exposure when white balanced, either toward green or toward the "correct" white balance.  Finally, because we want to expose, more or less, the sensor for highlight detail, we need to be careful with respect to the actual color of the critical highlights (either the inherent color of the highlight surface or the lighting falling on and reflecting off of the surface) when evaluating the per-channel clipping.

Unless one understands how the JPEG histogram relates to the raw exposure, simply setting a green WB (akin to UniWB) will only get you so far - other in-camera settings will affect where the tones are placed on the histogram such that, even if one fixes a constant exposure (aperture and SS), changes to the in-camera settings that affect the JPEG (color profiles, DR settings, etc.) will change the JPEG histogram even when sensor exposure remains constant.

I have found that I can get optimal exposure 99% of the time by metering the critical highlights (and it works better if these are close to neutral, clouds being a good example) with a spot meter and adding approximately 3 to 3.3 EV to the metered reading (usually be changing SS).  On my Sekonic meter I can enter this offset as a filter compensation value and, at the press of a button, get the compensation in SS that I need to expose for the metered highlights.

If you really want to push things to the edge, you can test your various raw converters and see how each responds to clipping in one or more channels (raw saturation) and evaluate how much extra compensation you can get away with if the highlights are neutral and the raw converter can RECONSTRUCT the partially clipped highlights from the intact raw channel data.  In some circumstances I can dial in over 4EV on my Fujifilm X-H1 with CaptureOne and reconstruct neutral highlights, for example.  I do not advise setting exposure based on your raw converter though, but it is good to know what result a combination of camera exposure plus software will give you for the times when you really need to turn exposure up to 11.

All of this is predicated upon studying the content of the raw data using Raw Digger or a similar tool.

Kirk

[faberryman]

I have found that I can get optimal exposure 99% of the time by metering the critical highlights (and it works better if these are close to neutral, clouds being a good example) with a spot meter and adding approximately 3 to 3.3 EV to the metered reading (usually be changing SS).  On my Sekonic meter I can enter this offset as a filter compensation value and, at the press of a button, get the compensation in SS that I need to expose for the metered highlights.

In the old days we would call this placing the highlights in Zone VIII and letting the shadows fall where they may. It is how you expose transparency film. The more things change, the more they stay the same. Hardly an epiphany.

[Tim Lookingbill]

Good luck in capturing the entire tonal scale of a scene with the 200D: there is just not enough sensor DR for many scenes so you make a compromise sacrifice shadow detail to capture highlight detail or vice versa or add light to the scene or use HDR.

Its a shame when threads turn out to be argumentative and others PoV are denigrated.

Two mentions here of UniWB, dismissed as irrelevant when the purpose of the OP seemed to indicate looking for ways to make the JPEG histogram more useful.
I do not have a 10 year old Pentax to make a comparison but AFAIK all DSLR offered some form of Custom WB and also probably some form of Picture mode e.g. Natural, Bright, Vibrant etc.
Any of which when implemented by the user alter the rendering of the raw WITH the obvious change to the histogram.  This does not change the exposure but merely the rendering. 

If you decide to use the camera LCD view including histogram as an indicator of raw exposure then you may be better selecting one of the seven scene modes  of the 200D, maybe Portrait or  Landscape bearing in mind that your raw converter of choice will have added its own values and results will be different e.g. Adobe to RT etc.

I know my Pentax can't capture all the tonal scale of a scene. I wrote that in my response! But thanks for writing out such a lengthy response to something I didn't say. It must've taken a lot  of effort and I appreciate it.

Now to ask again about YOUR camera and whether you've tried the jpeg RGB histogram experiment in my OP since I'm ASSUMING your camera is more modern and can take in MORE DYNAMIC RANGE THAN MY OLD PENTAX.

Let's see if I finally get one of you to answer this which was pointed out in my OP.

[digitaldog]

Let's see if I finally get one of you to answer this which was pointed out in my OP.

This question was answered here long ago:

Has anyone tried this?

[TonyW]

I know my Pentax can't capture all the tonal scale of a scene. I wrote that in my response! But thanks for writing out such a lengthy response to something I didn't say. It must've taken a lot  of effort and I appreciate it.

Sarcasm really you are already on a loser here why continue?  What do you think you did not say?

Now to ask again about YOUR camera and whether you've tried the jpeg RGB histogram experiment in my OP since I'm ASSUMING your camera is more modern and can take in MORE DYNAMIC RANGE THAN MY OLD PENTAX....

NO, I HAVE NOT TRIED NOR DO I NEED TO TRY YOUR VOODOO EXPERIMENTS. going down another damn pointless rabbit hole

[Tim Lookingbill]

Sarcasm really you are already on a loser here why continue?  What do you think you did not say?
NO, I HAVE NOT TRIED NOR DO I NEED TO TRY YOUR VOODOO EXPERIMENTS. going down another damn pointless rabbit hole

Thanks for the confirmation and reply. Very helpful.

[bjanes]

Exposure 101 for digital don’t blow out the highlights.
Again, the demonstration is to illustrate that higher ISO doesn’t necessarily equate to more noise.
Edit. Don’t blow out the highlights you the photographer in control of your exposure do not wish to blow out!

