In-camera RAW histograms

[Frans Waterlander]

As far as I know, there are no cameras available with in-camera RAW histograms. So why is that? How difficult could it be to write the code? After all, code has been created for demosaicing the RAW data and creating jpeg histograms. Why only jpeg? Isn't RAW way more important for the more serious photographer? How fast asleep are the camera manufacturers? I don't get it; do you?

[digitaldog]

As far as I know, there are no cameras available with in-camera RAW histograms.

Nope.

How difficult could it be to write the code?

How much coding do you understand? It wasn't difficult for these guys: https://magiclantern.fm and that goes back to the nope statement above, if you own the cameras supported.

Isn't RAW way more important for the more serious photographer?

How serious a photographer are you?    Own Rawdigger?

[digitaldog]

Some of us optimally exposed transparency film, without a Histogram of course, often as paid professionals (that's how serious our photography was), within a 1/4 of a stop or better. It's kind of interesting that the youngster's who've only known digital capture find the need to use Histograms to expose but yeah, since most digital cameras that capture raw provide a Histogram, it be nice if they didn't lie and actually showed us the raw Histogram. But is it necessary? No more than it was when some of us professionally, seriously photographed on transparency film, a media that's far, far less forgiving than a digital capture.
Food for thought.... 

[fdisilvestro]

In the film era it was common to use spot and/or incident meters & shoot Polaroid proofs. It was also common to do exposure bracketing especially for transparency film. If histograms were available then, I’m sure that many professional photographers would have used them.

Back to the OP, i don’t  think it is difficult to code a raw histogram, but it would probably confuse the majority of casual users. Also, as sensors get better and the industry shifts to EVFs, the need for a raw histogram in camera is diminished

[digitaldog]

In the film era it was common to use spot and/or incident meters & shoot Polaroid proofs. It was also common to do exposure bracketing especially for transparency film.

And in the digital era, we can and still do use spot and incident meters, and the LCD is even better than a Polaroid. As for bracketing, no, never for shooting people. Never for shooting sports or anything that wasn't say a studio setup (and why not?). Snip tests yes. A12 backs dedicated for one test shot per setup, yes. And again, nailing exposure without a Histogram.

[Frans Waterlander]

Again that condescending tone, Andrew. It gets really, really old.

At least you agree that there are no cameras that come from the supplier with RAW histograms.
The question remains why camera manufacturers don't supply that function. If Magic Lantern can do it, then Canon, Nikon, Sony, etc. certainly can do it, but they don't. Why? I think it would be a great tool and selling point for a lot of people. Of course, serious professionals like you, Andrew, don't need it.

[E. Dinur]

Maybe it is because the makers believe "serious professionals" and even "serious enthusiasts" constitute an insignificant market share. So until one sees a profit-producing reason to break the ice, what incentive do the others have? It would be interesting to see figures about the percentage of camera-users who shoot jpg only. Who regard the Raw option in the menu as incomprehensible and frightening or a useless waste of time and mental resources. Perhaps an indicator is the number of "experienced Raw shooters" and self-declared gurus one sees on various forums, who also admit to always shooting Raw + jpg, "just in case I screw the Raw up", when in fact the opposite mindset would be more appropriate. 

[Martin Kristiansen]

I never shoot Jpg and never have. Personally have zero interest in a raw histogram. Modern cameras have huge dynamic range and really good metering systems. It’s pretty hard to screw it up actually. If I have time to play with histograms I probably have time to bracket if I really need it. I can also set to picture effect on on my mirrorless camera.

 If I’m shooting in studio I’m tethered to a computer running C1 with a calibrated monitor. If I manage  to screw that up I don’t think the lack of an in camera raw histogram is my biggest problem.

[Rob C]

I never shoot Jpg and never have. Personally have zero interest in a raw histogram. Modern cameras have huge dynamic range and really good metering systems. It’s pretty hard to screw it up actually. If I have time to play with histograms I probably have time to bracket if I really need it. I can also set to picture effect on on my mirrorless camera.

 If I’m shooting in studio I’m tethered to a computer running C1 with a calibrated monitor. If I manage  to screw that up I don’t think the lack of an in camera raw histogram is my biggest problem.

