Setting for JPEG Camera display-5DIII, to get best histogram and RAW capture

[ThomasR99]

Hi, all, thanks for stopping by the beginners' area to help with this.

I believe I read somewhere in the past that to get a histogram display that best matches the RAW image capture, saturation, sharpening and/or contrast should be set to levels below 'default' or the middle of the available range on the menu.  I can no longer recall when/where I read that, and if others think it's still true or not.  For some time I'd set these all 2-3 clicks/marks/notches (however you define it) below middle on the menu bar.  Naturally, this made my JPEG's somewhat 'bland' looking, and I needed to do almost as much work to make them presentable for others to see on my Flickr site as I would have if I'd just used the RAW image.  I do use an ETTR method so they tend to be a bit washed-out as well.

So, the questions are, how do you set your camera for JPEG preview (and the displayed histogram) in order to assist you in optimizing the RAW capture?

I mostly shoot in 'neutral' and 'monochrome' picture styles.  And as long as I"m able to, I will continue to post-process in Aperture.

Thanks for your time.  Links to articles discussing this for me (and others) to review are welcome as well.

[Tony Jay]

The main issue is tone and tonal levels.
Choose a Picture Style that does not alter contrast - like 'Faithful'.
This gives the best approximation on what the raw histogram will look like.

Tony Jay

[BobShaw]

The best advise for a beginner is to tape over the back screen and ever worry about. Just kidding, but there is far to much looking at the back screen and overthinking it. Everything you need is in the viewfinder. As long as the raw file is in range then you are good. Faithful is what I use also. Never use monochrome if you ever shoot video, as the video will be monochrome permanently.

[digitaldog]

Each camera is different but by and large, just ignore the histogram on the camera when shooting raw. It's mostly a lie.

Links to articles discussing this for me (and others) to review are welcome as well.

Everything you thought you wanted to know about Histograms

Another exhaustive 40 minute video examining:

What are histograms. In Photoshop, ACR, Lightroom.
Histograms: clipping color and tones, color spaces and color gamut.
Histogram and Photoshop’s Level’s command.
Histograms don’t tell us our images are good (examples).
Misconceptions about histograms. How they lie.
Histograms and Expose To The Right (ETTR).
Are histograms useful and if so, how?

Low rez (YouTube): http://www.youtube.com/watch?v=EjPsP4HhHhE
High rez: http://digitaldog.net/files/Histogram_Video.mov

[SZRitter]

Each camera is different but by and large, just ignore the histogram on the camera when shooting raw. It's mostly a lie. Everything you thought you wanted to know about Histograms

Another exhaustive 40 minute video examining:

What are histograms. In Photoshop, ACR, Lightroom.
Histograms: clipping color and tones, color spaces and color gamut.
Histogram and Photoshop’s Level’s command.
Histograms don’t tell us our images are good (examples).
Misconceptions about histograms. How they lie.
Histograms and Expose To The Right (ETTR).
Are histograms useful and if so, how?

Low rez (YouTube): http://www.youtube.com/watch?v=EjPsP4HhHhE
High rez: http://digitaldog.net/files/Histogram_Video.mov

Lie or not, it can give you valuable information for troubleshooting exposure on the fly. You can see if you are pushing towards under/over exposure, or if your scene may push the boundaries of your sensor's dynamic range. But, learning if you can blow the highlights, and by how much, on the histogram is a camera dependent thing, so experiment. Just remember, they are more of guidelines, not rules.

[AlterEgo]

Lie or not, it can give you valuable information for troubleshooting exposure on the fly. You can see if you are pushing towards under/over exposure, or if your scene may push the boundaries of your sensor's dynamic range. But, learning if you can blow the highlights, and by how much, on the histogram is a camera dependent thing, so experiment. Just remember, they are more of guidelines, not rules.

cameras with blinkies/zebra in EVF/LV can typically be tuned (if you sacrifices good colors in EVF/LV by using UniWB, flat tone curve, etc) to show clipping in raw channels with not worse than 1/3 EV precision... even w/o true raw histogram

[jferrari]

You can see if you are pushing towards under/over exposure,

Relatively moot if you're shooting RAW. Ever hear of ETTR.     - Jim

[Hans Kruse]

Hi, all, thanks for stopping by the beginners' area to help with this.

I believe I read somewhere in the past that to get a histogram display that best matches the RAW image capture, saturation, sharpening and/or contrast should be set to levels below 'default' or the middle of the available range on the menu.  I can no longer recall when/where I read that, and if others think it's still true or not.  For some time I'd set these all 2-3 clicks/marks/notches (however you define it) below middle on the menu bar.  Naturally, this made my JPEG's somewhat 'bland' looking, and I needed to do almost as much work to make them presentable for others to see on my Flickr site as I would have if I'd just used the RAW image.  I do use an ETTR method so they tend to be a bit washed-out as well.

So, the questions are, how do you set your camera for JPEG preview (and the displayed histogram) in order to assist you in optimizing the RAW capture?