Of course, you don't want to blow out important highlights. However, if your scene dynamic range is 9 stops and the DR of your camera is only 8 stops, as it is with your camera at ISO 800 (according to Bill Claff), you can't capture the full DR of the scene. Depending on the subject, you can preserve the highlights or the shadows. However, if you had used base ISO you could have captured the full DR of the scene without clipping of highlights.

[digitaldog]

Of course, you don't want to blow out important highlights. However, if your scene dynamic range is 9 stops and the DR of your camera is only 8 stops, as it is with your camera at ISO 800 (according to Bill Claff), you can't capture the full DR of the scene. Depending on the subject, you can preserve the highlights or the shadows. However, if you had used base ISO you could have captured the full DR of the scene without clipping of highlights.

I don't think anyone, certainly not I dispute this. The entire discussion about ISO 800 (higher ISO) is to dismiss the OP's comments and a general concept (also mentioned) that higher ISO equates to higher noise. For a lot of cameras, that isn't the case. There were also those useful comments from the 2011 post about where a higher ISO with less noise can be useful. Certainly one should take DR into account in such cases.This is a technique that isn't going to work for everyone in every situation. But it can be useful in some cases and more importantly dismisses the wrong idea about high ISO and noise. It's also important to understand your comments about DR and blowing out highlights; photographers simply have to understand how their specific tools work (and why). Some here clearly don't. So yeah, DR, ISO, what meters assume how differing meter technology operate (Reflective vs. Incident), and how to avoid clipping in raw are all important to understand for photographers who wish to obtain the highest quality captures. And far more important IMHO than attempting silly hacks with JPEG Histograms for raw shooters.
I suppose we could ask, why do we still not have a raw camera Histogram (with exception of a firmware 'hack' for Canon)?
Much of the above was true for film.

[TonyW]

Thanks for the confirmation and reply. Very helpful.

So pleased to hear that you found this very helpful!

[kirkt]

In the old days we would call this placing the highlights in Zone VIII and letting the shadows fall where they may. It is how you expose transparency film. The more things change, the more they stay the same. Hardly an epiphany.

No epiphany, but a simple reality of exposing for a digital sensor.

Computers aren’t an epiphany either, just a bunch of switches controlling the flow of electrons.

Kirk

[DP]

It seems to me that that thread, and hence this one, is misleading on the issue of noise without some clarification. Yes, noise will be lower (but by how much differs from camera to camera) if correctly exposed at a higher ISO than underexposed (with the same shutter speed and aperture) at a lower ISO.

I'd say this way: with the equal exposures S/N in deep shadows can be increased (when/if possible - for example w/o clipping important highlights) by selecting a certain higher nominal ISO values (before shooting !) _BUT_ more exposure when / if possible is always better than using higher gains (higher nominal ISOs) ...

[Garnick]

"BUT_ more exposure when / if possible is always better than using higher gains (higher nominal ISOs)"

I agree absolutely, but of course the caveat "if possible" always lingers in the background.  I have bracketing sequences set up for situations that I deem "possible" for using the "Optimal Exposure" routine as mentioned in Bob DiNitale's excellent article.  And, if the exposure I want to use for that shot has clipped some important area I simply stack it with the one containing that information and integrate it in the final image.  More or less akin to HDR I suppose, but it works quite well when necessary.  As a matter of fact, although somewhat OT, about 12 years ago I was scanning a number of Kodachromes of the slot canyons in the south west US.  As you might imagine, the suns rays were almost blown out at the top of the canyon.  I "underexposed" those chromes buy approximately 1.5 EV along with the proper exposure and stacked them.  I was then able to brush in the otherwise blown out areas very successfully.  Long before HDR, but it worked very well. 

Gary       

[bjanes]

I'd say this way: with the equal exposures S/N in deep shadows can be increased (when/if possible - for example w/o clipping important highlights) by selecting a certain higher nominal ISO values (before shooting !) _BUT_ more exposure when / if possible is always better than using higher gains (higher nominal ISOs) ...

+1. These special techniques are often used when more exposure is not feasible because of depth of field considerations or freezing of action requires decreasing the aperture (larger f/stop) and/or using faster shutter speeds. In this situation with an ISO-invariant camera such as the Nikon D800 with ISOs above 400, it us often advantageous to expose as much as possible using the aperture and shutter speed that conditions require and then use a lower than indicated ISO in order to protect the highlights. One then brightens the image in post. Jim Kasson discusses this strategy here.

This confirms Andrew's assertion that ISO has nothing to do with exposure, although a higher ISO will result in less exposure if one uses the metered value. Of course, one does not have to use the metered exposure. Andrew does not like the term ETTR and instead prefers "proper exposure", but the latter term does not indicate how that exposure is obtained. Jim restricts use of the term ETTR to base ISO. Above base ISO when one uses the smaller aperture and the faster shutter speed that conditions require as discussed above, moving the histogram to the right by increasing the ISO does not affect exposure, but it does reduce highlight headroom.