Yeah, I never shoot anything but RAW either. What's the point? Coming from film, I believe in getting the best digital "negative or tranny" that I possibly can. As a Nikon user, I settled on Matrix metering right from the start, and found it so reliable that I only go off auto ISO when it's a case of serious backlighting, when the rear screen is very helpful in advising when I have it best, and the only times I chimp - or even think about doing it. I still have a Minolta Flash Meter 111, but I wonder if it still works after all this time doing nothing?

I had a Polaroid back, but almost never used it.

As a P.S.

On chimping for anything but exposure tests: for myself, I see no advantage at all, because the way the image looks as it comes out of the camera has very little to do with the way it's going to look when I have finished messing about with it and trying to get it to reveal something of what I think I see in the shot. To my surprise and glee. I sometimes discover something on the monitor that I had failed to see at the time of shooting.

On the very rare occasions, post-retirement, that I have shot with people, I hate it when they ask to have a look at the back of the camera. They might as well be asking to have a look at the negatives. They will be disappointed. I rather they be happy.

[KLaban]

I also have never shot a jpg in my life, use matrix metering and find it to be generally very good.

The advantage with mirrorless is that we see what the sensor sees and therefore see the exposure before capture.

[Bart_van_der_Wolf]

To those who say they never shoot JPEGs, your Raw file's thumbnail is a JPEG... 

But I agree with Rob, it won't look much like the finished image in most cases.

And to those who can remember shooting film, I would have loved a histogram based exposure when I shot 4x5inch (Agfachrome and E6) transparencies... Incident light meters additionally required to know the lighting contrast, and spot meters could not always meter how much of the specular highlights would shift in color, assuming one could have consistent filmprocessing (not all labs achieved good enough process control).

But I'm digressing. Yeah, it would be a useful feature to have raw-based histograms or better yet, exposure control. 

Raw histograms as such are not too helpful without a Gamma adjustment (otherwise most histogram data would usually be in the lower bins, and very little in the higher bins), but clipping indicators could be very useful in nailing the optimal exposure, e.g. in letting the specular highlights clip (or not, depending on whether their color needs to be retained).

Cheers,
Bart

[digitaldog]

Again that condescending tone, Andrew. It gets really, really old.

“If my answers frighten you then you should cease asking scary questions.”
― Quentin Tarantino, Pulp Fiction
Or questions when you refuse to accept the answers that don't fit your predetermined desires.

At least you agree that there are no cameras that come from the supplier with RAW histograms.

Was that the question, NO. There are cameras with available in-camera raw Histograms.

If Magic Lantern can do it, then Canon, Nikon, Sony, etc. certainly can do it, but they don't. Why?

Simple and correct answer you'll never accept: They choose not to. Now go ask each manufacturer yourself, then come back with their answers if you're really serious about this question and really serious about an answer from the source.

Of course, serious professionals like you, Andrew, don't need it.

That is correct and I explained why. That doesn't mean I'd not like to see one offered. I'd like to see a camera I can afford with a DR that equally matches a human; I'm not holding my breath.

Now that the question has been answered, here's a new one: where can we see examples of your 'serious photography'? 

[DP]

As far as I know, there are no cameras available with in-camera RAW histograms. So why is that? How difficult could it be to write the code? After all, code has been created for demosaicing the RAW data and creating jpeg histograms. Why only jpeg? Isn't RAW way more important for the more serious photographer? How fast asleep are the camera manufacturers? I don't get it; do you?

histograms (raw or not raw) are quite useless - if you want to see where the clipping (in raw channels) is in your frame is use properly tuned (UniWB + other settings) blinkies/zebra

[Frans Waterlander]

“If my answers frighten you then you should cease asking scary questions.”
― Quentin Tarantino, Pulp Fiction
Or questions when you refuse to accept the answers that don't fit your predetermined desires. Was that the question, NO. There are cameras with available in-camera raw Histograms. Simple and correct answer you'll never accept: They choose not to. Now go ask each manufacturer yourself, then come back with their answers if you're really serious about this question and really serious about an answer from the source. That is correct and I explained why. That doesn't mean I'd not like to see one offered. I'd like to see a camera I can afford with a DR that equally matches a human; I'm not holding my breath.

