Pentax Camera Image Controls vs Computer Editing

I wanted to lay down some thoughts on the recurring theme of 'in-camera' capture and computer image editing. It would be good to set out some goalposts and try and establish some common understanding for these debates, as I feel they are often confused by misunderstanding, termonologies and people actually not talking about the same things.

So, this is the world as I see it .... tell me what you think!

Our Pentax cameras have two distinct types of settings and controls ... and I'll group them according to my newly established definition terms ..... 1) Input Controls ..... 2) Output Controls

1) Input Controls directly affect the 'physical' optical image capture ...how the camera optics perform and the way the sensor records that optical image data.

So these input controls are .... a) Aperture b) Shutter Time Value c) ISO d) The focussing controls, whether manual or using AF points e) Flash use and flash exposure settings

These 5 key input parameters will determine the 'physical' characteristics that are burned into the digital capture. They determine the depth of field, the depiction of movement, the noise characteristics and other ISO related quality issues, the in (and out of) focus zones, and the lighting balance and colour qualities recorded. (Remember that exposure compensation is only a secondary control, as it will cause one of the primary input controls to change in order to make its effect.)

There is clearly one other powerful input factor, and that is the photographer, who chooses where to point the camera, what lens, and what to include or exclude from the scene ....

2) Output Controls act to adjust and alter the results from the input control settings ... ie, they apply computer controlled software adjustments to the digital imput file, which is effectively a RAW file inside the camera.

So these output controls include ... a) White Balance b) All the Custom Image settings ie Preset, saturation, hue, high/low contrast, sharpness, fine sharpness etc, filters for Black & White or other effects etc c) the highlight / shadow corrections, lens distortion corrections d) the HDR controls

All of these controls tell the Prime Image processor to carry out image editing software functions directly on the recorded image file, and then produce a new output JPEG file according to the quality and size settings made by the user. If the camera was set for JPEG only, then after this in-camera image editing has occurred, the stored 'RAW' file is discarded ... if both were set then both are kept ... and if RAW only then an output JPEG is not created, although the editing settings are all applied, but only a very small copy of the JPEG is created and 'embedded' within the RAW file for use as a thumbnail in cataloguing programs if needed.


So, there's my definitions of the two key sets of camera controls .... what do you think? Input controls, and output controls.

Now .... the fun.... how do these controls relate to computer image editing options and choices? My take on it is this .....

1) Input controls burn characteristics into the image in a 'physical' way - With the exception of minor brightness adjustment, any image editing that alters the effects of these input controls would be considered 'Image Manipulation', involving significant changes to the original material recorded by the camera .. eg, focussing, motion and blur, lighting changes and effects applied. Now, none of these are likely to add any quality to an image, so I would concur generally with those who say this is to be avoided ... much better to get this stuff right in the camera.


2) Output controls are simply settings for the camera processing software, so they have their computer software equivalents in almost every image editor. In most cases a skilled and experienced user will be able to improve on the results of this in-camera image processing by editing their images with a quality RAW converter and image editing program. These adjustment processes affecting saturation, contrast, hue, sharpness, shadow recovery, white balance etc etc, are all normal and required extensions to the image capture process .... the output must follow the input or we don't have a photograph.

So because all of these things can be done in more detail, with more control, on a computer, then that act of computer editing should be regarded in exactly the same manner as someone who sets their 'output controls' on their camera, and the same respect accorded to them for that. It will depend on the individual, their experience with either the camera output controls or computer software, as to the final quality of output they can achieve. Some will achieve better by precisely setting the camera controls, others will do better with computer editing.

Here endith the lesson ! I hope this may set the tone for future debates, and I hope these definitions of the 'input' and 'output' controls can help us better understand what we are talking about, and find common terms to use when talking about image editing vs camera capture settings.

I'm glad to see you thinking about things Nigel. You might need different names for your terms, as we already have input and output used in a slightly different way. Input sharpening is the sharpening set in the RAW converter, creative sharpening is done whilst processing an image in Photoshop (et al) and output sharpening is that applied just before outputting the image to the printer.

