Whatever happened to Camera-craft?

Depends what you are using it for.

White balance.... I'd made that perfectly clear. It's academic anyhow, as the interior of a lowepro bag is neither 18% nor grey. In fact, in the above image, you can see the camera's auto white balance has probably used the bag interior as it's white point any way, and it's made the radiator, grey carpet and the real grey card appear with a magenta cast.


Just buy a proper grey card.
 
Last edited:
Why is grey used for white balance anyway? White would be more logical. Although it's just as difficult to get a pure white as it is to get a pure grey.


Steve.
 
Why is grey used for white balance anyway? White would be more logical. Although it's just as difficult to get a pure white as it is to get a pure grey.


Steve.

As a means of white balance: Why would white better better? :) White pushes it to the extremes of exposure, and in many cases would mean it's just rendered white... literally... 255, 255, 255... which essentially means it contains NO information. Being grey makes it far more useful.
 
As a means of white balance: Why would white better better? :) White pushes it to the extremes of exposure, and in many cases would mean it's just rendered white... literally... 255, 255, 255... which essentially means it contains NO information. Being grey makes it far more useful.

Not really. For white balance, you are looking at relative levels of red, green and blue. You can push grey out to 255, 255, 255 just as easily as white. It just needs more light or a lower EV.

Equally, you can record either as 127, 127, 127 or any other value from 0 to 255.


Steve.
 
Not really. For white balance, you are looking at relative levels of red, green and blue. You can push grey out to 255, 255, 255 just as easily as white. It just needs more light or a lower EV.

Equally, you can record either as 127, 127, 127 or any other value from 0 to 255.


Steve.

It is possible that the actual exposure you have chosen for your shot will blow one of the rgb channels on a white card more easily than a grey card. You can underexpose to make your white card grey, but that is not an efficient workflow or better in any way that I can see.
 
As far as exposure goes, it makes no difference. As long as you get all three channels into the high area of their range, a colour temperature can be worked out.

Using white instead of grey has an advantage for low level light metering as there are an extra two and a bit stops of light (theoretically). If a meter is struggling with a grey card as it's near the low limit of its ability, that extra bit can help.

But it would require people to work out things in their heads. Now they're spoiled with automation, many can't do simple things like this!


Steve.
 
But it would require people to work out things in their heads. Now they're spoiled with automation, many can't do simple things like this!

Steady on! You're talking Camera Craft here :naughty:
 
As far as exposure goes, it makes no difference. As long as you get all three channels into the high area of their range, a colour temperature can be worked out.

That assumes that the sensor responds in the same way throughout its recording range to the red green and blue channels. There is (somewhere; I can't find it having just spent some time looking) a graph in The Manual of Photography 10th edition showing that this is not the case. In the absence of the graph, I fell back on looking at my own measurements where I certainly found this. The RGB curves were "in step" but separated throughout the mid tones, but came together at the high exposure end. This would seem to imply that if you're concerned at white balance in the mid tones, you should use a mid tone as the area to balance - hence a grey card.

I find myself wondering if basic sensitometry is camera craft, since it hasn't really anything to do with a camera. And if we allow this sort of thing to be covered as camera craft, then (if you're digitally based) jpg compression, tone mapping and a load of Photoshop stuff is going to be included as well. (Note that we film photographers have similar concerns with film.)
 
OK I surrender
maybe i should have worded my post a little better, after reading it again i do offer my appologies to anyone who was or has been offended by it, Hands up i give in
Please take the blindfold off me and get me away from the firing squad LOL :runaway:
 
As far as exposure goes, it makes no difference. As long as you get all three channels into the high area of their range, a colour temperature can be worked out.

Using white instead of grey has an advantage for low level light metering as there are an extra two and a bit stops of light (theoretically). If a meter is struggling with a grey card as it's near the low limit of its ability, that extra bit can help.

But it would require people to work out things in their heads. Now they're spoiled with automation, many can't do simple things like this!


Steve.

I've used a white handkerchief inside Holy Trinity Church in York to get a reading on my Sekonic spot meter which wouldn't give a reading from the dark pew. All I needed to do was adjust the reading to place the pew where I wanted on the tonal range.

Steady on! You're talking Camera Craft here :naughty:

If this sort of adjustment is "camera craft" then isn't it a fair extension to suggest that being able to assess the light without using a meter is camera craft par excellence? Using a meter in any shape or form then becomes a degeneration of craft skills. :D
 
This would seem to imply that if you're concerned at white balance in the mid tones, you should use a mid tone as the area to balance - hence a grey card.

Automatic exposure of a white card will give you exactly that. An 18% grey. As will automatic exposure of an 18% grey card!

Steady on! You're talking Camera Craft here

Sorry. I didn't mean to veer your thread back on topic. I will try not to do it again!


Steve.
 