Now that the question has been answered, here's a new one: where can we see examples of your 'serious photography'? 

Snappy come-backs only make your condescending attitude that more obvious. People would respect you more and would value your advise more if you dropped the annoying attitude, but I'm not holding my breathe.

[digitaldog]

Snappy come-backs only make your condescending attitude that more obvious. People would respect you more and would value your advise more if you dropped the annoying attitude, but I'm not holding my breathe.

I'm not at all interested in respect; it's earned by knowledge, by actions, by doing. I don't care about your idea of my attitude or the value of my posts as we've been though too many such experiences where you ask a question and ignore the answers. Do you want me to paste that paper trail here again?
You came here asking a question of members who, as far as I know, not a single one either works directly for a major camera manufacture or those of us who are beta's for these companies can speak outside our NDA's so you are asking a question that you should be asking to the actual manufacturers. See how far that gets you....

So the only answers you'll get are either based solely on opinions lacking in direct facts from the manufacturers or answers you will dismiss. Why ask? Just tell us the answer you wish to hear and maybe someone will tell you what you want to hear. One answer you got and it appears can't accept is, they don't offer a raw histogram because they choose not to provide a raw histogram. I know that probably either doesn't make sense to you or it isn't the answer you wish to hear. Would you be happy hearing that they don't because they all conspire to make photographers think a JPEG histogram is better than a raw Histogram? Or that every time a bell rings, or a photographer views a JPEG Histogram, an angel gets it's wings? Just tell us what you wish to hear, then you'll hear an answer that makes you happy.

Meanwhile, other's have told you how to get by without a raw Histogram, some have told you (and I agree), it's not useful nor necessary. Are you having problems learning to properly expose your raw data? If so ask; some here can teach you about this fundamental part of the science of Photography.

Now that the question has been answered, here's a new one you apparently don't wish to answer: where can we see examples of your 'serious photography'? 

[luxborealis]

I would be curious about how much more accurate a raw histogram would be from the jpeg histogram we currently see. Although LCDs have increased in pixel resolution, given the typically small size of histograms, would the difference be significant enough to determine more accurate exposure? Would, for example, channel highlights be within, say, 1 stop? ½ stop? ⅓ stop? Remembering back to my Pentax Digital Spotmeter, wasn’t it in ⅓-stop intervals?

Also, would the additional processing power required to produce a RAW histogram as instantly as we have come to expect take away from the speed at which RAW files are written?

Maybe engineers have made a trade-off: jpeg histograms are within allowable tolerances for determining exposure without slowing write speeds? I’m not a techie, so I have no idea, but in the absence of any other theories . . .

[Frans Waterlander]

I'm not at all interested in respect; it's earned by knowledge, by actions, by doing.

Yes, some people definitely have a deficiency in the respect area, failing to understand it's an asset and self-fulfillment issue, and as a result fail to treat others with respect as well.

[digitaldog]

Yes, some people definitely have a deficiency in the respect area, failing to understand it's an asset and self-fulfillment issue, and as a result fail to treat others with respect as well.

Respect has to be earned; earn it.
Franz, I know a little of your background (and nothing of your abilities or lack thereof as a photographer) so to answer your question as you'll happily accept it; the reason there are no cameras that have a raw Histogram directly from manufacturers is because Obama placed a sanction on the generation of raw Histograms. Evil right?
Happy now? 

[digitaldog]

I would be curious about how much more accurate a raw histogram would be from the jpeg histogram we currently see.

Download a demo of RawDigger; you'll know based on your camera and based on how you expose for the raw data vs. the JPEG data.

[Chris Kern]

I would be curious about how much more accurate a raw histogram would be from the jpeg histogram we currently see.

Well, they're both accurate―in the sense that the JPEG histogram represents the distribution of the in-camera-processed JPEG and the raw histogram represents the distribution of light reaching the sensor.  But I presume your question is how different they are.  Typically quite a lot.  Attached: (1) a JPEG extracted from a raw file produced by a Fuji X-T3, (2) the histogram of the JPEG, (3) the histogram of the raw image data.