I mostly shoot in 'neutral' and 'monochrome' picture styles.  And as long as I"m able to, I will continue to post-process in Aperture.

Thanks for your time.  Links to articles discussing this for me (and others) to review are welcome as well.

If you want to optimize your exposure on the 5D mkIII to avoid noise and banding and you shoot at low ISO, then do the following:

Bracket your shots with 1 stop between them and as many as needed so have at least one exposure with blinkies (which means it is potentially overexposed) and at least one thay is not blinking. Shoot RAW, of course. Then in Lightroom or ACR choose the most exposed that does not have overexposure warnings. You check that in Lightroom in the develop module by turning on the highlight clipping indicators. You can use the shortcut J.

Bracketing can be done in continous shooting when there is light enough to avoid blur from mirror slap or shake if you handhold. On the tripod shoot in live view and continous.

Don't waste your time on histograms in the camera as they are not optimal. You need to do the work on selecting the best exposure in pp with this approach but it is worth it, if you want optimal IQ from the Canon. You can also blend exposures afterwards if the dynamic range is too large to avoid noise and banding in then shadows.

[digitaldog]

Don't waste your time on histograms in the camera as they are not optimal.

Exactly! Something many of us did with film, long before we could imagine and worse, depend on a camera Histogram. A Histogram isn't necessary to produce optimal exposure and it's a heck of a lot more forgiving then trying to nail transparency film exposure.

It would be nice IF the camera manufactures would give us a Histogram that wasn't a fat lie when we expect to shoot raw, but that's apparently not on their radar.

[SZRitter]

cameras with blinkies/zebra in EVF/LV can typically be tuned (if you sacrifices good colors in EVF/LV by using UniWB, flat tone curve, etc) to show clipping in raw channels with not worse than 1/3 EV precision... even w/o true raw histogram

Basically, achieving the same thing. I prefer a histogram over blinkies, but to each their own.

Relatively moot if you're shooting RAW. Ever hear of ETTR>     - Jim

RAW still has a point it clips. You need to be even more conscious of this fact when doing ETTR as you are purposefully pushing towards that upper clip point.

[digitaldog]

You can see if you are pushing towards under/over exposure, or if your scene may push the boundaries of your sensor's dynamic range.

When shooting JPEG, yes, when shooting raw, no. It's just a big lie.

[digitaldog]

RAW still has a point it clips. You need to be even more conscious of this fact when doing ETTR as you are purposefully pushing towards that upper clip point.

Yes it does have a point that clips, just like transparency film, ever shoot it? This is photography 101 (optimal exposure) and what makes this condition an issue is people depend on feedback (the camera histogram) which has little to do with what they are capturing (raw not JPEG).

It's like getting into a car who's speedometer is 10 MPH too low. Don't complain if you get pulled over for speeding by using the lie of this feedback as factual. The camera histogram based on JPEG isn't based on raw and vise versa. One's gamma correct, one isn't. One's processed (baked), one isn't (raw).

[SZRitter]

When shooting JPEG, yes, when shooting raw, no. It's just a big lie.

If you read the part where I said it was more a guideline, you would realize that I don't expect it to be 100% faithful to the actual data. On my E-m5, with the default jpg settings, I can get about one or two stops past where the upper bound of the histogram is before the white clips.

But, I can also get a generalized idea of where my midtones are falling, how wide the DR is on the scene, and where my exposure could be getting in trouble.

[SZRitter]

Yes it does have a point that clips, just like transparency film, ever shoot it? This is photography 101 (optimal exposure) and what makes this condition an issue is people depend on feedback (the camera histogram) which has little to do with what they are capturing (raw not JPEG).

It's like getting into a car who's speedometer is 10 MPH too low. Don't complain if you get pulled over for speeding by using the lie of this feedback as factual. The camera histogram based on JPEG isn't based on raw and vise versa. One's gamma correct, one isn't. One's processed (baked), one isn't (raw).

Trust me, I have shot plenty of film (and just started B&W again after about a 10 year break). I 100% understand the difference between a RAW file and a jpeg. You don't seem to understand that a person can look at a relative scale to get an impression of what is going on to make some quick tweaks. If you understand the disconnect, you can understand how to interpret the data.

[digitaldog]

If you read the part where I said it was more a guideline, you would realize that I don't expect it to be 100% faithful to the actual data.

It's not even CLOSE to half that value (100% faithful). Depending on the camera system, it's a mile off.
Look, on my 5DMII, I can expose 1.5 stops OVER what the camera histogram and 'blinkies' say is proper exposure and end up without a lick of clipping in the raw converter:
http://www.digitalphotopro.com/technique/camera-technique/exposing-for-raw.html
You're not even in the ball park with the JPEG histogram, just like that speedometer that's 10 MPH or actually more off.

Trust me, I have shot plenty of film

Transparency? Good. If you can nail an exposure on a routine basis to 1/4 of a stop without a camera histogram, you've proven you don't need to pay attention to the lie it tells you about exposure for raw. As others have correctly stated here, if you shoot raw, ignore it, it's not telling you the truth.