Last edited:
Automatic exposure of a white card will give you exactly that. An 18% grey. As will automatic exposure of an 18% grey card!

Steve.

Without worrying about the exact percentage, I'll agree. I was assuming that because you used the phrase get all three channels into the high area of their range you were implying that the white card would be exposed to give white.
 
Automatic exposure of a white card will give you exactly that. An 18% grey. As will automatic exposure of an 18% grey card!

I was driving back from Leeds the other day and Listening to Radio 4 (I know - getting old). There was a whole debate about Black - from Physics to Psycho!

It would appear, technically, black is just a very low level of white! Get away - it was this type of thinking that made Newtown want to break light down into its constituent parts!
 
Yes. Black is just very dark white.

It could equally be a very dark shade of any other colour though!


Steve.
 
Automatic exposure of a white card will give you exactly that. An 18% grey. As will automatic exposure of an 18% grey card!

Steve.

......as would an automatic exposure of a pitch black night sky (if you had a shutter speed long enough) or, more usefully, a green field or a lawn or a tarmac road, or dozens of other things that have helpfully been left lying around the landscape for us to use.;)
 
Or, as I posted earlier. the palm of your hand - even more conveniently placed! Actually reflects a stop more than 18% so you have to open up a stop to compensate. Weston meters had a setting mark for this purpose.


Steve.
 
The idea of the "unprocessed, in-camera shot" is at best a myth and at worst a meaningless absurdity. With both digital and film.

Let's talk about digital images. So you think an 'unprocessed' in-camera shot is at best a myth and at worst a meaningless absurdity ?

Let me dispel that myth right now for you. If you use the cameras meter and expose the subject correctly, and mange to get it in focus, save it as a JPEG and then print the image or have it printed with out any post processing it is evidently possible you can achieve excellent results. If you've never experienced this enigma then you must be doing something wrong.

Plenty of examples out there, notably the Fuji and Sony systems which tender excellent image quality and dynamic range. Whilst I fully accept that many images do benefit from PP and notably when shooting RAW, it is perfectly possible to achieve very good to to excellent results using JPEGS straight out of the camera, and very quickly without having to resort to any post-processing.
 
Let's talk about digital images. So you think an 'unprocessed' in-camera shot is at best a myth and at worst a meaningless absurdity ?.

If you read the paragraph before the one you quoted, you'd see that jpgs are processed images.
 
Duplicate post. Browser paused for a while with some display in top right, then came back with post not made. Hitting the post button again gave two posts.
 
Last edited:
It could equally be a very dark shade of any other colour though

Apparently not Steve, at least according to the programme I was listening to. True black needs the total spectrum. I found it fascinating but it was a bit high-brow for driving so I may have missed some salient aspect :(

Might be available on i-Player?
 
Apparently not Steve, at least according to the programme I was listening to. True black needs the total spectrum.

That's just academic nonsense though. True black needs equal parts of all colours but at a zero level. That can be achieved by multiplying any combination by zero.


Steve.
 
If you read the paragraph before the one you quoted, you'd see that jpgs are processed images.

They are processed somehow whatever the output format. Unless you have a Fovian sensor in a Sigma camera, each pixel can only reproduce a single colour (red, green or blue) so some in camera trickery is required even if there is no other processing done to colour balance, contrast and sharpness.


Steve.
 
Let's talk about digital images. So you think an 'unprocessed' in-camera shot is at best a myth and at worst a meaningless absurdity ?

Let me dispel that myth right now for you. If you use the cameras meter and expose the subject correctly, and mange to get it in focus, save it as a JPEG and then print the image or have it printed with out any post processing it is evidently possible you can achieve excellent results. If you've never experienced this enigma then you must be doing something wrong.

Plenty of examples out there, notably the Fuji and Sony systems which tender excellent image quality and dynamic range. Whilst I fully accept that many images do benefit from PP and notably when shooting RAW, it is perfectly possible to achieve very good to to excellent results using JPEGS straight out of the camera, and very quickly without having to resort to any post-processing.
As StephenM has already pointed out, you've missed out a very important part of my point. JPG's are post-processed. If you shoot JPG your camera processes the RAW data for you, using the one-size-fits-all protocol that the software engineers have put into your camera. Some cameras give you a little bit of control over that protocol, allowing you to set a certain "look" for your "straight out of camera" JPGs. This is still post-processing. You have the most creative control shooting RAW and doing the post-processing yourself in Lightroom or whatever, rather than just delegating that part of the creative process to whoever wrote your camera's software. Nothing wrong with shooting JPG but you're kidding yourself if you think they're "unprocessed".

It's as much of a meaningless absurdity as showing a photograph you shot on film and claiming it's "unprocessed" because you took it to Boots instead of developing it yourself in a darkroom.
 