All of this is at the computer stage. The camera is a capture device that produces a RAW file, but also has limited processing power to produce a JPEG file as well. I say limited, but it's all relative and it can be quite powerful in its own right.

Yes John, and we should probably all try and make use of the camera 'output' controls as well as use software. It can be interesting to see the effects of adjustments when using the custom image settings.

You will see that I remain committed to my camera control definitions though ! If we restrict the thinking just to an actual camera and its controls, then I think 'input' and 'output' has a logic to it ... input directly affect the optics and the image data captured in a RAW form .... 'output' then adjusts and adapts that 'input' data to produce a JPEG file, the thing we actually view from the camera. By the same logic, then a RAW file would be considered the result of the 'Input' controls, and then its subsequent transformation via a RAW converter into a viewable image format would be considered 'output' .

The actual controls on the camera fall logically (as I have detailed) into these two distict groups.

Just a small point, Nigel. After the JPEG has been processed and saved in a Pentax K-5, you can preview the image and then, if you wish, opt to save the Raw data file also, by pressing the AE-L button. But I don't know whether this is an option in other Pentax cameras.

Cheers.
Philip

It is a useful practical option Philip, I suppose if you suddenly realised for some reason you wanted a RAW .... It is the same on the K7 also.

Why not set the camera to save in jpeg + RAW - I know it takes up more space on the card but then you will always get both files? ( that's what I do).

I find it's simply not necessary most of the time. But to each his/her own.

Cheers.
Philip

You could if there's a likely use for both files. Personally I generally decide before starting a shooting session, thinking about the subject type and likely processing approach I'll take, and then stick with one format throughout.

I don't see RAW as a kind of 'get out clause' for making exposure errors in camera ... I believe in fact that the ' input' stage settings (as I defined them earlier) are equally important in determining ultimate quality no matter which format we save the file in. Any limitations become more significant at the 'output' stage, when making those settings on the camera, or with computer editing, if you only have a JPEG ... But this is dependent really on the intended use for the image and the final size required also.

If this is to genuinely cause confusion, then I'd suggest conceptualising this as pre shot and Post/after shot control options.

Pre being everything you are referring to as Input Nigel, and Post being everything you are considering as Output.

Pre is what you (the photographer) can influence before the shot is captured, and Post being what the camera handles from there on.

Much of this will then flow onto work flow choices. For me for instance I set my camera to jpeg + raw and use the jpegs for quick skimming and image selection. Anything important will be edited in a subsequent image manipulation software application. To what degree depends on what I feel is required. Anything not important is likely taken on an iPhone and is generally shared, uploaded to Facebook or sent via email or Whatsapp directly, with no subsequent image manipulation done.

I only edit the RAW file as I'm not interested in any image manipulation choice the camera made on the day, at the time of the shot. I use my DSLR as an image capture device only, and then develop the photo as I desire afterwards. In that respect my workflow matches the manual development of the film days very closely. I see in camera JPEG as being the equivalent to the 1 hour photo development of which I have no control over, so this is recreated on the iPhone now. I want my DSLR to capture the widest range of detail possible, which gives me the most image detail to work with in development. That's akin to shooting with a flat picture profile in video terms, so the final work can be colour graded to suit the overall tone requirements of the whole piece, without any camera decisions burnt in.

All good points Richard .... I'm going to stick with ' input' and' ' output', as I'm restricting this to camera controls specifically .... My camera output controls are not 'post shoot' at all, in fact they are set up prior to taking the shots, just like the input settings ... Both input and output are pre-set at time of capture ... That's why I'm not saying 'pre' and 'post' which would make more sense when talking about the shooting and post processing stages.

My definitions are ways to divide the camera controls into those that directly affect the RAW image data (input) and those that only affect the JPEG output. The output controls all have parallels in computer software, but the inputs don't ... Unless you begin to perform ' manipulations' . This is the point of this theory ... To give us a firmer ground on which to judge the type of image adjustment and how we view it as being an essential component of the process, or rather something that some worked call ' cheating '.