[SZRitter]

I can look at a histogram of a camera I know, and get a pretty good idea of what I will have to work with when I get to Lightroom or Photoshop. It's like working in a professional kitchen, yes, you can use tools to get somethings exact, but when you are in the 7pm rush, you don't always have time, and you just learn your tools and instincts. When shooting events, it is the exact same thing. I can glance at the histogram and in a short time know if my exposure is going to give me workable results. It isn't about being exact, it's about getting a point that is workable.

I'll break it down on how I use the histogram.

1. Are white and black all the way in. If they are, then great, if not, how badly are they out.
2. How do the different "zones" of the histogram look? If I have a lot of room to the top, I can shift exposure to over expose knowing this will reduce noise in the shadows.
3. I hate clipped highlights on digital (the roll off is too harsh, while film feels much better, at least to me), so if I see way too much data on the high end of the scale, and lots of extra room at the bottom (I don't mind clipped shadows, and almost always process towards an "inky" black), I'll shift exposure down.



In about a second, I can do those three steps in my head. If I want precision, I'll meter with an external meter.

*edited: removed a slightly inflammatory intro. I mean no ill will, and value DigitalDog's different perspective.*

[SZRitter]

It's not even CLOSE to half that value (100% faithful). Depending on the camera system, it's a mile off.
Look, on my 5DMII, I can expose 1.5 stops OVER what the camera histogram and 'blinkies' say is proper exposure and end up without a lick of clipping in the raw converter:
http://www.digitalphotopro.com/technique/camera-technique/exposing-for-raw.html
You're not even in the ball park with the JPEG histogram, just like that speedometer that's 10 MPH or actually more off. Transparency? Good. If you can nail an exposure on a routine basis to 1/4 of a stop without a camera histogram, you've proven you don't need to pay attention to the lie it tells you about exposure for raw. As others have correctly stated here, if you shoot raw, ignore it, it's not telling you the truth.

Just an FYI, your link is broken.

[digitaldog]

I can look at a histogram of a camera I know, and get a pretty good idea of what I will have to work with when I get to Lightroom or Photoshop.

If you prefer to use a kitchen knife as a screwdriver, by all means use that kludge. You say you can properly expose film and you must have done so without a Histogram (maybe a Polaroid which is just about equally a lie about the E6 film). Why then would you view incorrect data when you state you didn't have to in the past? This JPEG histogram is a lie. I can easily illustrate this and actually did in my article. The meter says I'm 1.5 stops over exposed. ITS WRONG! The JPEG would be, the raw isn't at all and in fact, produces superior data because I didn't treat the raw exposure as if it were a JPEG.

1. Are white and black all the way in. If they are, then great, if not, how badly are they out.
2. How do the different "zones" of the histogram look? If I have a lot of room to the top, I can shift exposure to over expose knowing this will reduce noise in the shadows.
3. I hate clipped highlights on digital (the roll off is too harsh, while film feels much better, at least to me), so if I see way too much data on the high end of the scale, and lots of extra room at the bottom (I don't mind clipped shadows, and almost always process towards an "inky" black), I'll shift exposure down.

What you report is only correct for the JPEG. That's the bottom line and other's here have attempted to point this out. Do you have a raw converter that will show you what a real raw histogram looks like? Because you should seriously look at one versus what you see on the camera. Why on earth would you waste your time looking at this lie when you've admitted it was unnecessary capturing a far more difficult media in terms of exposure (transparency)?
As far as I'm concerned go ahead and futz with a Histogram that has little resemblance to the data you are trying to ideally capture. I'm AOK with that. Suggesting it's a useful workflow for others? Too many holes in that to accept. Again, if you have a Kitchen Knife and a screwdriver and you actually understand how and why each tool works as it does, you're kind of foolish recommending someone carve a turkey using a screw driver. But it's your turkey.

If I want precision, I'll meter with an external meter.

Sorry, no that's not correct. In the article provided, all exposures were based on a Minolta III flash meter, Incident mode. That meter was wrong, it is based again on a rendered (non linear) capture like the JPEG. I used it's initial setting to provide the base exposure which was 1.5 stops TOO DARK. The meter can't be taken at face value any more than your JPEG histogram. It's like setting it to ISO 100 when the capture media is really ISO 25! You cannot separate exposure and development, raw, JPEG or film and each is different. You have to test the media, as some of us did in the film days. ONLY when you know the correct ISO or how to compensate for ideal exposure (which is all ETTR is), that meter is as dumb about this process as your JPEG histogram is for raw data.

[digitaldog]

Just an FYI, your link is broken.

Yup, seems DPP mag has deleted years of older articles.
Here is a PDF:http://digitaldog.net/files/ExposeForRaw.pdf

[AlterEgo]

Basically, achieving the same thing. I prefer a histogram over blinkies, but to each their own.

not exactly - blinkies or zebra will show you where the clipping is exactly... histogram not... you certainly can make an educated guess, but the thing is that you have blinkies/zebra together with histogram vs just a histogram... a no brainer IMHO