Or, as I posted earlier. the palm of your hand - even more conveniently placed! Actually reflects a stop more than 18% so you have to open up a stop to compensate. Weston meters had a setting mark for this purpose.


Steve.

It might be conveniently placed but not that easy to actually use..... oooops , I dropped the camera.
 
Let's talk about digital images. So you think an 'unprocessed' in-camera shot is at best a myth and at worst a meaningless absurdity ?

Let me dispel that myth right now for you. If you use the cameras meter and expose the subject correctly, and mange to get it in focus, save it as a JPEG and then print the image or have it printed with out any post processing it is evidently possible you can achieve excellent results. If you've never experienced this enigma then you must be doing something wrong.

Plenty of examples out there, notably the Fuji and Sony systems which tender excellent image quality and dynamic range. Whilst I fully accept that many images do benefit from PP and notably when shooting RAW, it is perfectly possible to achieve very good to to excellent results using JPEGS straight out of the camera, and very quickly without having to resort to any post-processing.

When my camera is left on full auto that includes auto DRO (Dynamic Range Optimisation), which lifts the shadows a bit in the jpeg if the dynamic range in the image is too big for printing (it does more than that, but the details don't matter in this argument). I can also switch to in-camera manual control of DRO, or in-camera HDR processing from a few different exposures which are combined in camera. For static nightime scenes when a single handheld exposure at the low ISO I want would be too long for handheld I can switch to multiple exposure noise reduction, where it shoots and combines several short exposure images to produce a sharper cleaner image than would otherwise be possible handheld. It can also do handheld panorama pans and stitch the images together in the camera. That's all in-camera jpeg processing. Of course I could also not use these in-camera jpeg features and instead do the same thing in post-processing.

So if there's something special about ex-camera jpegs, what's the important difference between letting my camera produce these jpegs, or doing exactly the same thing from an ex-camera RAW file or files in post processing to get exactly the same end image?
 
When my camera is left on full auto that includes auto DRO (Dynamic Range Optimisation), which lifts the shadows a bit in the jpeg if the dynamic range in the image is too big for printing (it does more than that, but the details don't matter in this argument). I can also switch to in-camera manual control of DRO, or in-camera HDR processing from a few different exposures which are combined in camera. For static nightime scenes when a single handheld exposure at the low ISO I want would be too long for handheld I can switch to multiple exposure noise reduction, where it shoots and combines several short exposure images to produce a sharper cleaner image than would otherwise be possible handheld. It can also do handheld panorama pans and stitch the images together in the camera. That's all in-camera jpeg processing. Of course I could also not use these in-camera jpeg features and instead do the same thing in post-processing.

So if there's something special about ex-camera jpegs, what's the important difference between letting my camera produce these jpegs, or doing exactly the same thing from an ex-camera RAW file or files in post processing to get exactly the same end image?

It's all about choices Chris ! Personally I would never shoot a nighttime scene without a tripod. Low ISO, small aperture, self timer or cable release. Job done. Never handled.
 
To go back to the film analogy, letting your camera process your images for you is the equivalent of using one particular type of film. You are restricted to the characteristics of that film. Taking RAW's and processing them yourself is like having a full range of film choices to hand after the fact; and thousands / millions more which would never have been available as film.
 
When my camera is left on full auto that includes auto DRO (Dynamic Range Optimisation), which lifts the shadows a bit in the jpeg if the dynamic range in the image is too big for printing (it does more than that, but the details don't matter in this argument). I can also switch to in-camera manual control of DRO, or in-camera HDR processing from a few different exposures which are combined in camera. For static nightime scenes when a single handheld exposure at the low ISO I want would be too long for handheld I can switch to multiple exposure noise reduction, where it shoots and combines several short exposure images to produce a sharper cleaner image than would otherwise be possible handheld. It can also do handheld panorama pans and stitch the images together in the camera. That's all in-camera jpeg processing. Of course I could also not use these in-camera jpeg features and instead do the same thing in post-processing.

So if there's something special about ex-camera jpegs, what's the important difference between letting my camera produce these jpegs, or doing exactly the same thing from an ex-camera RAW file or files in post processing to get exactly the same end image?
Even if you don't use any "features", your camera still processes the image. The JPG from an individual digital camera (or model of digital camera) has a "look" which is the result of decisions made by the software team about what happens to the raw data between capture and output. You can take those decisions yourself by shooting RAW and using something like Lightroom. If you want. I'm not arguing that you have to. But if you don't, you still can't claim that your image is unprocessed.
 
Last edited:
Even if you don't use any "features", your camera still processes the image. The JPG from an individual digital camera (or model of digital camera) has a "look" which is the result of decisions made by the software team about what happens to the raw data between capture and output. You can take those decisions yourself by shooting RAW and using something like Lightroom. If you want. I'm not arguing that you have to. But if you don't, you still can't claim that your image is unprocessed.
This and more, an often overlooked feature of in camera processing are the different styles of image, that are not only selectable, but customisable. You can set your camera up to produce a 'look' with a preset as customisable as most people's home cooked PP presets.
 