I don't agree with your characterisation of camera jpegs as 1 hour photos ... We do have quite respectable on camera controls for adjusting the shots ...I call these the camera 'output' controls.

+1

Also, 1 hour photos from a shop are not usually developed by the photographer, whereas JPEGs, particularly the large 4-star files from our DSLRs, can be extensively adjusted (although 'extensively' is rarely necessary) in photo software, including ACR.

Cheers.
Philip

Agreed MrB ... let me give two practical examples of how I see this theory proving useful ...

Say we had an image that was shot at f8, but the photographer realised later that it needed less DOF, and should really have been shot at f3.5. But he's got a RAW file of it, so he uses software to create a blur effect and works at applying it in a natural way with brushes and masking etc ... its OK, but not really totally 'natural' .. what we might have expected had it been shot at the required aperture in the first place.

Now this is an example of an 'input' control not being set correctly at capture. The editing work required could be described as 'manipulation' , and indeed some people might call it 'cheating'. But at least we all understand the definition and can apply our own standards to the matter ... we have a level playing field for discussion .. ie that an 'input' control was wrong, and it has been put right in software. Now most would agree that is not the ideal way to work, that it would be better to 'get it right in camera' ....

Next example, we have a camera JPEG file where the photographer feels the saturation is not enough ... when setting her 'output' control for saturation (in the custom image setting) she left the slider on plus 1 ... it should ideally have been on plus 3 for the particular image.

Now, in software, our photographer uses a saturation control to boost the colours, and the result looks much better. Now, we can also agree surely, that this is a positive and 'OK' example of image adjustment using the computer ... because it apples to one of my camera 'output' settings (see the definitions in the OP). These are things that generally are duplicated both on the camera and in computer software. Again, we an understand the acceptability and simple extension this step makes to the normal photo creating process... and hopefully more understanding and agreement!

So, this is the rule ..... 'input' control faults corrected in software ...equals not too good, should have been done on the camera ....

'Output' control faults corrected in software .... perfectly OK types of adjustment, normal to do this to every image ... Now surely we can all agree to adhere to this new rule

I agree this is not strictly correct and want to make sure this doesn't derail the discussion.

Clearly you have far more control over the settings you want to apply in camera than you ever did with a 1 hour photo processing shop. What I was referring to though was that you have to make those decisions in place in the field, some of which are baked onto the final output. You're making those decisions before you've seen what is captured, and they are often harder to correct, especially if you're applying any in camera filters etc.

Nigel has alluded to this though, in that physical settings are harder to correct to appear natural than image forming software settings. Like I say, I'm not looking to derail this conversation with the 1 hour lab comment, just to put forward my take that image manipulation should be done later when you 'develop' your image, rather than in camera. This of course is central to my own use of the image capture tool and subsequent development workflow. Your results may vary as they say.

Something I would like to know is which (if any) of the Output controls has a physical effect if the image is shot in RAW. I was always of the opinion that customisable output functions such as saturation, contrast, hue, sharpness, shadow recovery, white balance etc only take effect on a JPEG image. Am I correct here ?

Many thanks
Nigel

The digital filters and so on only apply to the JPEG file. In fact, we can't shoot RAW at all in some modes, such as HDR.

There is another argument of course that, far from being a weak processor, the JPEG output of a camera is carefully tweaked by the manufacturer to get the optimum from that particular model. This could be in ways that we couldn't hope to guess at or emulate as we don't have the information. So the quirks of a particular sensor design, the shortfalls of a lens design, can all be corrected using JPEG.

In one sense it would be interesting to actually pit the RAW route against the JPEG route, but in the end it's the image that counts and there's no real world evidence that anyone can spot the difference. I tried this with the *istDS, making large A3 prints from RAW and JPEG files of the same scene. Even looking closely, very few people could see any difference, and some of those guessed the wrong way round.

Now the game has changed, and in our last print competition the judge was complimenting the staggering detail (his words) in an image produced by the Q-S1, a JPEG file by the way. In the end, it's all about the final image.