Regards,

Bill

[DP]

ISO has nothing to do with exposure

probably people shall start saying that "ISO" (quote marks intended) with the current state of technology in digital cameras is a part of a decision process about the exposure (while not part of that exposure process) which is done before exposure itself starts by a human operator and/or firmware based on metering and some other observations/experience/etc ... and we still can't say that never ever a technology will come that really change the "sensitivity" of the sensor or the amount of light reaching it (during the same exposure time & aperture setting) with a simple dial of ISO control... may be one day it happens... and then what ?

[digitaldog]

Setting ISO speed does not change the sensitivity of the sensor to light (Photons striking it), just like volume control does not change the sensitivity of a radio receiving it's signal. In each cases the setting (ISO or volume) controls only the signal processing, while the input stage (sensor, antenna) provides the same input signal.
That's why when ISO setting is cranked up, automatic exposure can result in more noise - automatic exposure in this case decreases the exposure (that is, the combination of aperture and shutter speed is set to allow less light captured by the sensor). Less exposure, less light, more noise. Again, avoid automatic exposure; it's easily fooled. No Histogram necessary. For Franz who may have left the building (or Tim), in the case of my ISO 800vs.100 example, I didn't follow the automatic exposure for the ISO 800 capture and due to amplification in this type of camera, less noise.

[DP]

Setting ISO speed does not change the sensitivity of the sensor to light

with the fine print - with the current technology ... you can't be sure what comes next or may be already lurking in the labs ...

PS: a funny example when somebody can claim that ISO is actually a part of exposure... shooting with speedlites (or strobes) with TTL... well there ISO you dial actually does control (TTL !) the amount of light emitted by speedlite/strobe and thus hitting the sensor with fixed exposure time and fixed aperture 

[digitaldog]

with the fine print - with the current technology ... you can't be sure what comes next or may be already lurking in the labs ...

Yeah, like anti-gravity machines and photon torpedoes; you're right. Let's stick to what's available to the people posting here today if we can. 

[Garnick]

Setting ISO speed does not change the sensitivity of the sensor to light (Photons striking it), just like volume control does not change the sensitivity of a radio receiving it's signal. In each cases the setting (ISO or volume) controls only the signal processing, while the input stage (sensor, antenna) provides the same input signal.
That's why when ISO setting is cranked up, automatic exposure can result in more noise - automatic exposure in this case decreases the exposure (that is, the combination of aperture and shutter speed is set to allow less light captured by the sensor). Less exposure, less light, more noise. Again, avoid automatic exposure; it's easily fooled. No Histogram necessary. For Franz who may have left the building (or Tim), in the case of my ISO 800vs.100 example, I didn't follow the automatic exposure for the ISO 800 capture and due to amplification in this type of camera, less noise.

Hmmm...I was expecting that Andrew would correct me on part of the two replies I offered, but I wasn't expecting that it would necessarily be this part.  The ISO part of my replies was from an article I read probably 5 years ago or more.  I have always remembered that and its association with why and how Film ISO actually works.  Therefore, since I am not a tech, nor have I ever pretended to be, it did seem plausible to me at the time.  Since I read this latest offering from Andrew I have been trying to locate that article, but to no avail unfortunately.  However, that is not important now, since it obviously contained some rather  egregious errors.  This particular topic had never been a focal point of my experience while using a digital DSLR until I read Bob DiNitale's article here on LuLa a couple of years ago, and then following his routine to prove it to myself.  Indeed, "the proof was in the pudding", or in Bob's case, "the milk".  Of course only those who have actually read the article will understand my reference to "milk". 

Thanks for correcting my error Andrew, and your explanation is certainly very understandable.  The only downside is that from this point on I'll have one more setting to deal with and hope I haven't overcooked that file.  But I have also "overexposed" from the meter reading by two stops and been able to bring it back with no apparent issues, and definitely NO noise that I can determine.  In ACR, if one is not pulling the "exposure" slider to a negative value, the file has not yet reached Optimum Exposure.  That's my personal standard for evaluating the exposure of an image.

Gary         

[DP]

Yeah, like anti-gravity machines and photon torpedoes; you're right. Let's stick to what's available to the people posting here today if we can. 

still with all those organic sensors in the making (like Panasonic/Fuji) - there is no assurance that sensitiviy of the sensor can't be changed... or Eric Fossum's new toy... where you can in fact control what an individual sensel (they call it - jot) will do (count or not) with an individual incoming photon so to say... so you can basically from the practical standpoint "control sensitivity" of the sensor based on those jots differently across the frame ___DURING THE EXPOSURE TIME__ ... and they have working prototypes - it is not photon torpedo

[digitaldog]

still with all those organic sensors in the making (like Panasonic/Fuji) - there is no assurance that sensitiviy of the sensor can't be changed... or Eric Fossum's new toy... where you can in fact control what an individual sensel (they call it - jot) will do (count or not) with an individual incoming photon so to say... so you can basically from the practical standpoint "control sensitivity" of the sensor based on those jots differently across the frame ___DURING THE EXPOSURE TIME__ ... and they have working prototypes - it is not photon torpedo

https://futurism.com/we-might-actually-be-able-to-make-star-treks-photon-torpedoes