This and more, an often overlooked feature of in camera processing are the different styles of image, that are not only selectable, but customisable. You can set your camera up to produce a 'look' with a preset as customisable as most people's home cooked PP presets.

So, Phil... a Camera-Craft? It seems to me you can completely ruin your photos if you get it in wrong 'in camera' and you only use JPEGs :)
 
So, Phil... a Camera-Craft? It seems to me you can completely ruin your photos if you get it in wrong 'in camera' and you only use JPEGs :)
You can completely ruin your photos if you get it wrong in camera and shoot RAW! ;)

I think many photographers magnify the amount of influence a bit of processing can have on an image.

Back to the beginning of this thread, a 'great picture' comes from the mind of an artist, not from the twiddling of a few knobs on a camera or sliders in Lightroom.

The knob twiddling is craft, and it has its place, but a perfectly executed boring photo is still a boring photo.

ETA. The point of processing isn't to 'rescue' an image that wasn't shot properly in the first place. Again a massively popular misconception.
 
Last edited:
So, Phil... a Camera-Craft? It seems to me you can completely ruin your photos if you get it in wrong 'in camera' and you only use JPEGs :)

And film too for that matter ! - even a perfectly exposed film and developed negative can produce appalling work under the enlarger with inappropriate over-burning and vignetting. Entirely subjective though, beauty is in the eye of the beholder !
 
Not really. For white balance, you are looking at relative levels of red, green and blue. You can push grey out to 255, 255, 255 just as easily as white.

With all due respect, no you can't, because white is so much closer to the limit of over-exposure.


It just needs more light or a lower EV.

Exactly... so it's not as easy to lose the data, as it requires so much more exposure.

Equally, you can record either as 127, 127, 127 or any other value from 0 to 255.


Steve.

Yes you can, but 127 is mid grey, so much harder to over expose. Everything I'm talking about here is regarding white balance. not exposure. The only reason I suggest using a grey card is because not only is the reflected light levels guaranteed (not important for white balance) but so is its neutrality, which IS important for white balance.

Using a white object for white balance makes zero sense.
 
Last edited:
double post due to browser not responding. Ignore.


While I'm here editing though:

There are many people in this thread who have strong technical knowledge, and hence camera "craft". Do they all take good photographs?

Technical craft is just one aspect of photography, and those who only value that almost always become very boring photographers who shoot the same old formulaic tropes over and over again in pursuit of only measurable, technical excellence - sharpness, "composition" according to prescribed rules, exposure etc. Not only does that not guarantee great imagery, it's actually a recipe for never actually attaining it because you're so obsessed with it. The idea of practising all this crap is so you can basically operate intuitively on auto pilot and not THINK about them. Do great drivers THINK about what they're hands and feet are doing? Once you know all this stuff... just get on with it, and start thinking about the imagery you're about to create instead of how ****ing sharp it is, or whether it has "leading lines" or s**t like that. If you need to THINK about getting a sharp image, then you're probably not as great as you think you are.
 
Last edited:
With all due respect, no you can't, because white is so much closer to the limit of over-exposure.

Surely that depends on your camera settings. Whatever you set to get the grey correctly exposed will expose white as grey if the exposure is reduced by 2.5 stops.

Using a white object for white balance makes zero sense.

That sounds like one of the most illogical statements ever!

While I'm here editing though:

There are many people in this thread who have strong technical knowledge, and hence camera "craft". Do they all take good photographs?

Some do and some don't. The most important thing in photography is pointing the camera at something interesting.


Steve.
 
Last edited:
Surely that depends on your camera settings. Whatever you set to get the grey correctly exposed will expose white as grey if the exposure is reduced by 2.5 stops.


Steve.


You are missing the point. When you use a grey card to set white balance, you take a reference shot with the grey card in the shot. You don't expose for the grey card, or meter off it (although with a proper grey card you certainly CAN if you want)... it's just "there"... you meter for whatever you need to meter for to get the shot, hold a grey card in the frame, take a reference, then get on with your job. WIth a "white" card you can't meter off it, as you'd have to RE-meter when you removed it from the frame.


If it was white, you would be running the risk of it getting close to pure white - it may end up being 240 RGB or something, and that would NOT be a great sample to white balance from, or depending on the rest of the image and how you've metered it, it may end up being 255RGB, which is useless. Using white would mean you need to meter twice.... once to correctly expose your "white" card, and then again to actually take the photo. Why would you do this?

White is just NOT a great reference point for setting white balance... it's just not :) Go try it if you don't believe me. Why you arguing? LOL.. just use a grey card like everyone else.
 
Last edited:
Back
Top